diff options
Diffstat (limited to 'fs')
228 files changed, 8967 insertions, 4538 deletions
diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c index 392c5dac1981..d934f04e7736 100644 --- a/fs/9p/v9fs.c +++ b/fs/9p/v9fs.c @@ -184,10 +184,20 @@ static int v9fs_parse_options(struct v9fs_session_info *v9ses, char *opts) v9ses->afid = option; break; case Opt_uname: - match_strlcpy(v9ses->uname, &args[0], PATH_MAX); + kfree(v9ses->uname); + v9ses->uname = match_strdup(&args[0]); + if (!v9ses->uname) { + ret = -ENOMEM; + goto free_and_return; + } break; case Opt_remotename: - match_strlcpy(v9ses->aname, &args[0], PATH_MAX); + kfree(v9ses->aname); + v9ses->aname = match_strdup(&args[0]); + if (!v9ses->aname) { + ret = -ENOMEM; + goto free_and_return; + } break; case Opt_nodevmap: v9ses->nodev = 1; @@ -287,21 +297,21 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, struct p9_fid *fid; int rc; - v9ses->uname = __getname(); + v9ses->uname = kstrdup(V9FS_DEFUSER, GFP_KERNEL); if (!v9ses->uname) return ERR_PTR(-ENOMEM); - v9ses->aname = __getname(); + v9ses->aname = kstrdup(V9FS_DEFANAME, GFP_KERNEL); if (!v9ses->aname) { - __putname(v9ses->uname); + kfree(v9ses->uname); return ERR_PTR(-ENOMEM); } init_rwsem(&v9ses->rename_sem); rc = bdi_setup_and_register(&v9ses->bdi, "9p", BDI_CAP_MAP_COPY); if (rc) { - __putname(v9ses->aname); - __putname(v9ses->uname); + kfree(v9ses->aname); + kfree(v9ses->uname); return ERR_PTR(rc); } @@ -309,8 +319,6 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, list_add(&v9ses->slist, &v9fs_sessionlist); spin_unlock(&v9fs_sessionlist_lock); - strcpy(v9ses->uname, V9FS_DEFUSER); - strcpy(v9ses->aname, V9FS_DEFANAME); v9ses->uid = ~0; v9ses->dfltuid = V9FS_DEFUID; v9ses->dfltgid = V9FS_DEFGID; @@ -412,8 +420,8 @@ void v9fs_session_close(struct v9fs_session_info *v9ses) kfree(v9ses->cachetag); } #endif - __putname(v9ses->uname); - __putname(v9ses->aname); + kfree(v9ses->uname); + kfree(v9ses->aname); bdi_destroy(&v9ses->bdi); diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index dd6f7ee1e312..c2483e97beee 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -738,6 +738,7 @@ v9fs_cached_file_write(struct file *filp, const char __user * data, static const struct vm_operations_struct v9fs_file_vm_ops = { .fault = filemap_fault, .page_mkwrite = v9fs_vm_page_mkwrite, + .remap_pages = generic_file_remap_pages, }; diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index cbf9dbb1b2a2..890bed538f9b 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -1276,12 +1276,12 @@ static int v9fs_readlink(struct dentry *dentry, char *buffer, int buflen) } /* copy extension buffer into buffer */ - strncpy(buffer, st->extension, buflen); + retval = min(strlen(st->extension)+1, (size_t)buflen); + memcpy(buffer, st->extension, retval); - p9_debug(P9_DEBUG_VFS, "%s -> %s (%s)\n", - dentry->d_name.name, st->extension, buffer); + p9_debug(P9_DEBUG_VFS, "%s -> %s (%.*s)\n", + dentry->d_name.name, st->extension, buflen, buffer); - retval = strnlen(buffer, buflen); done: p9stat_free(st); kfree(st); diff --git a/fs/Kconfig.binfmt b/fs/Kconfig.binfmt index 022574202749..0efd1524b977 100644 --- a/fs/Kconfig.binfmt +++ b/fs/Kconfig.binfmt @@ -164,3 +164,11 @@ config BINFMT_MISC You may say M here for module support and later load the module when you have use for it; the module is called binfmt_misc. If you don't know what to answer at this point, say Y. + +config COREDUMP + bool "Enable core dump support" if EXPERT + default y + help + This option enables support for performing core dumps. You almost + certainly want to say Y here. Not necessary on systems that never + need debugging or only ever run flawless code. diff --git a/fs/Makefile b/fs/Makefile index 8938f8250320..1d7af79288a0 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -11,7 +11,7 @@ obj-y := open.o read_write.o file_table.o super.o \ attr.o bad_inode.o file.o filesystems.o namespace.o \ seq_file.o xattr.o libfs.o fs-writeback.o \ pnode.o drop_caches.o splice.o sync.o utimes.o \ - stack.o fs_struct.o statfs.o coredump.o + stack.o fs_struct.o statfs.o ifeq ($(CONFIG_BLOCK),y) obj-y += buffer.o bio.o block_dev.o direct-io.o mpage.o ioprio.o @@ -48,6 +48,7 @@ obj-$(CONFIG_FS_MBCACHE) += mbcache.o obj-$(CONFIG_FS_POSIX_ACL) += posix_acl.o xattr_acl.o obj-$(CONFIG_NFS_COMMON) += nfs_common/ obj-$(CONFIG_GENERIC_ACL) += generic_acl.o +obj-$(CONFIG_COREDUMP) += coredump.o obj-$(CONFIG_FHANDLE) += fhandle.o diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index e7396cfdb109..91b11650722e 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -392,10 +392,12 @@ static struct vfsmount *autofs4_d_automount(struct path *path) ino->flags |= AUTOFS_INF_PENDING; spin_unlock(&sbi->fs_lock); status = autofs4_mount_wait(dentry); - if (status) - return ERR_PTR(status); spin_lock(&sbi->fs_lock); ino->flags &= ~AUTOFS_INF_PENDING; + if (status) { + spin_unlock(&sbi->fs_lock); + return ERR_PTR(status); + } } done: if (!(ino->flags & AUTOFS_INF_EXPIRING)) { diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index d146e181d10d..0e7a6f81ae36 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -32,31 +32,8 @@ static int load_aout_binary(struct linux_binprm *, struct pt_regs * regs); static int load_aout_library(struct file*); -static int aout_core_dump(struct coredump_params *cprm); - -static struct linux_binfmt aout_format = { - .module = THIS_MODULE, - .load_binary = load_aout_binary, - .load_shlib = load_aout_library, - .core_dump = aout_core_dump, - .min_coredump = PAGE_SIZE -}; - -#define BAD_ADDR(x) ((unsigned long)(x) >= TASK_SIZE) - -static int set_brk(unsigned long start, unsigned long end) -{ - start = PAGE_ALIGN(start); - end = PAGE_ALIGN(end); - if (end > start) { - unsigned long addr; - addr = vm_brk(start, end - start); - if (BAD_ADDR(addr)) - return addr; - } - return 0; -} +#ifdef CONFIG_COREDUMP /* * Routine writes a core dump image in the current directory. * Currently only a stub-function. @@ -66,7 +43,6 @@ static int set_brk(unsigned long start, unsigned long end) * field, which also makes sure the core-dumps won't be recursive if the * dumping of the process results in another error.. */ - static int aout_core_dump(struct coredump_params *cprm) { struct file *file = cprm->file; @@ -89,7 +65,7 @@ static int aout_core_dump(struct coredump_params *cprm) current->flags |= PF_DUMPCORE; strncpy(dump.u_comm, current->comm, sizeof(dump.u_comm)); dump.u_ar0 = offsetof(struct user, regs); - dump.signal = cprm->signr; + dump.signal = cprm->siginfo->si_signo; aout_dump_thread(cprm->regs, &dump); /* If the size of the dump file exceeds the rlimit, then see what would happen @@ -135,6 +111,32 @@ end_coredump: set_fs(fs); return has_dumped; } +#else +#define aout_core_dump NULL +#endif + +static struct linux_binfmt aout_format = { + .module = THIS_MODULE, + .load_binary = load_aout_binary, + .load_shlib = load_aout_library, + .core_dump = aout_core_dump, + .min_coredump = PAGE_SIZE +}; + +#define BAD_ADDR(x) ((unsigned long)(x) >= TASK_SIZE) + +static int set_brk(unsigned long start, unsigned long end) +{ + start = PAGE_ALIGN(start); + end = PAGE_ALIGN(end); + if (end > start) { + unsigned long addr; + addr = vm_brk(start, end - start); + if (BAD_ADDR(addr)) + return addr; + } + return 0; +} /* * create_aout_tables() parses the env- and arg-strings in new user diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 0225fddf49b7..fbd9f60bd763 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -27,6 +27,7 @@ #include <linux/compiler.h> #include <linux/highmem.h> #include <linux/pagemap.h> +#include <linux/vmalloc.h> #include <linux/security.h> #include <linux/random.h> #include <linux/elf.h> @@ -35,7 +36,13 @@ #include <asm/uaccess.h> #include <asm/param.h> #include <asm/page.h> -#include <asm/exec.h> + +#ifndef user_long_t +#define user_long_t long +#endif +#ifndef user_siginfo_t +#define user_siginfo_t siginfo_t +#endif static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs); static int load_elf_library(struct file *); @@ -881,7 +888,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) } if (elf_interpreter) { - unsigned long uninitialized_var(interp_map_addr); + unsigned long interp_map_addr = 0; elf_entry = load_elf_interp(&loc->interp_elf_ex, interpreter, @@ -1115,7 +1122,7 @@ static unsigned long vma_dump_size(struct vm_area_struct *vma, if (always_dump_vma(vma)) goto whole; - if (vma->vm_flags & VM_NODUMP) + if (vma->vm_flags & VM_DONTDUMP) return 0; /* Hugetlb memory check */ @@ -1127,7 +1134,7 @@ static unsigned long vma_dump_size(struct vm_area_struct *vma, } /* Do not dump I/O mapped devices or special mappings */ - if (vma->vm_flags & (VM_IO | VM_RESERVED)) + if (vma->vm_flags & VM_IO) return 0; /* By default, dump shared memory if mapped from an anonymous file. */ @@ -1372,6 +1379,103 @@ static void fill_auxv_note(struct memelfnote *note, struct mm_struct *mm) fill_note(note, "CORE", NT_AUXV, i * sizeof(elf_addr_t), auxv); } +static void fill_siginfo_note(struct memelfnote *note, user_siginfo_t *csigdata, + siginfo_t *siginfo) +{ + mm_segment_t old_fs = get_fs(); + set_fs(KERNEL_DS); + copy_siginfo_to_user((user_siginfo_t __user *) csigdata, siginfo); + set_fs(old_fs); + fill_note(note, "CORE", NT_SIGINFO, sizeof(*csigdata), csigdata); +} + +#define MAX_FILE_NOTE_SIZE (4*1024*1024) +/* + * Format of NT_FILE note: + * + * long count -- how many files are mapped + * long page_size -- units for file_ofs + * array of [COUNT] elements of + * long start + * long end + * long file_ofs + * followed by COUNT filenames in ASCII: "FILE1" NUL "FILE2" NUL... + */ +static void fill_files_note(struct memelfnote *note) +{ + struct vm_area_struct *vma; + unsigned count, size, names_ofs, remaining, n; + user_long_t *data; + user_long_t *start_end_ofs; + char *name_base, *name_curpos; + + /* *Estimated* file count and total data size needed */ + count = current->mm->map_count; + size = count * 64; + + names_ofs = (2 + 3 * count) * sizeof(data[0]); + alloc: + if (size >= MAX_FILE_NOTE_SIZE) /* paranoia check */ + goto err; + size = round_up(size, PAGE_SIZE); + data = vmalloc(size); + if (!data) + goto err; + + start_end_ofs = data + 2; + name_base = name_curpos = ((char *)data) + names_ofs; + remaining = size - names_ofs; + count = 0; + for (vma = current->mm->mmap; vma != NULL; vma = vma->vm_next) { + struct file *file; + const char *filename; + + file = vma->vm_file; + if (!file) + continue; + filename = d_path(&file->f_path, name_curpos, remaining); + if (IS_ERR(filename)) { + if (PTR_ERR(filename) == -ENAMETOOLONG) { + vfree(data); + size = size * 5 / 4; + goto alloc; + } + continue; + } + + /* d_path() fills at the end, move name down */ + /* n = strlen(filename) + 1: */ + n = (name_curpos + remaining) - filename; + remaining = filename - name_curpos; + memmove(name_curpos, filename, n); + name_curpos += n; + + *start_end_ofs++ = vma->vm_start; + *start_end_ofs++ = vma->vm_end; + *start_end_ofs++ = vma->vm_pgoff; + count++; + } + + /* Now we know exact count of files, can store it */ + data[0] = count; + data[1] = PAGE_SIZE; + /* + * Count usually is less than current->mm->map_count, + * we need to move filenames down. + */ + n = current->mm->map_count - count; + if (n != 0) { + unsigned shift_bytes = n * 3 * sizeof(data[0]); + memmove(name_base - shift_bytes, name_base, + name_curpos - name_base); + name_curpos -= shift_bytes; + } + + size = name_curpos - (char *)data; + fill_note(note, "CORE", NT_FILE, size, data); + err: ; +} + #ifdef CORE_DUMP_USE_REGSET #include <linux/regset.h> @@ -1385,7 +1489,10 @@ struct elf_thread_core_info { struct elf_note_info { struct elf_thread_core_info *thread; struct memelfnote psinfo; + struct memelfnote signote; struct memelfnote auxv; + struct memelfnote files; + user_siginfo_t csigdata; size_t size; int thread_notes; }; @@ -1480,7 +1587,7 @@ static int fill_thread_core_info(struct elf_thread_core_info *t, static int fill_note_info(struct elfhdr *elf, int phdrs, struct elf_note_info *info, - long signr, struct pt_regs *regs) + siginfo_t *siginfo, struct pt_regs *regs) { struct task_struct *dump_task = current; const struct user_regset_view *view = task_user_regset_view(dump_task); @@ -1550,7 +1657,7 @@ static int fill_note_info(struct elfhdr *elf, int phdrs, * Now fill in each thread's information. */ for (t = info->thread; t != NULL; t = t->next) - if (!fill_thread_core_info(t, view, signr, &info->size)) + if (!fill_thread_core_info(t, view, siginfo->si_signo, &info->size)) return 0; /* @@ -1559,9 +1666,15 @@ static int fill_note_info(struct elfhdr *elf, int phdrs, fill_psinfo(psinfo, dump_task->group_leader, dump_task->mm); info->size += notesize(&info->psinfo); + fill_siginfo_note(&info->signote, &info->csigdata, siginfo); + info->size += notesize(&info->signote); + fill_auxv_note(&info->auxv, current->mm); info->size += notesize(&info->auxv); + fill_files_note(&info->files); + info->size += notesize(&info->files); + return 1; } @@ -1588,8 +1701,12 @@ static int write_note_info(struct elf_note_info *info, if (first && !writenote(&info->psinfo, file, foffset)) return 0; + if (first && !writenote(&info->signote, file, foffset)) + return 0; if (first && !writenote(&info->auxv, file, foffset)) return 0; + if (first && !writenote(&info->files, file, foffset)) + return 0; for (i = 1; i < info->thread_notes; ++i) if (t->notes[i].data && @@ -1616,6 +1733,7 @@ static void free_note_info(struct elf_note_info *info) kfree(t); } kfree(info->psinfo.data); + vfree(info->files.data); } #else @@ -1681,6 +1799,7 @@ struct elf_note_info { #ifdef ELF_CORE_COPY_XFPREGS elf_fpxregset_t *xfpu; #endif + user_siginfo_t csigdata; int thread_status_size; int numnote; }; @@ -1690,8 +1809,8 @@ static int elf_note_info_init(struct elf_note_info *info) memset(info, 0, sizeof(*info)); INIT_LIST_HEAD(&info->thread_list); - /* Allocate space for six ELF notes */ - info->notes = kmalloc(6 * sizeof(struct memelfnote), GFP_KERNEL); + /* Allocate space for ELF notes */ + info->notes = kmalloc(8 * sizeof(struct memelfnote), GFP_KERNEL); if (!info->notes) return 0; info->psinfo = kmalloc(sizeof(*info->psinfo), GFP_KERNEL); @@ -1713,14 +1832,14 @@ static int elf_note_info_init(struct elf_note_info *info) static int fill_note_info(struct elfhdr *elf, int phdrs, struct elf_note_info *info, - long signr, struct pt_regs *regs) + siginfo_t *siginfo, struct pt_regs *regs) { struct list_head *t; if (!elf_note_info_init(info)) return 0; - if (signr) { + if (siginfo->si_signo) { struct core_thread *ct; struct elf_thread_status *ets; @@ -1738,13 +1857,13 @@ static int fill_note_info(struct elfhdr *elf, int phdrs, int sz; ets = list_entry(t, struct elf_thread_status, list); - sz = elf_dump_thread_status(signr, ets); + sz = elf_dump_thread_status(siginfo->si_signo, ets); info->thread_status_size += sz; } } /* now collect the dump for the current */ memset(info->prstatus, 0, sizeof(*info->prstatus)); - fill_prstatus(info->prstatus, current, signr); + fill_prstatus(info->prstatus, current, siginfo->si_signo); elf_core_copy_regs(&info->prstatus->pr_reg, regs); /* Set up header */ @@ -1761,9 +1880,11 @@ static int fill_note_info(struct elfhdr *elf, int phdrs, fill_note(info->notes + 1, "CORE", NT_PRPSINFO, sizeof(*info->psinfo), info->psinfo); - info->numnote = 2; + fill_siginfo_note(info->notes + 2, &info->csigdata, siginfo); + fill_auxv_note(info->notes + 3, current->mm); + fill_files_note(info->notes + 4); - fill_auxv_note(&info->notes[info->numnote++], current->mm); + info->numnote = 5; /* Try to dump the FPU. */ info->prstatus->pr_fpvalid = elf_core_copy_task_fpregs(current, regs, @@ -1825,6 +1946,9 @@ static void free_note_info(struct elf_note_info *info) kfree(list_entry(tmp, struct elf_thread_status, list)); } + /* Free data allocated by fill_files_note(): */ + vfree(info->notes[4].data); + kfree(info->prstatus); kfree(info->psinfo); kfree(info->notes); @@ -1951,7 +2075,7 @@ static int elf_core_dump(struct coredump_params *cprm) * Collect all the non-memory information about the process for the * notes. This also sets up the file header. */ - if (!fill_note_info(elf, e_phnum, &info, cprm->signr, cprm->regs)) + if (!fill_note_info(elf, e_phnum, &info, cprm->siginfo, cprm->regs)) goto cleanup; has_dumped = 1; diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 3d77cf81ba3c..a46049154107 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -39,7 +39,6 @@ #include <asm/uaccess.h> #include <asm/param.h> #include <asm/pgalloc.h> -#include <asm/exec.h> typedef char *elf_caddr_t; @@ -1205,7 +1204,7 @@ static int maydump(struct vm_area_struct *vma, unsigned long mm_flags) int dump_ok; /* Do not dump I/O mapped devices or special mappings */ - if (vma->vm_flags & (VM_IO | VM_RESERVED)) { + if (vma->vm_flags & VM_IO) { kdcore("%08lx: %08lx: no (IO)", vma->vm_start, vma->vm_flags); return 0; } @@ -1642,7 +1641,7 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm) goto cleanup; #endif - if (cprm->signr) { + if (cprm->siginfo->si_signo) { struct core_thread *ct; struct elf_thread_status *tmp; @@ -1661,13 +1660,13 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm) int sz; tmp = list_entry(t, struct elf_thread_status, list); - sz = elf_dump_thread_status(cprm->signr, tmp); + sz = elf_dump_thread_status(cprm->siginfo->si_signo, tmp); thread_status_size += sz; } } /* now collect the dump for the current */ - fill_prstatus(prstatus, current, cprm->signr); + fill_prstatus(prstatus, current, cprm->siginfo->si_signo); elf_core_copy_regs(&prstatus->pr_reg, cprm->regs); segs = current->mm->map_count; diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 178cb70acc26..e280352b28f9 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -107,7 +107,7 @@ static struct linux_binfmt flat_format = { static int flat_core_dump(struct coredump_params *cprm) { printk("Process %s:%d received signr %d and should have core dumped\n", - current->comm, current->pid, (int) cprm->signr); + current->comm, current->pid, (int) cprm->siginfo->si_signo); return(1); } diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c index e85c04b9f61c..a3f28f331b2b 100644 --- a/fs/bio-integrity.c +++ b/fs/bio-integrity.c @@ -70,23 +70,25 @@ static inline int use_bip_pool(unsigned int idx) } /** - * bio_integrity_alloc_bioset - Allocate integrity payload and attach it to bio + * bio_integrity_alloc - Allocate integrity payload and attach it to bio * @bio: bio to attach integrity metadata to * @gfp_mask: Memory allocation mask * @nr_vecs: Number of integrity metadata scatter-gather elements - * @bs: bio_set to allocate from * * Description: This function prepares a bio for attaching integrity * metadata. nr_vecs specifies the maximum number of pages containing * integrity metadata that can be attached. */ -struct bio_integrity_payload *bio_integrity_alloc_bioset(struct bio *bio, - gfp_t gfp_mask, - unsigned int nr_vecs, - struct bio_set *bs) +struct bio_integrity_payload *bio_integrity_alloc(struct bio *bio, + gfp_t gfp_mask, + unsigned int nr_vecs) { struct bio_integrity_payload *bip; unsigned int idx = vecs_to_idx(nr_vecs); + struct bio_set *bs = bio->bi_pool; + + if (!bs) + bs = fs_bio_set; BUG_ON(bio == NULL); bip = NULL; @@ -114,37 +116,22 @@ struct bio_integrity_payload *bio_integrity_alloc_bioset(struct bio *bio, return bip; } -EXPORT_SYMBOL(bio_integrity_alloc_bioset); - -/** - * bio_integrity_alloc - Allocate integrity payload and attach it to bio - * @bio: bio to attach integrity metadata to - * @gfp_mask: Memory allocation mask - * @nr_vecs: Number of integrity metadata scatter-gather elements - * - * Description: This function prepares a bio for attaching integrity - * metadata. nr_vecs specifies the maximum number of pages containing - * integrity metadata that can be attached. - */ -struct bio_integrity_payload *bio_integrity_alloc(struct bio *bio, - gfp_t gfp_mask, - unsigned int nr_vecs) -{ - return bio_integrity_alloc_bioset(bio, gfp_mask, nr_vecs, fs_bio_set); -} EXPORT_SYMBOL(bio_integrity_alloc); /** * bio_integrity_free - Free bio integrity payload * @bio: bio containing bip to be freed - * @bs: bio_set this bio was allocated from * * Description: Used to free the integrity portion of a bio. Usually * called from bio_free(). */ -void bio_integrity_free(struct bio *bio, struct bio_set *bs) +void bio_integrity_free(struct bio *bio) { struct bio_integrity_payload *bip = bio->bi_integrity; + struct bio_set *bs = bio->bi_pool; + + if (!bs) + bs = fs_bio_set; BUG_ON(bip == NULL); @@ -730,19 +717,18 @@ EXPORT_SYMBOL(bio_integrity_split); * @bio: New bio * @bio_src: Original bio * @gfp_mask: Memory allocation mask - * @bs: bio_set to allocate bip from * * Description: Called to allocate a bip when cloning a bio */ int bio_integrity_clone(struct bio *bio, struct bio *bio_src, - gfp_t gfp_mask, struct bio_set *bs) + gfp_t gfp_mask) { struct bio_integrity_payload *bip_src = bio_src->bi_integrity; struct bio_integrity_payload *bip; BUG_ON(bip_src == NULL); - bip = bio_integrity_alloc_bioset(bio, gfp_mask, bip_src->bip_vcnt, bs); + bip = bio_integrity_alloc(bio, gfp_mask, bip_src->bip_vcnt); if (bip == NULL) return -EIO; @@ -55,6 +55,7 @@ static struct biovec_slab bvec_slabs[BIOVEC_NR_POOLS] __read_mostly = { * IO code that does not need private memory pools. */ struct bio_set *fs_bio_set; +EXPORT_SYMBOL(fs_bio_set); /* * Our slab pool management @@ -233,26 +234,37 @@ fallback: return bvl; } -void bio_free(struct bio *bio, struct bio_set *bs) +static void __bio_free(struct bio *bio) { + bio_disassociate_task(bio); + + if (bio_integrity(bio)) + bio_integrity_free(bio); +} + +static void bio_free(struct bio *bio) +{ + struct bio_set *bs = bio->bi_pool; void *p; - if (bio_has_allocated_vec(bio)) - bvec_free_bs(bs, bio->bi_io_vec, BIO_POOL_IDX(bio)); + __bio_free(bio); - if (bio_integrity(bio)) - bio_integrity_free(bio, bs); + if (bs) { + if (bio_has_allocated_vec(bio)) + bvec_free_bs(bs, bio->bi_io_vec, BIO_POOL_IDX(bio)); - /* - * If we have front padding, adjust the bio pointer before freeing - */ - p = bio; - if (bs->front_pad) + /* + * If we have front padding, adjust the bio pointer before freeing + */ + p = bio; p -= bs->front_pad; - mempool_free(p, bs->bio_pool); + mempool_free(p, bs->bio_pool); + } else { + /* Bio was allocated by bio_kmalloc() */ + kfree(bio); + } } -EXPORT_SYMBOL(bio_free); void bio_init(struct bio *bio) { @@ -263,48 +275,85 @@ void bio_init(struct bio *bio) EXPORT_SYMBOL(bio_init); /** + * bio_reset - reinitialize a bio + * @bio: bio to reset + * + * Description: + * After calling bio_reset(), @bio will be in the same state as a freshly + * allocated bio returned bio bio_alloc_bioset() - the only fields that are + * preserved are the ones that are initialized by bio_alloc_bioset(). See + * comment in struct bio. + */ +void bio_reset(struct bio *bio) +{ + unsigned long flags = bio->bi_flags & (~0UL << BIO_RESET_BITS); + + __bio_free(bio); + + memset(bio, 0, BIO_RESET_BYTES); + bio->bi_flags = flags|(1 << BIO_UPTODATE); +} +EXPORT_SYMBOL(bio_reset); + +/** * bio_alloc_bioset - allocate a bio for I/O * @gfp_mask: the GFP_ mask given to the slab allocator * @nr_iovecs: number of iovecs to pre-allocate * @bs: the bio_set to allocate from. * * Description: - * bio_alloc_bioset will try its own mempool to satisfy the allocation. - * If %__GFP_WAIT is set then we will block on the internal pool waiting - * for a &struct bio to become free. + * If @bs is NULL, uses kmalloc() to allocate the bio; else the allocation is + * backed by the @bs's mempool. * - * Note that the caller must set ->bi_destructor on successful return - * of a bio, to do the appropriate freeing of the bio once the reference - * count drops to zero. - **/ + * When @bs is not NULL, if %__GFP_WAIT is set then bio_alloc will always be + * able to allocate a bio. This is due to the mempool guarantees. To make this + * work, callers must never allocate more than 1 bio at a time from this pool. + * Callers that need to allocate more than 1 bio must always submit the + * previously allocated bio for IO before attempting to allocate a new one. + * Failure to do so can cause deadlocks under memory pressure. + * + * RETURNS: + * Pointer to new bio on success, NULL on failure. + */ struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs) { + unsigned front_pad; + unsigned inline_vecs; unsigned long idx = BIO_POOL_NONE; struct bio_vec *bvl = NULL; struct bio *bio; void *p; - p = mempool_alloc(bs->bio_pool, gfp_mask); + if (!bs) { + if (nr_iovecs > UIO_MAXIOV) + return NULL; + + p = kmalloc(sizeof(struct bio) + + nr_iovecs * sizeof(struct bio_vec), + gfp_mask); + front_pad = 0; + inline_vecs = nr_iovecs; + } else { + p = mempool_alloc(bs->bio_pool, gfp_mask); + front_pad = bs->front_pad; + inline_vecs = BIO_INLINE_VECS; + } + if (unlikely(!p)) return NULL; - bio = p + bs->front_pad; + bio = p + front_pad; bio_init(bio); - if (unlikely(!nr_iovecs)) - goto out_set; - - if (nr_iovecs <= BIO_INLINE_VECS) { - bvl = bio->bi_inline_vecs; - nr_iovecs = BIO_INLINE_VECS; - } else { + if (nr_iovecs > inline_vecs) { bvl = bvec_alloc_bs(gfp_mask, nr_iovecs, &idx, bs); if (unlikely(!bvl)) goto err_free; - - nr_iovecs = bvec_nr_vecs(idx); + } else if (nr_iovecs) { + bvl = bio->bi_inline_vecs; } -out_set: + + bio->bi_pool = bs; bio->bi_flags |= idx << BIO_POOL_OFFSET; bio->bi_max_vecs = nr_iovecs; bio->bi_io_vec = bvl; @@ -316,80 +365,6 @@ err_free: } EXPORT_SYMBOL(bio_alloc_bioset); -static void bio_fs_destructor(struct bio *bio) -{ - bio_free(bio, fs_bio_set); -} - -/** - * bio_alloc - allocate a new bio, memory pool backed - * @gfp_mask: allocation mask to use - * @nr_iovecs: number of iovecs - * - * bio_alloc will allocate a bio and associated bio_vec array that can hold - * at least @nr_iovecs entries. Allocations will be done from the - * fs_bio_set. Also see @bio_alloc_bioset and @bio_kmalloc. - * - * If %__GFP_WAIT is set, then bio_alloc will always be able to allocate - * a bio. This is due to the mempool guarantees. To make this work, callers - * must never allocate more than 1 bio at a time from this pool. Callers - * that need to allocate more than 1 bio must always submit the previously - * allocated bio for IO before attempting to allocate a new one. Failure to - * do so can cause livelocks under memory pressure. - * - * RETURNS: - * Pointer to new bio on success, NULL on failure. - */ -struct bio *bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs) -{ - struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, fs_bio_set); - - if (bio) - bio->bi_destructor = bio_fs_destructor; - - return bio; -} -EXPORT_SYMBOL(bio_alloc); - -static void bio_kmalloc_destructor(struct bio *bio) -{ - if (bio_integrity(bio)) - bio_integrity_free(bio, fs_bio_set); - kfree(bio); -} - -/** - * bio_kmalloc - allocate a bio for I/O using kmalloc() - * @gfp_mask: the GFP_ mask given to the slab allocator - * @nr_iovecs: number of iovecs to pre-allocate - * - * Description: - * Allocate a new bio with @nr_iovecs bvecs. If @gfp_mask contains - * %__GFP_WAIT, the allocation is guaranteed to succeed. - * - **/ -struct bio *bio_kmalloc(gfp_t gfp_mask, unsigned int nr_iovecs) -{ - struct bio *bio; - - if (nr_iovecs > UIO_MAXIOV) - return NULL; - - bio = kmalloc(sizeof(struct bio) + nr_iovecs * sizeof(struct bio_vec), - gfp_mask); - if (unlikely(!bio)) - return NULL; - - bio_init(bio); - bio->bi_flags |= BIO_POOL_NONE << BIO_POOL_OFFSET; - bio->bi_max_vecs = nr_iovecs; - bio->bi_io_vec = bio->bi_inline_vecs; - bio->bi_destructor = bio_kmalloc_destructor; - - return bio; -} -EXPORT_SYMBOL(bio_kmalloc); - void zero_fill_bio(struct bio *bio) { unsigned long flags; @@ -420,11 +395,8 @@ void bio_put(struct bio *bio) /* * last put frees it */ - if (atomic_dec_and_test(&bio->bi_cnt)) { - bio_disassociate_task(bio); - bio->bi_next = NULL; - bio->bi_destructor(bio); - } + if (atomic_dec_and_test(&bio->bi_cnt)) + bio_free(bio); } EXPORT_SYMBOL(bio_put); @@ -466,26 +438,28 @@ void __bio_clone(struct bio *bio, struct bio *bio_src) EXPORT_SYMBOL(__bio_clone); /** - * bio_clone - clone a bio + * bio_clone_bioset - clone a bio * @bio: bio to clone * @gfp_mask: allocation priority + * @bs: bio_set to allocate from * * Like __bio_clone, only also allocates the returned bio */ -struct bio *bio_clone(struct bio *bio, gfp_t gfp_mask) +struct bio *bio_clone_bioset(struct bio *bio, gfp_t gfp_mask, + struct bio_set *bs) { - struct bio *b = bio_alloc_bioset(gfp_mask, bio->bi_max_vecs, fs_bio_set); + struct bio *b; + b = bio_alloc_bioset(gfp_mask, bio->bi_max_vecs, bs); if (!b) return NULL; - b->bi_destructor = bio_fs_destructor; __bio_clone(b, bio); if (bio_integrity(bio)) { int ret; - ret = bio_integrity_clone(b, bio, gfp_mask, fs_bio_set); + ret = bio_integrity_clone(b, bio, gfp_mask); if (ret < 0) { bio_put(b); @@ -495,7 +469,7 @@ struct bio *bio_clone(struct bio *bio, gfp_t gfp_mask) return b; } -EXPORT_SYMBOL(bio_clone); +EXPORT_SYMBOL(bio_clone_bioset); /** * bio_get_nr_vecs - return approx number of vecs @@ -1501,7 +1475,7 @@ struct bio_pair *bio_split(struct bio *bi, int first_sectors) trace_block_split(bdev_get_queue(bi->bi_bdev), bi, bi->bi_sector + first_sectors); - BUG_ON(bi->bi_vcnt != 1); + BUG_ON(bi->bi_vcnt != 1 && bi->bi_vcnt != 0); BUG_ON(bi->bi_idx != 0); atomic_set(&bp->cnt, 3); bp->error = 0; @@ -1511,17 +1485,22 @@ struct bio_pair *bio_split(struct bio *bi, int first_sectors) bp->bio2.bi_size -= first_sectors << 9; bp->bio1.bi_size = first_sectors << 9; - bp->bv1 = bi->bi_io_vec[0]; - bp->bv2 = bi->bi_io_vec[0]; - bp->bv2.bv_offset += first_sectors << 9; - bp->bv2.bv_len -= first_sectors << 9; - bp->bv1.bv_len = first_sectors << 9; + if (bi->bi_vcnt != 0) { + bp->bv1 = bi->bi_io_vec[0]; + bp->bv2 = bi->bi_io_vec[0]; + + if (bio_is_rw(bi)) { + bp->bv2.bv_offset += first_sectors << 9; + bp->bv2.bv_len -= first_sectors << 9; + bp->bv1.bv_len = first_sectors << 9; + } - bp->bio1.bi_io_vec = &bp->bv1; - bp->bio2.bi_io_vec = &bp->bv2; + bp->bio1.bi_io_vec = &bp->bv1; + bp->bio2.bi_io_vec = &bp->bv2; - bp->bio1.bi_max_vecs = 1; - bp->bio2.bi_max_vecs = 1; + bp->bio1.bi_max_vecs = 1; + bp->bio2.bi_max_vecs = 1; + } bp->bio1.bi_end_io = bio_pair_end_1; bp->bio2.bi_end_io = bio_pair_end_2; diff --git a/fs/block_dev.c b/fs/block_dev.c index 38e721b35d45..1a1e5e3b1eaf 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -116,6 +116,8 @@ EXPORT_SYMBOL(invalidate_bdev); int set_blocksize(struct block_device *bdev, int size) { + struct address_space *mapping; + /* Size must be a power of two, and between 512 and PAGE_SIZE */ if (size > PAGE_SIZE || size < 512 || !is_power_of_2(size)) return -EINVAL; @@ -124,6 +126,19 @@ int set_blocksize(struct block_device *bdev, int size) if (size < bdev_logical_block_size(bdev)) return -EINVAL; + /* Prevent starting I/O or mapping the device */ + percpu_down_write(&bdev->bd_block_size_semaphore); + + /* Check that the block device is not memory mapped */ + mapping = bdev->bd_inode->i_mapping; + mutex_lock(&mapping->i_mmap_mutex); + if (mapping_mapped(mapping)) { + mutex_unlock(&mapping->i_mmap_mutex); + percpu_up_write(&bdev->bd_block_size_semaphore); + return -EBUSY; + } + mutex_unlock(&mapping->i_mmap_mutex); + /* Don't change the size if it is same as current */ if (bdev->bd_block_size != size) { sync_blockdev(bdev); @@ -131,6 +146,9 @@ int set_blocksize(struct block_device *bdev, int size) bdev->bd_inode->i_blkbits = blksize_bits(size); kill_bdev(bdev); } + + percpu_up_write(&bdev->bd_block_size_semaphore); + return 0; } @@ -441,6 +459,12 @@ static struct inode *bdev_alloc_inode(struct super_block *sb) struct bdev_inode *ei = kmem_cache_alloc(bdev_cachep, GFP_KERNEL); if (!ei) return NULL; + + if (unlikely(percpu_init_rwsem(&ei->bdev.bd_block_size_semaphore))) { + kmem_cache_free(bdev_cachep, ei); + return NULL; + } + return &ei->vfs_inode; } @@ -449,6 +473,8 @@ static void bdev_i_callback(struct rcu_head *head) struct inode *inode = container_of(head, struct inode, i_rcu); struct bdev_inode *bdi = BDEV_I(inode); + percpu_free_rwsem(&bdi->bdev.bd_block_size_semaphore); + kmem_cache_free(bdev_cachep, bdi); } @@ -1567,6 +1593,22 @@ static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg) return blkdev_ioctl(bdev, mode, cmd, arg); } +ssize_t blkdev_aio_read(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos) +{ + ssize_t ret; + struct block_device *bdev = I_BDEV(iocb->ki_filp->f_mapping->host); + + percpu_down_read(&bdev->bd_block_size_semaphore); + + ret = generic_file_aio_read(iocb, iov, nr_segs, pos); + + percpu_up_read(&bdev->bd_block_size_semaphore); + + return ret; +} +EXPORT_SYMBOL_GPL(blkdev_aio_read); + /* * Write data to the block device. Only intended for the block device itself * and the raw driver which basically is a fake block device. @@ -1578,12 +1620,16 @@ ssize_t blkdev_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) { struct file *file = iocb->ki_filp; + struct block_device *bdev = I_BDEV(file->f_mapping->host); struct blk_plug plug; ssize_t ret; BUG_ON(iocb->ki_pos != pos); blk_start_plug(&plug); + + percpu_down_read(&bdev->bd_block_size_semaphore); + ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); if (ret > 0 || ret == -EIOCBQUEUED) { ssize_t err; @@ -1592,11 +1638,62 @@ ssize_t blkdev_aio_write(struct kiocb *iocb, const struct iovec *iov, if (err < 0 && ret > 0) ret = err; } + + percpu_up_read(&bdev->bd_block_size_semaphore); + blk_finish_plug(&plug); + return ret; } EXPORT_SYMBOL_GPL(blkdev_aio_write); +static int blkdev_mmap(struct file *file, struct vm_area_struct *vma) +{ + int ret; + struct block_device *bdev = I_BDEV(file->f_mapping->host); + + percpu_down_read(&bdev->bd_block_size_semaphore); + + ret = generic_file_mmap(file, vma); + + percpu_up_read(&bdev->bd_block_size_semaphore); + + return ret; +} + +static ssize_t blkdev_splice_read(struct file *file, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags) +{ + ssize_t ret; + struct block_device *bdev = I_BDEV(file->f_mapping->host); + + percpu_down_read(&bdev->bd_block_size_semaphore); + + ret = generic_file_splice_read(file, ppos, pipe, len, flags); + + percpu_up_read(&bdev->bd_block_size_semaphore); + + return ret; +} + +static ssize_t blkdev_splice_write(struct pipe_inode_info *pipe, + struct file *file, loff_t *ppos, size_t len, + unsigned int flags) +{ + ssize_t ret; + struct block_device *bdev = I_BDEV(file->f_mapping->host); + + percpu_down_read(&bdev->bd_block_size_semaphore); + + ret = generic_file_splice_write(pipe, file, ppos, len, flags); + + percpu_up_read(&bdev->bd_block_size_semaphore); + + return ret; +} + + /* * Try to release a page associated with block device when the system * is under memory pressure. @@ -1627,16 +1724,16 @@ const struct file_operations def_blk_fops = { .llseek = block_llseek, .read = do_sync_read, .write = do_sync_write, - .aio_read = generic_file_aio_read, + .aio_read = blkdev_aio_read, .aio_write = blkdev_aio_write, - .mmap = generic_file_mmap, + .mmap = blkdev_mmap, .fsync = blkdev_fsync, .unlocked_ioctl = block_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = compat_blkdev_ioctl, #endif - .splice_read = generic_file_splice_read, - .splice_write = generic_file_splice_write, + .splice_read = blkdev_splice_read, + .splice_write = blkdev_splice_write, }; int ioctl_by_bdev(struct block_device *bdev, unsigned cmd, unsigned long arg) diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index ff6475f409d6..208d8aa5b07e 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -16,6 +16,7 @@ * Boston, MA 021110-1307, USA. */ +#include <linux/vmalloc.h> #include "ctree.h" #include "disk-io.h" #include "backref.h" @@ -231,7 +232,7 @@ static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path, } if (!ret) { ret = ulist_add(parents, eb->start, - (unsigned long)eie, GFP_NOFS); + (uintptr_t)eie, GFP_NOFS); if (ret < 0) break; if (!extent_item_pos) { @@ -282,9 +283,7 @@ static int __resolve_indirect_ref(struct btrfs_fs_info *fs_info, goto out; } - rcu_read_lock(); - root_level = btrfs_header_level(root->node); - rcu_read_unlock(); + root_level = btrfs_old_root_level(root, time_seq); if (root_level + 1 == level) goto out; @@ -363,8 +362,8 @@ static int __resolve_indirect_refs(struct btrfs_fs_info *fs_info, ULIST_ITER_INIT(&uiter); node = ulist_next(parents, &uiter); ref->parent = node ? node->val : 0; - ref->inode_list = - node ? (struct extent_inode_elem *)node->aux : 0; + ref->inode_list = node ? + (struct extent_inode_elem *)(uintptr_t)node->aux : 0; /* additional parents require new refs being added here */ while ((node = ulist_next(parents, &uiter))) { @@ -375,8 +374,8 @@ static int __resolve_indirect_refs(struct btrfs_fs_info *fs_info, } memcpy(new_ref, ref, sizeof(*ref)); new_ref->parent = node->val; - new_ref->inode_list = - (struct extent_inode_elem *)node->aux; + new_ref->inode_list = (struct extent_inode_elem *) + (uintptr_t)node->aux; list_add(&new_ref->list, &ref->list); } ulist_reinit(parents); @@ -914,8 +913,8 @@ again: free_extent_buffer(eb); } ret = ulist_add_merge(refs, ref->parent, - (unsigned long)ref->inode_list, - (unsigned long *)&eie, GFP_NOFS); + (uintptr_t)ref->inode_list, + (u64 *)&eie, GFP_NOFS); if (!ret && extent_item_pos) { /* * we've recorded that parent, so we must extend @@ -959,7 +958,7 @@ static void free_leaf_list(struct ulist *blocks) while ((node = ulist_next(blocks, &uiter))) { if (!node->aux) continue; - eie = (struct extent_inode_elem *)node->aux; + eie = (struct extent_inode_elem *)(uintptr_t)node->aux; for (; eie; eie = eie_next) { eie_next = eie->next; kfree(eie); @@ -1108,44 +1107,97 @@ static int inode_ref_info(u64 inum, u64 ioff, struct btrfs_root *fs_root, found_key); } -/* - * this iterates to turn a btrfs_inode_ref into a full filesystem path. elements - * of the path are separated by '/' and the path is guaranteed to be - * 0-terminated. the path is only given within the current file system. - * Therefore, it never starts with a '/'. the caller is responsible to provide - * "size" bytes in "dest". the dest buffer will be filled backwards. finally, - * the start point of the resulting string is returned. this pointer is within - * dest, normally. - * in case the path buffer would overflow, the pointer is decremented further - * as if output was written to the buffer, though no more output is actually - * generated. that way, the caller can determine how much space would be - * required for the path to fit into the buffer. in that case, the returned - * value will be smaller than dest. callers must check this! - */ -char *btrfs_iref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path, - struct btrfs_inode_ref *iref, - struct extent_buffer *eb_in, u64 parent, - char *dest, u32 size) +int btrfs_find_one_extref(struct btrfs_root *root, u64 inode_objectid, + u64 start_off, struct btrfs_path *path, + struct btrfs_inode_extref **ret_extref, + u64 *found_off) +{ + int ret, slot; + struct btrfs_key key; + struct btrfs_key found_key; + struct btrfs_inode_extref *extref; + struct extent_buffer *leaf; + unsigned long ptr; + + key.objectid = inode_objectid; + btrfs_set_key_type(&key, BTRFS_INODE_EXTREF_KEY); + key.offset = start_off; + + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) + return ret; + + while (1) { + leaf = path->nodes[0]; + slot = path->slots[0]; + if (slot >= btrfs_header_nritems(leaf)) { + /* + * If the item at offset is not found, + * btrfs_search_slot will point us to the slot + * where it should be inserted. In our case + * that will be the slot directly before the + * next INODE_REF_KEY_V2 item. In the case + * that we're pointing to the last slot in a + * leaf, we must move one leaf over. + */ + ret = btrfs_next_leaf(root, path); + if (ret) { + if (ret >= 1) + ret = -ENOENT; + break; + } + continue; + } + + btrfs_item_key_to_cpu(leaf, &found_key, slot); + + /* + * Check that we're still looking at an extended ref key for + * this particular objectid. If we have different + * objectid or type then there are no more to be found + * in the tree and we can exit. + */ + ret = -ENOENT; + if (found_key.objectid != inode_objectid) + break; + if (btrfs_key_type(&found_key) != BTRFS_INODE_EXTREF_KEY) + break; + + ret = 0; + ptr = btrfs_item_ptr_offset(leaf, path->slots[0]); + extref = (struct btrfs_inode_extref *)ptr; + *ret_extref = extref; + if (found_off) + *found_off = found_key.offset; + break; + } + + return ret; +} + +char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path, + u32 name_len, unsigned long name_off, + struct extent_buffer *eb_in, u64 parent, + char *dest, u32 size) { - u32 len; int slot; u64 next_inum; int ret; - s64 bytes_left = size - 1; + s64 bytes_left = ((s64)size) - 1; struct extent_buffer *eb = eb_in; struct btrfs_key found_key; int leave_spinning = path->leave_spinning; + struct btrfs_inode_ref *iref; if (bytes_left >= 0) dest[bytes_left] = '\0'; path->leave_spinning = 1; while (1) { - len = btrfs_inode_ref_name_len(eb, iref); - bytes_left -= len; + bytes_left -= name_len; if (bytes_left >= 0) read_extent_buffer(eb, dest + bytes_left, - (unsigned long)(iref + 1), len); + name_off, name_len); if (eb != eb_in) { btrfs_tree_read_unlock_blocking(eb); free_extent_buffer(eb); @@ -1155,6 +1207,7 @@ char *btrfs_iref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path, ret = -ENOENT; if (ret) break; + next_inum = found_key.offset; /* regular exit ahead */ @@ -1170,8 +1223,11 @@ char *btrfs_iref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path, btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK); } btrfs_release_path(path); - iref = btrfs_item_ptr(eb, slot, struct btrfs_inode_ref); + + name_len = btrfs_inode_ref_name_len(eb, iref); + name_off = (unsigned long)(iref + 1); + parent = next_inum; --bytes_left; if (bytes_left >= 0) @@ -1188,12 +1244,39 @@ char *btrfs_iref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path, } /* + * this iterates to turn a btrfs_inode_ref into a full filesystem path. elements + * of the path are separated by '/' and the path is guaranteed to be + * 0-terminated. the path is only given within the current file system. + * Therefore, it never starts with a '/'. the caller is responsible to provide + * "size" bytes in "dest". the dest buffer will be filled backwards. finally, + * the start point of the resulting string is returned. this pointer is within + * dest, normally. + * in case the path buffer would overflow, the pointer is decremented further + * as if output was written to the buffer, though no more output is actually + * generated. that way, the caller can determine how much space would be + * required for the path to fit into the buffer. in that case, the returned + * value will be smaller than dest. callers must check this! + */ +char *btrfs_iref_to_path(struct btrfs_root *fs_root, + struct btrfs_path *path, + struct btrfs_inode_ref *iref, + struct extent_buffer *eb_in, u64 parent, + char *dest, u32 size) +{ + return btrfs_ref_to_path(fs_root, path, + btrfs_inode_ref_name_len(eb_in, iref), + (unsigned long)(iref + 1), + eb_in, parent, dest, size); +} + +/* * this makes the path point to (logical EXTENT_ITEM *) * returns BTRFS_EXTENT_FLAG_DATA for data, BTRFS_EXTENT_FLAG_TREE_BLOCK for * tree blocks and <0 on error. */ int extent_from_logical(struct btrfs_fs_info *fs_info, u64 logical, - struct btrfs_path *path, struct btrfs_key *found_key) + struct btrfs_path *path, struct btrfs_key *found_key, + u64 *flags_ret) { int ret; u64 flags; @@ -1237,10 +1320,17 @@ int extent_from_logical(struct btrfs_fs_info *fs_info, u64 logical, (unsigned long long)found_key->objectid, (unsigned long long)found_key->offset, (unsigned long long)flags, item_size); - if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) - return BTRFS_EXTENT_FLAG_TREE_BLOCK; - if (flags & BTRFS_EXTENT_FLAG_DATA) - return BTRFS_EXTENT_FLAG_DATA; + + WARN_ON(!flags_ret); + if (flags_ret) { + if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) + *flags_ret = BTRFS_EXTENT_FLAG_TREE_BLOCK; + else if (flags & BTRFS_EXTENT_FLAG_DATA) + *flags_ret = BTRFS_EXTENT_FLAG_DATA; + else + BUG_ON(1); + return 0; + } return -EIO; } @@ -1404,12 +1494,13 @@ int iterate_extent_inodes(struct btrfs_fs_info *fs_info, ULIST_ITER_INIT(&root_uiter); while (!ret && (root_node = ulist_next(roots, &root_uiter))) { pr_debug("root %llu references leaf %llu, data list " - "%#lx\n", root_node->val, ref_node->val, - ref_node->aux); - ret = iterate_leaf_refs( - (struct extent_inode_elem *)ref_node->aux, - root_node->val, extent_item_objectid, - iterate, ctx); + "%#llx\n", root_node->val, ref_node->val, + (long long)ref_node->aux); + ret = iterate_leaf_refs((struct extent_inode_elem *) + (uintptr_t)ref_node->aux, + root_node->val, + extent_item_objectid, + iterate, ctx); } ulist_free(roots); roots = NULL; @@ -1432,15 +1523,15 @@ int iterate_inodes_from_logical(u64 logical, struct btrfs_fs_info *fs_info, { int ret; u64 extent_item_pos; + u64 flags = 0; struct btrfs_key found_key; int search_commit_root = path->search_commit_root; - ret = extent_from_logical(fs_info, logical, path, - &found_key); + ret = extent_from_logical(fs_info, logical, path, &found_key, &flags); btrfs_release_path(path); if (ret < 0) return ret; - if (ret & BTRFS_EXTENT_FLAG_TREE_BLOCK) + if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) return -EINVAL; extent_item_pos = logical - found_key.objectid; @@ -1451,9 +1542,12 @@ int iterate_inodes_from_logical(u64 logical, struct btrfs_fs_info *fs_info, return ret; } -static int iterate_irefs(u64 inum, struct btrfs_root *fs_root, - struct btrfs_path *path, - iterate_irefs_t *iterate, void *ctx) +typedef int (iterate_irefs_t)(u64 parent, u32 name_len, unsigned long name_off, + struct extent_buffer *eb, void *ctx); + +static int iterate_inode_refs(u64 inum, struct btrfs_root *fs_root, + struct btrfs_path *path, + iterate_irefs_t *iterate, void *ctx) { int ret = 0; int slot; @@ -1470,7 +1564,7 @@ static int iterate_irefs(u64 inum, struct btrfs_root *fs_root, while (!ret) { path->leave_spinning = 1; ret = inode_ref_info(inum, parent ? parent+1 : 0, fs_root, path, - &found_key); + &found_key); if (ret < 0) break; if (ret) { @@ -1498,7 +1592,8 @@ static int iterate_irefs(u64 inum, struct btrfs_root *fs_root, "tree %llu\n", cur, (unsigned long long)found_key.objectid, (unsigned long long)fs_root->objectid); - ret = iterate(parent, iref, eb, ctx); + ret = iterate(parent, name_len, + (unsigned long)(iref + 1), eb, ctx); if (ret) break; len = sizeof(*iref) + name_len; @@ -1513,12 +1608,98 @@ static int iterate_irefs(u64 inum, struct btrfs_root *fs_root, return ret; } +static int iterate_inode_extrefs(u64 inum, struct btrfs_root *fs_root, + struct btrfs_path *path, + iterate_irefs_t *iterate, void *ctx) +{ + int ret; + int slot; + u64 offset = 0; + u64 parent; + int found = 0; + struct extent_buffer *eb; + struct btrfs_inode_extref *extref; + struct extent_buffer *leaf; + u32 item_size; + u32 cur_offset; + unsigned long ptr; + + while (1) { + ret = btrfs_find_one_extref(fs_root, inum, offset, path, &extref, + &offset); + if (ret < 0) + break; + if (ret) { + ret = found ? 0 : -ENOENT; + break; + } + ++found; + + slot = path->slots[0]; + eb = path->nodes[0]; + /* make sure we can use eb after releasing the path */ + atomic_inc(&eb->refs); + + btrfs_tree_read_lock(eb); + btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK); + btrfs_release_path(path); + + leaf = path->nodes[0]; + item_size = btrfs_item_size_nr(leaf, path->slots[0]); + ptr = btrfs_item_ptr_offset(leaf, path->slots[0]); + cur_offset = 0; + + while (cur_offset < item_size) { + u32 name_len; + + extref = (struct btrfs_inode_extref *)(ptr + cur_offset); + parent = btrfs_inode_extref_parent(eb, extref); + name_len = btrfs_inode_extref_name_len(eb, extref); + ret = iterate(parent, name_len, + (unsigned long)&extref->name, eb, ctx); + if (ret) + break; + + cur_offset += btrfs_inode_extref_name_len(leaf, extref); + cur_offset += sizeof(*extref); + } + btrfs_tree_read_unlock_blocking(eb); + free_extent_buffer(eb); + + offset++; + } + + btrfs_release_path(path); + + return ret; +} + +static int iterate_irefs(u64 inum, struct btrfs_root *fs_root, + struct btrfs_path *path, iterate_irefs_t *iterate, + void *ctx) +{ + int ret; + int found_refs = 0; + + ret = iterate_inode_refs(inum, fs_root, path, iterate, ctx); + if (!ret) + ++found_refs; + else if (ret != -ENOENT) + return ret; + + ret = iterate_inode_extrefs(inum, fs_root, path, iterate, ctx); + if (ret == -ENOENT && found_refs) + return 0; + + return ret; +} + /* * returns 0 if the path could be dumped (probably truncated) * returns <0 in case of an error */ -static int inode_to_path(u64 inum, struct btrfs_inode_ref *iref, - struct extent_buffer *eb, void *ctx) +static int inode_to_path(u64 inum, u32 name_len, unsigned long name_off, + struct extent_buffer *eb, void *ctx) { struct inode_fs_paths *ipath = ctx; char *fspath; @@ -1531,20 +1712,16 @@ static int inode_to_path(u64 inum, struct btrfs_inode_ref *iref, ipath->fspath->bytes_left - s_ptr : 0; fspath_min = (char *)ipath->fspath->val + (i + 1) * s_ptr; - fspath = btrfs_iref_to_path(ipath->fs_root, ipath->btrfs_path, iref, eb, - inum, fspath_min, bytes_left); + fspath = btrfs_ref_to_path(ipath->fs_root, ipath->btrfs_path, name_len, + name_off, eb, inum, fspath_min, bytes_left); if (IS_ERR(fspath)) return PTR_ERR(fspath); if (fspath > fspath_min) { - pr_debug("path resolved: %s\n", fspath); ipath->fspath->val[i] = (u64)(unsigned long)fspath; ++ipath->fspath->elem_cnt; ipath->fspath->bytes_left = fspath - fspath_min; } else { - pr_debug("missed path, not enough space. missing bytes: %lu, " - "constructed so far: %s\n", - (unsigned long)(fspath_min - fspath), fspath_min); ++ipath->fspath->elem_missed; ipath->fspath->bytes_missing += fspath_min - fspath; ipath->fspath->bytes_left = 0; @@ -1566,7 +1743,7 @@ static int inode_to_path(u64 inum, struct btrfs_inode_ref *iref, int paths_from_inode(u64 inum, struct inode_fs_paths *ipath) { return iterate_irefs(inum, ipath->fs_root, ipath->btrfs_path, - inode_to_path, ipath); + inode_to_path, ipath); } struct btrfs_data_container *init_data_container(u32 total_bytes) @@ -1575,7 +1752,7 @@ struct btrfs_data_container *init_data_container(u32 total_bytes) size_t alloc_bytes; alloc_bytes = max_t(size_t, total_bytes, sizeof(*data)); - data = kmalloc(alloc_bytes, GFP_NOFS); + data = vmalloc(alloc_bytes); if (!data) return ERR_PTR(-ENOMEM); @@ -1626,6 +1803,6 @@ void free_ipath(struct inode_fs_paths *ipath) { if (!ipath) return; - kfree(ipath->fspath); + vfree(ipath->fspath); kfree(ipath); } diff --git a/fs/btrfs/backref.h b/fs/btrfs/backref.h index 032f4dc7eab8..d61feca79455 100644 --- a/fs/btrfs/backref.h +++ b/fs/btrfs/backref.h @@ -33,14 +33,13 @@ struct inode_fs_paths { typedef int (iterate_extent_inodes_t)(u64 inum, u64 offset, u64 root, void *ctx); -typedef int (iterate_irefs_t)(u64 parent, struct btrfs_inode_ref *iref, - struct extent_buffer *eb, void *ctx); int inode_item_info(u64 inum, u64 ioff, struct btrfs_root *fs_root, struct btrfs_path *path); int extent_from_logical(struct btrfs_fs_info *fs_info, u64 logical, - struct btrfs_path *path, struct btrfs_key *found_key); + struct btrfs_path *path, struct btrfs_key *found_key, + u64 *flags); int tree_backref_for_extent(unsigned long *ptr, struct extent_buffer *eb, struct btrfs_extent_item *ei, u32 item_size, @@ -63,10 +62,19 @@ int btrfs_find_all_roots(struct btrfs_trans_handle *trans, char *btrfs_iref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path, struct btrfs_inode_ref *iref, struct extent_buffer *eb, u64 parent, char *dest, u32 size); +char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path, + u32 name_len, unsigned long name_off, + struct extent_buffer *eb_in, u64 parent, + char *dest, u32 size); struct btrfs_data_container *init_data_container(u32 total_bytes); struct inode_fs_paths *init_ipath(s32 total_bytes, struct btrfs_root *fs_root, struct btrfs_path *path); void free_ipath(struct inode_fs_paths *ipath); +int btrfs_find_one_extref(struct btrfs_root *root, u64 inode_objectid, + u64 start_off, struct btrfs_path *path, + struct btrfs_inode_extref **ret_extref, + u64 *found_off); + #endif diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index 5b2ad6bc4fe7..ed8ca7ca5eff 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -38,6 +38,7 @@ #define BTRFS_INODE_DELALLOC_META_RESERVED 4 #define BTRFS_INODE_HAS_ORPHAN_ITEM 5 #define BTRFS_INODE_HAS_ASYNC_EXTENT 6 +#define BTRFS_INODE_NEEDS_FULL_SYNC 7 /* in memory btrfs inode */ struct btrfs_inode { @@ -143,6 +144,9 @@ struct btrfs_inode { /* flags field from the on disk inode */ u32 flags; + /* a local copy of root's last_log_commit */ + unsigned long last_log_commit; + /* * Counters to keep track of the number of extent item's we may use due * to delalloc and such. outstanding_extents is the number of extent @@ -202,15 +206,10 @@ static inline bool btrfs_is_free_space_inode(struct inode *inode) static inline int btrfs_inode_in_log(struct inode *inode, u64 generation) { - struct btrfs_root *root = BTRFS_I(inode)->root; - int ret = 0; - - mutex_lock(&root->log_mutex); if (BTRFS_I(inode)->logged_trans == generation && - BTRFS_I(inode)->last_sub_trans <= root->last_log_commit) - ret = 1; - mutex_unlock(&root->log_mutex); - return ret; + BTRFS_I(inode)->last_sub_trans <= BTRFS_I(inode)->last_log_commit) + return 1; + return 0; } #endif diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c index 9197e2e33407..5a3e45db642a 100644 --- a/fs/btrfs/check-integrity.c +++ b/fs/btrfs/check-integrity.c @@ -37,8 +37,9 @@ * the file system was mounted, (i.e., they have been * referenced by the super block) or they have been * written since then and the write completion callback - * was called and a FLUSH request to the device where - * these blocks are located was received and completed. + * was called and no write error was indicated and a + * FLUSH request to the device where these blocks are + * located was received and completed. * 2b. All referenced blocks need to have a generation * number which is equal to the parent's number. * @@ -2601,6 +2602,17 @@ static int btrfsic_check_all_ref_blocks(struct btrfsic_state *state, (unsigned long long)l->block_ref_to->dev_bytenr, l->block_ref_to->mirror_num); ret = -1; + } else if (l->block_ref_to->iodone_w_error) { + printk(KERN_INFO "btrfs: attempt to write superblock" + " which references block %c @%llu (%s/%llu/%d)" + " which has write error!\n", + btrfsic_get_block_type(state, l->block_ref_to), + (unsigned long long) + l->block_ref_to->logical_bytenr, + l->block_ref_to->dev_state->name, + (unsigned long long)l->block_ref_to->dev_bytenr, + l->block_ref_to->mirror_num); + ret = -1; } else if (l->parent_generation != l->block_ref_to->generation && BTRFSIC_GENERATION_UNKNOWN != diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index 43d1c5a3a030..c6467aa88bee 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -577,6 +577,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio, u64 em_start; struct extent_map *em; int ret = -ENOMEM; + int faili = 0; u32 *sums; tree = &BTRFS_I(inode)->io_tree; @@ -626,9 +627,13 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio, for (pg_index = 0; pg_index < nr_pages; pg_index++) { cb->compressed_pages[pg_index] = alloc_page(GFP_NOFS | __GFP_HIGHMEM); - if (!cb->compressed_pages[pg_index]) + if (!cb->compressed_pages[pg_index]) { + faili = pg_index - 1; + ret = -ENOMEM; goto fail2; + } } + faili = nr_pages - 1; cb->nr_pages = nr_pages; add_ra_bio_pages(inode, em_start + em_len, cb); @@ -713,8 +718,10 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio, return 0; fail2: - for (pg_index = 0; pg_index < nr_pages; pg_index++) - free_page((unsigned long)cb->compressed_pages[pg_index]); + while (faili >= 0) { + __free_page(cb->compressed_pages[faili]); + faili--; + } kfree(cb->compressed_pages); fail1: diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index 6d183f60d63a..cdfb4c49a806 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -596,6 +596,11 @@ tree_mod_log_insert_move(struct btrfs_fs_info *fs_info, if (tree_mod_dont_log(fs_info, eb)) return 0; + /* + * When we override something during the move, we log these removals. + * This can only happen when we move towards the beginning of the + * buffer, i.e. dst_slot < src_slot. + */ for (i = 0; i + dst_slot < src_slot && i < nr_items; i++) { ret = tree_mod_log_insert_key_locked(fs_info, eb, i + dst_slot, MOD_LOG_KEY_REMOVE_WHILE_MOVING); @@ -647,8 +652,6 @@ tree_mod_log_insert_root(struct btrfs_fs_info *fs_info, if (tree_mod_dont_log(fs_info, NULL)) return 0; - __tree_mod_log_free_eb(fs_info, old_root); - ret = tree_mod_alloc(fs_info, flags, &tm); if (ret < 0) goto out; @@ -926,12 +929,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans, ret = btrfs_dec_ref(trans, root, buf, 1, 1); BUG_ON(ret); /* -ENOMEM */ } - /* - * don't log freeing in case we're freeing the root node, this - * is done by tree_mod_log_set_root_pointer later - */ - if (buf != root->node && btrfs_header_level(buf) != 0) - tree_mod_log_free_eb(root->fs_info, buf); + tree_mod_log_free_eb(root->fs_info, buf); clean_tree_block(trans, root, buf); *last_ref = 1; } @@ -1225,6 +1223,8 @@ tree_mod_log_rewind(struct btrfs_fs_info *fs_info, struct extent_buffer *eb, free_extent_buffer(eb); __tree_mod_log_rewind(eb_rewin, time_seq, tm); + WARN_ON(btrfs_header_nritems(eb_rewin) > + BTRFS_NODEPTRS_PER_BLOCK(fs_info->fs_root)); return eb_rewin; } @@ -1241,9 +1241,11 @@ get_old_root(struct btrfs_root *root, u64 time_seq) { struct tree_mod_elem *tm; struct extent_buffer *eb; + struct extent_buffer *old; struct tree_mod_root *old_root = NULL; u64 old_generation = 0; u64 logical; + u32 blocksize; eb = btrfs_read_lock_root_node(root); tm = __tree_mod_log_oldest_root(root->fs_info, root, time_seq); @@ -1259,14 +1261,32 @@ get_old_root(struct btrfs_root *root, u64 time_seq) } tm = tree_mod_log_search(root->fs_info, logical, time_seq); - if (old_root) + if (old_root && tm && tm->op != MOD_LOG_KEY_REMOVE_WHILE_FREEING) { + btrfs_tree_read_unlock(root->node); + free_extent_buffer(root->node); + blocksize = btrfs_level_size(root, old_root->level); + old = read_tree_block(root, logical, blocksize, 0); + if (!old) { + pr_warn("btrfs: failed to read tree block %llu from get_old_root\n", + logical); + WARN_ON(1); + } else { + eb = btrfs_clone_extent_buffer(old); + free_extent_buffer(old); + } + } else if (old_root) { + btrfs_tree_read_unlock(root->node); + free_extent_buffer(root->node); eb = alloc_dummy_extent_buffer(logical, root->nodesize); - else + } else { eb = btrfs_clone_extent_buffer(root->node); - btrfs_tree_read_unlock(root->node); - free_extent_buffer(root->node); + btrfs_tree_read_unlock(root->node); + free_extent_buffer(root->node); + } + if (!eb) return NULL; + extent_buffer_get(eb); btrfs_tree_read_lock(eb); if (old_root) { btrfs_set_header_bytenr(eb, eb->start); @@ -1279,11 +1299,28 @@ get_old_root(struct btrfs_root *root, u64 time_seq) __tree_mod_log_rewind(eb, time_seq, tm); else WARN_ON(btrfs_header_level(eb) != 0); - extent_buffer_get(eb); + WARN_ON(btrfs_header_nritems(eb) > BTRFS_NODEPTRS_PER_BLOCK(root)); return eb; } +int btrfs_old_root_level(struct btrfs_root *root, u64 time_seq) +{ + struct tree_mod_elem *tm; + int level; + + tm = __tree_mod_log_oldest_root(root->fs_info, root, time_seq); + if (tm && tm->op == MOD_LOG_ROOT_REPLACE) { + level = tm->old_root.level; + } else { + rcu_read_lock(); + level = btrfs_header_level(root->node); + rcu_read_unlock(); + } + + return level; +} + static inline int should_cow_block(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct extent_buffer *buf) @@ -1725,6 +1762,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans, goto enospc; } + tree_mod_log_free_eb(root->fs_info, root->node); tree_mod_log_set_root_pointer(root, child); rcu_assign_pointer(root->node, child); @@ -2970,8 +3008,10 @@ static int push_node_left(struct btrfs_trans_handle *trans, push_items * sizeof(struct btrfs_key_ptr)); if (push_items < src_nritems) { - tree_mod_log_eb_move(root->fs_info, src, 0, push_items, - src_nritems - push_items); + /* + * don't call tree_mod_log_eb_move here, key removal was already + * fully logged by tree_mod_log_eb_copy above. + */ memmove_extent_buffer(src, btrfs_node_key_ptr_offset(0), btrfs_node_key_ptr_offset(push_items), (src_nritems - push_items) * @@ -4402,149 +4442,6 @@ void btrfs_extend_item(struct btrfs_trans_handle *trans, } /* - * Given a key and some data, insert items into the tree. - * This does all the path init required, making room in the tree if needed. - * Returns the number of keys that were inserted. - */ -int btrfs_insert_some_items(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - struct btrfs_path *path, - struct btrfs_key *cpu_key, u32 *data_size, - int nr) -{ - struct extent_buffer *leaf; - struct btrfs_item *item; - int ret = 0; - int slot; - int i; - u32 nritems; - u32 total_data = 0; - u32 total_size = 0; - unsigned int data_end; - struct btrfs_disk_key disk_key; - struct btrfs_key found_key; - struct btrfs_map_token token; - - btrfs_init_map_token(&token); - - for (i = 0; i < nr; i++) { - if (total_size + data_size[i] + sizeof(struct btrfs_item) > - BTRFS_LEAF_DATA_SIZE(root)) { - break; - nr = i; - } - total_data += data_size[i]; - total_size += data_size[i] + sizeof(struct btrfs_item); - } - BUG_ON(nr == 0); - - ret = btrfs_search_slot(trans, root, cpu_key, path, total_size, 1); - if (ret == 0) - return -EEXIST; - if (ret < 0) - goto out; - - leaf = path->nodes[0]; - - nritems = btrfs_header_nritems(leaf); - data_end = leaf_data_end(root, leaf); - - if (btrfs_leaf_free_space(root, leaf) < total_size) { - for (i = nr; i >= 0; i--) { - total_data -= data_size[i]; - total_size -= data_size[i] + sizeof(struct btrfs_item); - if (total_size < btrfs_leaf_free_space(root, leaf)) - break; - } - nr = i; - } - - slot = path->slots[0]; - BUG_ON(slot < 0); - - if (slot != nritems) { - unsigned int old_data = btrfs_item_end_nr(leaf, slot); - - item = btrfs_item_nr(leaf, slot); - btrfs_item_key_to_cpu(leaf, &found_key, slot); - - /* figure out how many keys we can insert in here */ - total_data = data_size[0]; - for (i = 1; i < nr; i++) { - if (btrfs_comp_cpu_keys(&found_key, cpu_key + i) <= 0) - break; - total_data += data_size[i]; - } - nr = i; - - if (old_data < data_end) { - btrfs_print_leaf(root, leaf); - printk(KERN_CRIT "slot %d old_data %d data_end %d\n", - slot, old_data, data_end); - BUG_ON(1); - } - /* - * item0..itemN ... dataN.offset..dataN.size .. data0.size - */ - /* first correct the data pointers */ - for (i = slot; i < nritems; i++) { - u32 ioff; - - item = btrfs_item_nr(leaf, i); - ioff = btrfs_token_item_offset(leaf, item, &token); - btrfs_set_token_item_offset(leaf, item, - ioff - total_data, &token); - } - /* shift the items */ - memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + nr), - btrfs_item_nr_offset(slot), - (nritems - slot) * sizeof(struct btrfs_item)); - - /* shift the data */ - memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + - data_end - total_data, btrfs_leaf_data(leaf) + - data_end, old_data - data_end); - data_end = old_data; - } else { - /* - * this sucks but it has to be done, if we are inserting at - * the end of the leaf only insert 1 of the items, since we - * have no way of knowing whats on the next leaf and we'd have - * to drop our current locks to figure it out - */ - nr = 1; - } - - /* setup the item for the new data */ - for (i = 0; i < nr; i++) { - btrfs_cpu_key_to_disk(&disk_key, cpu_key + i); - btrfs_set_item_key(leaf, &disk_key, slot + i); - item = btrfs_item_nr(leaf, slot + i); - btrfs_set_token_item_offset(leaf, item, - data_end - data_size[i], &token); - data_end -= data_size[i]; - btrfs_set_token_item_size(leaf, item, data_size[i], &token); - } - btrfs_set_header_nritems(leaf, nritems + nr); - btrfs_mark_buffer_dirty(leaf); - - ret = 0; - if (slot == 0) { - btrfs_cpu_key_to_disk(&disk_key, cpu_key); - fixup_low_keys(trans, root, path, &disk_key, 1); - } - - if (btrfs_leaf_free_space(root, leaf) < 0) { - btrfs_print_leaf(root, leaf); - BUG(); - } -out: - if (!ret) - ret = nr; - return ret; -} - -/* * this is a helper for btrfs_insert_empty_items, the main goal here is * to save stack depth by doing the bulk of the work in a function * that doesn't call btrfs_search_slot @@ -5073,6 +4970,7 @@ static void tree_move_down(struct btrfs_root *root, struct btrfs_path *path, int *level, int root_level) { + BUG_ON(*level == 0); path->nodes[*level - 1] = read_node_slot(root, path->nodes[*level], path->slots[*level]); path->slots[*level - 1] = 0; @@ -5089,7 +4987,7 @@ static int tree_move_next_or_upnext(struct btrfs_root *root, path->slots[*level]++; - while (path->slots[*level] == nritems) { + while (path->slots[*level] >= nritems) { if (*level == root_level) return -1; @@ -5433,9 +5331,11 @@ int btrfs_compare_trees(struct btrfs_root *left_root, goto out; advance_right = ADVANCE; } else { + WARN_ON(!extent_buffer_uptodate(left_path->nodes[0])); ret = tree_compare_item(left_root, left_path, right_path, tmp_buf); if (ret) { + WARN_ON(!extent_buffer_uptodate(left_path->nodes[0])); ret = changed_cb(left_root, right_root, left_path, right_path, &left_key, diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 9821b672f5a2..c72ead869507 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -154,6 +154,13 @@ struct btrfs_ordered_sum; */ #define BTRFS_NAME_LEN 255 +/* + * Theoretical limit is larger, but we keep this down to a sane + * value. That should limit greatly the possibility of collisions on + * inode ref items. + */ +#define BTRFS_LINK_MAX 65535U + /* 32 bytes in various csum fields */ #define BTRFS_CSUM_SIZE 32 @@ -489,6 +496,8 @@ struct btrfs_super_block { */ #define BTRFS_FEATURE_INCOMPAT_BIG_METADATA (1ULL << 5) +#define BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF (1ULL << 6) + #define BTRFS_FEATURE_COMPAT_SUPP 0ULL #define BTRFS_FEATURE_COMPAT_RO_SUPP 0ULL #define BTRFS_FEATURE_INCOMPAT_SUPP \ @@ -496,7 +505,8 @@ struct btrfs_super_block { BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL | \ BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS | \ BTRFS_FEATURE_INCOMPAT_BIG_METADATA | \ - BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO) + BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO | \ + BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF) /* * A leaf is full of items. offset and size tell us where to find @@ -643,6 +653,14 @@ struct btrfs_inode_ref { /* name goes here */ } __attribute__ ((__packed__)); +struct btrfs_inode_extref { + __le64 parent_objectid; + __le64 index; + __le16 name_len; + __u8 name[0]; + /* name goes here */ +} __attribute__ ((__packed__)); + struct btrfs_timespec { __le64 sec; __le32 nsec; @@ -1028,12 +1046,22 @@ struct btrfs_space_info { wait_queue_head_t wait; }; +#define BTRFS_BLOCK_RSV_GLOBAL 1 +#define BTRFS_BLOCK_RSV_DELALLOC 2 +#define BTRFS_BLOCK_RSV_TRANS 3 +#define BTRFS_BLOCK_RSV_CHUNK 4 +#define BTRFS_BLOCK_RSV_DELOPS 5 +#define BTRFS_BLOCK_RSV_EMPTY 6 +#define BTRFS_BLOCK_RSV_TEMP 7 + struct btrfs_block_rsv { u64 size; u64 reserved; struct btrfs_space_info *space_info; spinlock_t lock; - unsigned int full; + unsigned short full; + unsigned short type; + unsigned short failfast; }; /* @@ -1127,6 +1155,9 @@ struct btrfs_block_group_cache { * Today it will only have one thing on it, but that may change */ struct list_head cluster_list; + + /* For delayed block group creation */ + struct list_head new_bg_list; }; /* delayed seq elem */ @@ -1240,7 +1271,6 @@ struct btrfs_fs_info { struct mutex reloc_mutex; struct list_head trans_list; - struct list_head hashers; struct list_head dead_roots; struct list_head caching_block_groups; @@ -1366,9 +1396,6 @@ struct btrfs_fs_info { struct rb_root defrag_inodes; atomic_t defrag_running; - spinlock_t ref_cache_lock; - u64 total_ref_cache_size; - /* * these three are in extended format (availability of single * chunks is denoted by BTRFS_AVAIL_ALLOC_BIT_SINGLE bit, other @@ -1441,6 +1468,8 @@ struct btrfs_fs_info { /* next backup root to be overwritten */ int backup_root_index; + + int num_tolerated_disk_barrier_failures; }; /* @@ -1481,9 +1510,9 @@ struct btrfs_root { wait_queue_head_t log_commit_wait[2]; atomic_t log_writers; atomic_t log_commit[2]; + atomic_t log_batch; unsigned long log_transid; unsigned long last_log_commit; - unsigned long log_batch; pid_t log_start_pid; bool log_multiple_pids; @@ -1592,6 +1621,7 @@ struct btrfs_ioctl_defrag_range_args { */ #define BTRFS_INODE_ITEM_KEY 1 #define BTRFS_INODE_REF_KEY 12 +#define BTRFS_INODE_EXTREF_KEY 13 #define BTRFS_XATTR_ITEM_KEY 24 #define BTRFS_ORPHAN_ITEM_KEY 48 /* reserve 2-15 close to the inode for later flexibility */ @@ -1978,6 +2008,13 @@ BTRFS_SETGET_STACK_FUNCS(block_group_flags, BTRFS_SETGET_FUNCS(inode_ref_name_len, struct btrfs_inode_ref, name_len, 16); BTRFS_SETGET_FUNCS(inode_ref_index, struct btrfs_inode_ref, index, 64); +/* struct btrfs_inode_extref */ +BTRFS_SETGET_FUNCS(inode_extref_parent, struct btrfs_inode_extref, + parent_objectid, 64); +BTRFS_SETGET_FUNCS(inode_extref_name_len, struct btrfs_inode_extref, + name_len, 16); +BTRFS_SETGET_FUNCS(inode_extref_index, struct btrfs_inode_extref, index, 64); + /* struct btrfs_inode_item */ BTRFS_SETGET_FUNCS(inode_generation, struct btrfs_inode_item, generation, 64); BTRFS_SETGET_FUNCS(inode_sequence, struct btrfs_inode_item, sequence, 64); @@ -2858,6 +2895,8 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 size); int btrfs_remove_block_group(struct btrfs_trans_handle *trans, struct btrfs_root *root, u64 group_start); +void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans, + struct btrfs_root *root); u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags); u64 btrfs_get_alloc_profile(struct btrfs_root *root, int data); void btrfs_clear_space_info_full(struct btrfs_fs_info *info); @@ -2874,8 +2913,9 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes); void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes); int btrfs_delalloc_reserve_space(struct inode *inode, u64 num_bytes); void btrfs_delalloc_release_space(struct inode *inode, u64 num_bytes); -void btrfs_init_block_rsv(struct btrfs_block_rsv *rsv); -struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_root *root); +void btrfs_init_block_rsv(struct btrfs_block_rsv *rsv, unsigned short type); +struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_root *root, + unsigned short type); void btrfs_free_block_rsv(struct btrfs_root *root, struct btrfs_block_rsv *rsv); int btrfs_block_rsv_add(struct btrfs_root *root, @@ -3080,6 +3120,7 @@ static inline u64 btrfs_inc_tree_mod_seq(struct btrfs_fs_info *fs_info) { return atomic_inc_return(&fs_info->tree_mod_seq); } +int btrfs_old_root_level(struct btrfs_root *root, u64 time_seq); /* root-item.c */ int btrfs_find_root_ref(struct btrfs_root *tree_root, @@ -3172,12 +3213,12 @@ int btrfs_del_inode_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root, const char *name, int name_len, u64 inode_objectid, u64 ref_objectid, u64 *index); -struct btrfs_inode_ref * -btrfs_lookup_inode_ref(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - struct btrfs_path *path, - const char *name, int name_len, - u64 inode_objectid, u64 ref_objectid, int mod); +int btrfs_get_inode_ref_index(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + struct btrfs_path *path, + const char *name, int name_len, + u64 inode_objectid, u64 ref_objectid, int mod, + u64 *ret_index); int btrfs_insert_empty_inode(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct btrfs_path *path, u64 objectid); @@ -3185,6 +3226,19 @@ int btrfs_lookup_inode(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct btrfs_path *path, struct btrfs_key *location, int mod); +struct btrfs_inode_extref * +btrfs_lookup_inode_extref(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + struct btrfs_path *path, + const char *name, int name_len, + u64 inode_objectid, u64 ref_objectid, int ins_len, + int cow); + +int btrfs_find_name_in_ext_backref(struct btrfs_path *path, + u64 ref_objectid, const char *name, + int name_len, + struct btrfs_inode_extref **extref_ret); + /* file-item.c */ int btrfs_del_csums(struct btrfs_trans_handle *trans, struct btrfs_root *root, u64 bytenr, u64 len); @@ -3249,6 +3303,8 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *dir, u64 objectid, const char *name, int name_len); +int btrfs_truncate_page(struct inode *inode, loff_t from, loff_t len, + int front); int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *inode, u64 new_size, @@ -3283,6 +3339,8 @@ struct extent_map *btrfs_get_extent(struct inode *inode, struct page *page, int btrfs_update_inode(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *inode); +int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans, + struct btrfs_root *root, struct inode *inode); int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode); int btrfs_orphan_del(struct btrfs_trans_handle *trans, struct inode *inode); int btrfs_orphan_cleanup(struct btrfs_root *root); @@ -3308,16 +3366,27 @@ void btrfs_inherit_iflags(struct inode *inode, struct inode *dir); int btrfs_defrag_file(struct inode *inode, struct file *file, struct btrfs_ioctl_defrag_range_args *range, u64 newer_than, unsigned long max_pages); +void btrfs_get_block_group_info(struct list_head *groups_list, + struct btrfs_ioctl_space_info *space); + /* file.c */ int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans, struct inode *inode); int btrfs_run_defrag_inodes(struct btrfs_fs_info *fs_info); int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync); -int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, - int skip_pinned); +void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, + int skip_pinned); +int btrfs_replace_extent_cache(struct inode *inode, struct extent_map *replace, + u64 start, u64 end, int skip_pinned, + int modified); extern const struct file_operations btrfs_file_operations; -int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode, - u64 start, u64 end, u64 *hint_byte, int drop_cache); +int __btrfs_drop_extents(struct btrfs_trans_handle *trans, + struct btrfs_root *root, struct inode *inode, + struct btrfs_path *path, u64 start, u64 end, + u64 *drop_end, int drop_cache); +int btrfs_drop_extents(struct btrfs_trans_handle *trans, + struct btrfs_root *root, struct inode *inode, u64 start, + u64 end, int drop_cache); int btrfs_mark_extent_written(struct btrfs_trans_handle *trans, struct inode *inode, u64 start, u64 end); int btrfs_release_file(struct inode *inode, struct file *file); @@ -3378,6 +3447,11 @@ static inline void __btrfs_set_fs_incompat(struct btrfs_fs_info *fs_info, } } +/* + * Call btrfs_abort_transaction as early as possible when an error condition is + * detected, that way the exact line number is reported. + */ + #define btrfs_abort_transaction(trans, root, errno) \ do { \ __btrfs_abort_transaction(trans, root, __func__, \ diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index 52c85e2b95d0..478f66bdc57b 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -29,7 +29,7 @@ static struct kmem_cache *delayed_node_cache; int __init btrfs_delayed_inode_init(void) { - delayed_node_cache = kmem_cache_create("delayed_node", + delayed_node_cache = kmem_cache_create("btrfs_delayed_node", sizeof(struct btrfs_delayed_node), 0, SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, @@ -650,7 +650,7 @@ static int btrfs_delayed_inode_reserve_metadata( * we're accounted for. */ if (!src_rsv || (!trans->bytes_reserved && - src_rsv != &root->fs_info->delalloc_block_rsv)) { + src_rsv->type != BTRFS_BLOCK_RSV_DELALLOC)) { ret = btrfs_block_rsv_add_noflush(root, dst_rsv, num_bytes); /* * Since we're under a transaction reserve_metadata_bytes could @@ -668,7 +668,7 @@ static int btrfs_delayed_inode_reserve_metadata( num_bytes, 1); } return ret; - } else if (src_rsv == &root->fs_info->delalloc_block_rsv) { + } else if (src_rsv->type == BTRFS_BLOCK_RSV_DELALLOC) { spin_lock(&BTRFS_I(inode)->lock); if (test_and_clear_bit(BTRFS_INODE_DELALLOC_META_RESERVED, &BTRFS_I(inode)->runtime_flags)) { diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 22e98e04c2ea..7cda51995c1e 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -46,6 +46,10 @@ #include "check-integrity.h" #include "rcu-string.h" +#ifdef CONFIG_X86 +#include <asm/cpufeature.h> +#endif + static struct extent_io_ops btree_extent_io_ops; static void end_workqueue_fn(struct btrfs_work *work); static void free_fs_root(struct btrfs_root *root); @@ -217,26 +221,16 @@ static struct extent_map *btree_get_extent(struct inode *inode, write_lock(&em_tree->lock); ret = add_extent_mapping(em_tree, em); if (ret == -EEXIST) { - u64 failed_start = em->start; - u64 failed_len = em->len; - free_extent_map(em); em = lookup_extent_mapping(em_tree, start, len); - if (em) { - ret = 0; - } else { - em = lookup_extent_mapping(em_tree, failed_start, - failed_len); - ret = -EIO; - } + if (!em) + em = ERR_PTR(-EIO); } else if (ret) { free_extent_map(em); - em = NULL; + em = ERR_PTR(ret); } write_unlock(&em_tree->lock); - if (ret) - em = ERR_PTR(ret); out: return em; } @@ -439,10 +433,6 @@ static int csum_dirty_buffer(struct btrfs_root *root, struct page *page) WARN_ON(1); return 0; } - if (eb->pages[0] != page) { - WARN_ON(1); - return 0; - } if (!PageUptodate(page)) { WARN_ON(1); return 0; @@ -869,10 +859,22 @@ static int __btree_submit_bio_done(struct inode *inode, int rw, struct bio *bio, return btrfs_map_bio(BTRFS_I(inode)->root, rw, bio, mirror_num, 1); } +static int check_async_write(struct inode *inode, unsigned long bio_flags) +{ + if (bio_flags & EXTENT_BIO_TREE_LOG) + return 0; +#ifdef CONFIG_X86 + if (cpu_has_xmm4_2) + return 0; +#endif + return 1; +} + static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio, int mirror_num, unsigned long bio_flags, u64 bio_offset) { + int async = check_async_write(inode, bio_flags); int ret; if (!(rw & REQ_WRITE)) { @@ -887,6 +889,12 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio, return ret; return btrfs_map_bio(BTRFS_I(inode)->root, rw, bio, mirror_num, 0); + } else if (!async) { + ret = btree_csum_one_bio(bio); + if (ret) + return ret; + return btrfs_map_bio(BTRFS_I(inode)->root, rw, bio, + mirror_num, 0); } /* @@ -1168,8 +1176,8 @@ static void __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize, atomic_set(&root->log_commit[0], 0); atomic_set(&root->log_commit[1], 0); atomic_set(&root->log_writers, 0); + atomic_set(&root->log_batch, 0); atomic_set(&root->orphan_inodes, 0); - root->log_batch = 0; root->log_transid = 0; root->last_log_commit = 0; extent_io_tree_init(&root->dirty_log_pages, @@ -1667,9 +1675,10 @@ static int transaction_kthread(void *arg) spin_unlock(&root->fs_info->trans_lock); /* If the file system is aborted, this will always fail. */ - trans = btrfs_join_transaction(root); + trans = btrfs_attach_transaction(root); if (IS_ERR(trans)) { - cannot_commit = true; + if (PTR_ERR(trans) != -ENOENT) + cannot_commit = true; goto sleep; } if (transid == trans->transid) { @@ -1994,13 +2003,11 @@ int open_ctree(struct super_block *sb, INIT_LIST_HEAD(&fs_info->trans_list); INIT_LIST_HEAD(&fs_info->dead_roots); INIT_LIST_HEAD(&fs_info->delayed_iputs); - INIT_LIST_HEAD(&fs_info->hashers); INIT_LIST_HEAD(&fs_info->delalloc_inodes); INIT_LIST_HEAD(&fs_info->ordered_operations); INIT_LIST_HEAD(&fs_info->caching_block_groups); spin_lock_init(&fs_info->delalloc_lock); spin_lock_init(&fs_info->trans_lock); - spin_lock_init(&fs_info->ref_cache_lock); spin_lock_init(&fs_info->fs_roots_radix_lock); spin_lock_init(&fs_info->delayed_iput_lock); spin_lock_init(&fs_info->defrag_inodes_lock); @@ -2014,12 +2021,15 @@ int open_ctree(struct super_block *sb, INIT_LIST_HEAD(&fs_info->space_info); INIT_LIST_HEAD(&fs_info->tree_mod_seq_list); btrfs_mapping_init(&fs_info->mapping_tree); - btrfs_init_block_rsv(&fs_info->global_block_rsv); - btrfs_init_block_rsv(&fs_info->delalloc_block_rsv); - btrfs_init_block_rsv(&fs_info->trans_block_rsv); - btrfs_init_block_rsv(&fs_info->chunk_block_rsv); - btrfs_init_block_rsv(&fs_info->empty_block_rsv); - btrfs_init_block_rsv(&fs_info->delayed_block_rsv); + btrfs_init_block_rsv(&fs_info->global_block_rsv, + BTRFS_BLOCK_RSV_GLOBAL); + btrfs_init_block_rsv(&fs_info->delalloc_block_rsv, + BTRFS_BLOCK_RSV_DELALLOC); + btrfs_init_block_rsv(&fs_info->trans_block_rsv, BTRFS_BLOCK_RSV_TRANS); + btrfs_init_block_rsv(&fs_info->chunk_block_rsv, BTRFS_BLOCK_RSV_CHUNK); + btrfs_init_block_rsv(&fs_info->empty_block_rsv, BTRFS_BLOCK_RSV_EMPTY); + btrfs_init_block_rsv(&fs_info->delayed_block_rsv, + BTRFS_BLOCK_RSV_DELOPS); atomic_set(&fs_info->nr_async_submits, 0); atomic_set(&fs_info->async_delalloc_pages, 0); atomic_set(&fs_info->async_submit_draining, 0); @@ -2491,6 +2501,8 @@ retry_root_backup: printk(KERN_ERR "Failed to read block groups: %d\n", ret); goto fail_block_groups; } + fs_info->num_tolerated_disk_barrier_failures = + btrfs_calc_num_tolerated_disk_barrier_failures(fs_info); fs_info->cleaner_kthread = kthread_run(cleaner_kthread, tree_root, "btrfs-cleaner"); @@ -2874,12 +2886,10 @@ static int write_dev_flush(struct btrfs_device *device, int wait) printk_in_rcu("btrfs: disabling barriers on dev %s\n", rcu_str_deref(device->name)); device->nobarriers = 1; - } - if (!bio_flagged(bio, BIO_UPTODATE)) { + } else if (!bio_flagged(bio, BIO_UPTODATE)) { ret = -EIO; - if (!bio_flagged(bio, BIO_EOPNOTSUPP)) - btrfs_dev_stat_inc_and_print(device, - BTRFS_DEV_STAT_FLUSH_ERRS); + btrfs_dev_stat_inc_and_print(device, + BTRFS_DEV_STAT_FLUSH_ERRS); } /* drop the reference from the wait == 0 run */ @@ -2918,14 +2928,15 @@ static int barrier_all_devices(struct btrfs_fs_info *info) { struct list_head *head; struct btrfs_device *dev; - int errors = 0; + int errors_send = 0; + int errors_wait = 0; int ret; /* send down all the barriers */ head = &info->fs_devices->devices; list_for_each_entry_rcu(dev, head, dev_list) { if (!dev->bdev) { - errors++; + errors_send++; continue; } if (!dev->in_fs_metadata || !dev->writeable) @@ -2933,13 +2944,13 @@ static int barrier_all_devices(struct btrfs_fs_info *info) ret = write_dev_flush(dev, 0); if (ret) - errors++; + errors_send++; } /* wait for all the barriers */ list_for_each_entry_rcu(dev, head, dev_list) { if (!dev->bdev) { - errors++; + errors_wait++; continue; } if (!dev->in_fs_metadata || !dev->writeable) @@ -2947,13 +2958,87 @@ static int barrier_all_devices(struct btrfs_fs_info *info) ret = write_dev_flush(dev, 1); if (ret) - errors++; + errors_wait++; } - if (errors) + if (errors_send > info->num_tolerated_disk_barrier_failures || + errors_wait > info->num_tolerated_disk_barrier_failures) return -EIO; return 0; } +int btrfs_calc_num_tolerated_disk_barrier_failures( + struct btrfs_fs_info *fs_info) +{ + struct btrfs_ioctl_space_info space; + struct btrfs_space_info *sinfo; + u64 types[] = {BTRFS_BLOCK_GROUP_DATA, + BTRFS_BLOCK_GROUP_SYSTEM, + BTRFS_BLOCK_GROUP_METADATA, + BTRFS_BLOCK_GROUP_DATA | BTRFS_BLOCK_GROUP_METADATA}; + int num_types = 4; + int i; + int c; + int num_tolerated_disk_barrier_failures = + (int)fs_info->fs_devices->num_devices; + + for (i = 0; i < num_types; i++) { + struct btrfs_space_info *tmp; + + sinfo = NULL; + rcu_read_lock(); + list_for_each_entry_rcu(tmp, &fs_info->space_info, list) { + if (tmp->flags == types[i]) { + sinfo = tmp; + break; + } + } + rcu_read_unlock(); + + if (!sinfo) + continue; + + down_read(&sinfo->groups_sem); + for (c = 0; c < BTRFS_NR_RAID_TYPES; c++) { + if (!list_empty(&sinfo->block_groups[c])) { + u64 flags; + + btrfs_get_block_group_info( + &sinfo->block_groups[c], &space); + if (space.total_bytes == 0 || + space.used_bytes == 0) + continue; + flags = space.flags; + /* + * return + * 0: if dup, single or RAID0 is configured for + * any of metadata, system or data, else + * 1: if RAID5 is configured, or if RAID1 or + * RAID10 is configured and only two mirrors + * are used, else + * 2: if RAID6 is configured, else + * num_mirrors - 1: if RAID1 or RAID10 is + * configured and more than + * 2 mirrors are used. + */ + if (num_tolerated_disk_barrier_failures > 0 && + ((flags & (BTRFS_BLOCK_GROUP_DUP | + BTRFS_BLOCK_GROUP_RAID0)) || + ((flags & BTRFS_BLOCK_GROUP_PROFILE_MASK) + == 0))) + num_tolerated_disk_barrier_failures = 0; + else if (num_tolerated_disk_barrier_failures > 1 + && + (flags & (BTRFS_BLOCK_GROUP_RAID1 | + BTRFS_BLOCK_GROUP_RAID10))) + num_tolerated_disk_barrier_failures = 1; + } + } + up_read(&sinfo->groups_sem); + } + + return num_tolerated_disk_barrier_failures; +} + int write_all_supers(struct btrfs_root *root, int max_mirrors) { struct list_head *head; @@ -2976,8 +3061,16 @@ int write_all_supers(struct btrfs_root *root, int max_mirrors) mutex_lock(&root->fs_info->fs_devices->device_list_mutex); head = &root->fs_info->fs_devices->devices; - if (do_barriers) - barrier_all_devices(root->fs_info); + if (do_barriers) { + ret = barrier_all_devices(root->fs_info); + if (ret) { + mutex_unlock( + &root->fs_info->fs_devices->device_list_mutex); + btrfs_error(root->fs_info, ret, + "errors while submitting device barriers."); + return ret; + } + } list_for_each_entry_rcu(dev, head, dev_list) { if (!dev->bdev) { @@ -3211,10 +3304,6 @@ int close_ctree(struct btrfs_root *root) printk(KERN_INFO "btrfs: at unmount delalloc count %llu\n", (unsigned long long)fs_info->delalloc_bytes); } - if (fs_info->total_ref_cache_size) { - printk(KERN_INFO "btrfs: at umount reference cache size %llu\n", - (unsigned long long)fs_info->total_ref_cache_size); - } free_extent_buffer(fs_info->extent_root->node); free_extent_buffer(fs_info->extent_root->commit_root); @@ -3360,52 +3449,6 @@ int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid) return btree_read_extent_buffer_pages(root, buf, 0, parent_transid); } -int btree_lock_page_hook(struct page *page, void *data, - void (*flush_fn)(void *)) -{ - struct inode *inode = page->mapping->host; - struct btrfs_root *root = BTRFS_I(inode)->root; - struct extent_buffer *eb; - - /* - * We culled this eb but the page is still hanging out on the mapping, - * carry on. - */ - if (!PagePrivate(page)) - goto out; - - eb = (struct extent_buffer *)page->private; - if (!eb) { - WARN_ON(1); - goto out; - } - if (page != eb->pages[0]) - goto out; - - if (!btrfs_try_tree_write_lock(eb)) { - flush_fn(data); - btrfs_tree_lock(eb); - } - btrfs_set_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN); - - if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)) { - spin_lock(&root->fs_info->delalloc_lock); - if (root->fs_info->dirty_metadata_bytes >= eb->len) - root->fs_info->dirty_metadata_bytes -= eb->len; - else - WARN_ON(1); - spin_unlock(&root->fs_info->delalloc_lock); - } - - btrfs_tree_unlock(eb); -out: - if (!trylock_page(page)) { - flush_fn(data); - lock_page(page); - } - return 0; -} - static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info, int read_only) { @@ -3608,7 +3651,7 @@ static int btrfs_destroy_marked_extents(struct btrfs_root *root, while (1) { ret = find_first_extent_bit(dirty_pages, start, &start, &end, - mark); + mark, NULL); if (ret) break; @@ -3663,7 +3706,7 @@ static int btrfs_destroy_pinned_extent(struct btrfs_root *root, again: while (1) { ret = find_first_extent_bit(unpin, 0, &start, &end, - EXTENT_DIRTY); + EXTENT_DIRTY, NULL); if (ret) break; @@ -3800,7 +3843,6 @@ int btrfs_cleanup_transaction(struct btrfs_root *root) } static struct extent_io_ops btree_extent_io_ops = { - .write_cache_pages_lock_hook = btree_lock_page_hook, .readpage_end_io_hook = btree_readpage_end_io_hook, .readpage_io_failed_hook = btree_io_failed_hook, .submit_bio_hook = btree_submit_bio_hook, diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h index c5b00a735fef..2025a9132c16 100644 --- a/fs/btrfs/disk-io.h +++ b/fs/btrfs/disk-io.h @@ -95,6 +95,8 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans, u64 objectid); int btree_lock_page_hook(struct page *page, void *data, void (*flush_fn)(void *)); +int btrfs_calc_num_tolerated_disk_barrier_failures( + struct btrfs_fs_info *fs_info); #ifdef CONFIG_DEBUG_LOCK_ALLOC void btrfs_init_lockdep(void); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index ba58024d40d3..3d3e2c17d8d1 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -94,8 +94,8 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, u64 flags, struct btrfs_disk_key *key, int level, struct btrfs_key *ins); static int do_chunk_alloc(struct btrfs_trans_handle *trans, - struct btrfs_root *extent_root, u64 alloc_bytes, - u64 flags, int force); + struct btrfs_root *extent_root, u64 flags, + int force); static int find_next_key(struct btrfs_path *path, int level, struct btrfs_key *key); static void dump_space_info(struct btrfs_space_info *info, u64 bytes, @@ -312,7 +312,8 @@ static u64 add_new_free_space(struct btrfs_block_group_cache *block_group, while (start < end) { ret = find_first_extent_bit(info->pinned_extents, start, &extent_start, &extent_end, - EXTENT_DIRTY | EXTENT_UPTODATE); + EXTENT_DIRTY | EXTENT_UPTODATE, + NULL); if (ret) break; @@ -2361,10 +2362,6 @@ static noinline int run_clustered_refs(struct btrfs_trans_handle *trans, } next: - do_chunk_alloc(trans, fs_info->extent_root, - 2 * 1024 * 1024, - btrfs_get_alloc_profile(root, 0), - CHUNK_ALLOC_NO_FORCE); cond_resched(); spin_lock(&delayed_refs->lock); } @@ -2478,10 +2475,6 @@ int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans, if (root == root->fs_info->extent_root) root = root->fs_info->tree_root; - do_chunk_alloc(trans, root->fs_info->extent_root, - 2 * 1024 * 1024, btrfs_get_alloc_profile(root, 0), - CHUNK_ALLOC_NO_FORCE); - btrfs_delayed_refs_qgroup_accounting(trans, root->fs_info); delayed_refs = &trans->transaction->delayed_refs; @@ -2551,6 +2544,12 @@ again: } if (run_all) { + if (!list_empty(&trans->new_bgs)) { + spin_unlock(&delayed_refs->lock); + btrfs_create_pending_block_groups(trans, root); + spin_lock(&delayed_refs->lock); + } + node = rb_first(&delayed_refs->root); if (!node) goto out; @@ -3406,7 +3405,6 @@ alloc: return PTR_ERR(trans); ret = do_chunk_alloc(trans, root->fs_info->extent_root, - bytes + 2 * 1024 * 1024, alloc_target, CHUNK_ALLOC_NO_FORCE); btrfs_end_transaction(trans, root); @@ -3488,8 +3486,7 @@ static void force_metadata_allocation(struct btrfs_fs_info *info) } static int should_alloc_chunk(struct btrfs_root *root, - struct btrfs_space_info *sinfo, u64 alloc_bytes, - int force) + struct btrfs_space_info *sinfo, int force) { struct btrfs_block_rsv *global_rsv = &root->fs_info->global_block_rsv; u64 num_bytes = sinfo->total_bytes - sinfo->bytes_readonly; @@ -3504,7 +3501,8 @@ static int should_alloc_chunk(struct btrfs_root *root, * and purposes it's used space. Don't worry about locking the * global_rsv, it doesn't change except when the transaction commits. */ - num_allocated += global_rsv->size; + if (sinfo->flags & BTRFS_BLOCK_GROUP_METADATA) + num_allocated += global_rsv->size; /* * in limited mode, we want to have some free space up to @@ -3518,15 +3516,8 @@ static int should_alloc_chunk(struct btrfs_root *root, if (num_bytes - num_allocated < thresh) return 1; } - thresh = btrfs_super_total_bytes(root->fs_info->super_copy); - /* 256MB or 2% of the FS */ - thresh = max_t(u64, 256 * 1024 * 1024, div_factor_fine(thresh, 2)); - /* system chunks need a much small threshold */ - if (sinfo->flags & BTRFS_BLOCK_GROUP_SYSTEM) - thresh = 32 * 1024 * 1024; - - if (num_bytes > thresh && sinfo->bytes_used < div_factor(num_bytes, 8)) + if (num_allocated + 2 * 1024 * 1024 < div_factor(num_bytes, 8)) return 0; return 1; } @@ -3576,8 +3567,7 @@ static void check_system_chunk(struct btrfs_trans_handle *trans, } static int do_chunk_alloc(struct btrfs_trans_handle *trans, - struct btrfs_root *extent_root, u64 alloc_bytes, - u64 flags, int force) + struct btrfs_root *extent_root, u64 flags, int force) { struct btrfs_space_info *space_info; struct btrfs_fs_info *fs_info = extent_root->fs_info; @@ -3601,7 +3591,7 @@ again: return 0; } - if (!should_alloc_chunk(extent_root, space_info, alloc_bytes, force)) { + if (!should_alloc_chunk(extent_root, space_info, force)) { spin_unlock(&space_info->lock); return 0; } else if (space_info->chunk_alloc) { @@ -3669,6 +3659,46 @@ out: return ret; } +static int can_overcommit(struct btrfs_root *root, + struct btrfs_space_info *space_info, u64 bytes, + int flush) +{ + u64 profile = btrfs_get_alloc_profile(root, 0); + u64 avail; + u64 used; + + used = space_info->bytes_used + space_info->bytes_reserved + + space_info->bytes_pinned + space_info->bytes_readonly + + space_info->bytes_may_use; + + spin_lock(&root->fs_info->free_chunk_lock); + avail = root->fs_info->free_chunk_space; + spin_unlock(&root->fs_info->free_chunk_lock); + + /* + * If we have dup, raid1 or raid10 then only half of the free + * space is actually useable. + */ + if (profile & (BTRFS_BLOCK_GROUP_DUP | + BTRFS_BLOCK_GROUP_RAID1 | + BTRFS_BLOCK_GROUP_RAID10)) + avail >>= 1; + + /* + * If we aren't flushing don't let us overcommit too much, say + * 1/8th of the space. If we can flush, let it overcommit up to + * 1/2 of the space. + */ + if (flush) + avail >>= 3; + else + avail >>= 1; + + if (used + bytes < space_info->total_bytes + avail) + return 1; + return 0; +} + /* * shrink metadata reservation for delalloc */ @@ -3693,7 +3723,7 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig, if (delalloc_bytes == 0) { if (trans) return; - btrfs_wait_ordered_extents(root, 0, 0); + btrfs_wait_ordered_extents(root, 0); return; } @@ -3703,11 +3733,15 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig, writeback_inodes_sb_nr_if_idle(root->fs_info->sb, nr_pages, WB_REASON_FS_FREE_SPACE); + /* + * We need to wait for the async pages to actually start before + * we do anything. + */ + wait_event(root->fs_info->async_submit_wait, + !atomic_read(&root->fs_info->async_delalloc_pages)); + spin_lock(&space_info->lock); - if (space_info->bytes_used + space_info->bytes_reserved + - space_info->bytes_pinned + space_info->bytes_readonly + - space_info->bytes_may_use + orig <= - space_info->total_bytes) { + if (can_overcommit(root, space_info, orig, !trans)) { spin_unlock(&space_info->lock); break; } @@ -3715,7 +3749,7 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig, loops++; if (wait_ordered && !trans) { - btrfs_wait_ordered_extents(root, 0, 0); + btrfs_wait_ordered_extents(root, 0); } else { time_left = schedule_timeout_killable(1); if (time_left) @@ -3784,11 +3818,12 @@ commit: } enum flush_state { - FLUSH_DELALLOC = 1, - FLUSH_DELALLOC_WAIT = 2, - FLUSH_DELAYED_ITEMS_NR = 3, - FLUSH_DELAYED_ITEMS = 4, - COMMIT_TRANS = 5, + FLUSH_DELAYED_ITEMS_NR = 1, + FLUSH_DELAYED_ITEMS = 2, + FLUSH_DELALLOC = 3, + FLUSH_DELALLOC_WAIT = 4, + ALLOC_CHUNK = 5, + COMMIT_TRANS = 6, }; static int flush_space(struct btrfs_root *root, @@ -3800,11 +3835,6 @@ static int flush_space(struct btrfs_root *root, int ret = 0; switch (state) { - case FLUSH_DELALLOC: - case FLUSH_DELALLOC_WAIT: - shrink_delalloc(root, num_bytes, orig_bytes, - state == FLUSH_DELALLOC_WAIT); - break; case FLUSH_DELAYED_ITEMS_NR: case FLUSH_DELAYED_ITEMS: if (state == FLUSH_DELAYED_ITEMS_NR) { @@ -3825,6 +3855,24 @@ static int flush_space(struct btrfs_root *root, ret = btrfs_run_delayed_items_nr(trans, root, nr); btrfs_end_transaction(trans, root); break; + case FLUSH_DELALLOC: + case FLUSH_DELALLOC_WAIT: + shrink_delalloc(root, num_bytes, orig_bytes, + state == FLUSH_DELALLOC_WAIT); + break; + case ALLOC_CHUNK: + trans = btrfs_join_transaction(root); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + break; + } + ret = do_chunk_alloc(trans, root->fs_info->extent_root, + btrfs_get_alloc_profile(root, 0), + CHUNK_ALLOC_NO_FORCE); + btrfs_end_transaction(trans, root); + if (ret == -ENOSPC) + ret = 0; + break; case COMMIT_TRANS: ret = may_commit_transaction(root, space_info, orig_bytes, 0); break; @@ -3856,10 +3904,9 @@ static int reserve_metadata_bytes(struct btrfs_root *root, struct btrfs_space_info *space_info = block_rsv->space_info; u64 used; u64 num_bytes = orig_bytes; - int flush_state = FLUSH_DELALLOC; + int flush_state = FLUSH_DELAYED_ITEMS_NR; int ret = 0; bool flushing = false; - bool committed = false; again: ret = 0; @@ -3922,57 +3969,12 @@ again: (orig_bytes * 2); } - if (ret) { - u64 profile = btrfs_get_alloc_profile(root, 0); - u64 avail; - - /* - * If we have a lot of space that's pinned, don't bother doing - * the overcommit dance yet and just commit the transaction. - */ - avail = (space_info->total_bytes - space_info->bytes_used) * 8; - do_div(avail, 10); - if (space_info->bytes_pinned >= avail && flush && !committed) { - space_info->flush = 1; - flushing = true; - spin_unlock(&space_info->lock); - ret = may_commit_transaction(root, space_info, - orig_bytes, 1); - if (ret) - goto out; - committed = true; - goto again; - } - - spin_lock(&root->fs_info->free_chunk_lock); - avail = root->fs_info->free_chunk_space; - - /* - * If we have dup, raid1 or raid10 then only half of the free - * space is actually useable. - */ - if (profile & (BTRFS_BLOCK_GROUP_DUP | - BTRFS_BLOCK_GROUP_RAID1 | - BTRFS_BLOCK_GROUP_RAID10)) - avail >>= 1; - - /* - * If we aren't flushing don't let us overcommit too much, say - * 1/8th of the space. If we can flush, let it overcommit up to - * 1/2 of the space. - */ - if (flush) - avail >>= 3; - else - avail >>= 1; - spin_unlock(&root->fs_info->free_chunk_lock); - - if (used + num_bytes < space_info->total_bytes + avail) { - space_info->bytes_may_use += orig_bytes; - trace_btrfs_space_reservation(root->fs_info, - "space_info", space_info->flags, orig_bytes, 1); - ret = 0; - } + if (ret && can_overcommit(root, space_info, orig_bytes, flush)) { + space_info->bytes_may_use += orig_bytes; + trace_btrfs_space_reservation(root->fs_info, "space_info", + space_info->flags, orig_bytes, + 1); + ret = 0; } /* @@ -4114,13 +4116,15 @@ static int block_rsv_migrate_bytes(struct btrfs_block_rsv *src, return 0; } -void btrfs_init_block_rsv(struct btrfs_block_rsv *rsv) +void btrfs_init_block_rsv(struct btrfs_block_rsv *rsv, unsigned short type) { memset(rsv, 0, sizeof(*rsv)); spin_lock_init(&rsv->lock); + rsv->type = type; } -struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_root *root) +struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_root *root, + unsigned short type) { struct btrfs_block_rsv *block_rsv; struct btrfs_fs_info *fs_info = root->fs_info; @@ -4129,7 +4133,7 @@ struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_root *root) if (!block_rsv) return NULL; - btrfs_init_block_rsv(block_rsv); + btrfs_init_block_rsv(block_rsv, type); block_rsv->space_info = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA); return block_rsv; @@ -4138,6 +4142,8 @@ struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_root *root) void btrfs_free_block_rsv(struct btrfs_root *root, struct btrfs_block_rsv *rsv) { + if (!rsv) + return; btrfs_block_rsv_release(root, rsv, (u64)-1); kfree(rsv); } @@ -4416,10 +4422,10 @@ int btrfs_snap_reserve_metadata(struct btrfs_trans_handle *trans, struct btrfs_block_rsv *src_rsv = get_block_rsv(trans, root); struct btrfs_block_rsv *dst_rsv = &pending->block_rsv; /* - * two for root back/forward refs, two for directory entries - * and one for root of the snapshot. + * two for root back/forward refs, two for directory entries, + * one for root of the snapshot and one for parent inode. */ - u64 num_bytes = btrfs_calc_trans_metadata_size(root, 5); + u64 num_bytes = btrfs_calc_trans_metadata_size(root, 6); dst_rsv->space_info = src_rsv->space_info; return block_rsv_migrate_bytes(src_rsv, dst_rsv, num_bytes); } @@ -5018,7 +5024,7 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans, while (1) { ret = find_first_extent_bit(unpin, 0, &start, &end, - EXTENT_DIRTY); + EXTENT_DIRTY, NULL); if (ret) break; @@ -5096,8 +5102,10 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, ret = remove_extent_backref(trans, extent_root, path, NULL, refs_to_drop, is_data); - if (ret) - goto abort; + if (ret) { + btrfs_abort_transaction(trans, extent_root, ret); + goto out; + } btrfs_release_path(path); path->leave_spinning = 1; @@ -5115,8 +5123,10 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, btrfs_print_leaf(extent_root, path->nodes[0]); } - if (ret < 0) - goto abort; + if (ret < 0) { + btrfs_abort_transaction(trans, extent_root, ret); + goto out; + } extent_slot = path->slots[0]; } } else if (ret == -ENOENT) { @@ -5130,7 +5140,8 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, (unsigned long long)owner_objectid, (unsigned long long)owner_offset); } else { - goto abort; + btrfs_abort_transaction(trans, extent_root, ret); + goto out; } leaf = path->nodes[0]; @@ -5140,8 +5151,10 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, BUG_ON(found_extent || extent_slot != path->slots[0]); ret = convert_extent_item_v0(trans, extent_root, path, owner_objectid, 0); - if (ret < 0) - goto abort; + if (ret < 0) { + btrfs_abort_transaction(trans, extent_root, ret); + goto out; + } btrfs_release_path(path); path->leave_spinning = 1; @@ -5158,8 +5171,11 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, (unsigned long long)bytenr); btrfs_print_leaf(extent_root, path->nodes[0]); } - if (ret < 0) - goto abort; + if (ret < 0) { + btrfs_abort_transaction(trans, extent_root, ret); + goto out; + } + extent_slot = path->slots[0]; leaf = path->nodes[0]; item_size = btrfs_item_size_nr(leaf, extent_slot); @@ -5196,8 +5212,10 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, ret = remove_extent_backref(trans, extent_root, path, iref, refs_to_drop, is_data); - if (ret) - goto abort; + if (ret) { + btrfs_abort_transaction(trans, extent_root, ret); + goto out; + } } } else { if (found_extent) { @@ -5214,27 +5232,29 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, ret = btrfs_del_items(trans, extent_root, path, path->slots[0], num_to_del); - if (ret) - goto abort; + if (ret) { + btrfs_abort_transaction(trans, extent_root, ret); + goto out; + } btrfs_release_path(path); if (is_data) { ret = btrfs_del_csums(trans, root, bytenr, num_bytes); - if (ret) - goto abort; + if (ret) { + btrfs_abort_transaction(trans, extent_root, ret); + goto out; + } } ret = update_block_group(trans, root, bytenr, num_bytes, 0); - if (ret) - goto abort; + if (ret) { + btrfs_abort_transaction(trans, extent_root, ret); + goto out; + } } out: btrfs_free_path(path); return ret; - -abort: - btrfs_abort_transaction(trans, extent_root, ret); - goto out; } /* @@ -5497,8 +5517,6 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans, struct btrfs_block_group_cache *used_block_group; u64 search_start = 0; int empty_cluster = 2 * 1024 * 1024; - int allowed_chunk_alloc = 0; - int done_chunk_alloc = 0; struct btrfs_space_info *space_info; int loop = 0; int index = 0; @@ -5530,9 +5548,6 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans, if (btrfs_mixed_space_info(space_info)) use_cluster = false; - if (orig_root->ref_cows || empty_size) - allowed_chunk_alloc = 1; - if (data & BTRFS_BLOCK_GROUP_METADATA && use_cluster) { last_ptr = &root->fs_info->meta_alloc_cluster; if (!btrfs_test_opt(root, SSD)) @@ -5806,10 +5821,6 @@ checks: trace_btrfs_reserve_extent(orig_root, block_group, search_start, num_bytes); - if (offset < search_start) - btrfs_add_free_space(used_block_group, offset, - search_start - offset); - BUG_ON(offset > search_start); if (used_block_group != block_group) btrfs_put_block_group(used_block_group); btrfs_put_block_group(block_group); @@ -5842,34 +5853,17 @@ loop: index = 0; loop++; if (loop == LOOP_ALLOC_CHUNK) { - if (allowed_chunk_alloc) { - ret = do_chunk_alloc(trans, root, num_bytes + - 2 * 1024 * 1024, data, - CHUNK_ALLOC_LIMITED); - /* - * Do not bail out on ENOSPC since we - * can do more things. - */ - if (ret < 0 && ret != -ENOSPC) { - btrfs_abort_transaction(trans, - root, ret); - goto out; - } - allowed_chunk_alloc = 0; - if (ret == 1) - done_chunk_alloc = 1; - } else if (!done_chunk_alloc && - space_info->force_alloc == - CHUNK_ALLOC_NO_FORCE) { - space_info->force_alloc = CHUNK_ALLOC_LIMITED; + ret = do_chunk_alloc(trans, root, data, + CHUNK_ALLOC_FORCE); + /* + * Do not bail out on ENOSPC since we + * can do more things. + */ + if (ret < 0 && ret != -ENOSPC) { + btrfs_abort_transaction(trans, + root, ret); + goto out; } - - /* - * We didn't allocate a chunk, go ahead and drop the - * empty size and loop again. - */ - if (!done_chunk_alloc) - loop = LOOP_NO_EMPTY_SIZE; } if (loop == LOOP_NO_EMPTY_SIZE) { @@ -5944,20 +5938,6 @@ int btrfs_reserve_extent(struct btrfs_trans_handle *trans, data = btrfs_get_alloc_profile(root, data); again: - /* - * the only place that sets empty_size is btrfs_realloc_node, which - * is not called recursively on allocations - */ - if (empty_size || root->ref_cows) { - ret = do_chunk_alloc(trans, root->fs_info->extent_root, - num_bytes + 2 * 1024 * 1024, data, - CHUNK_ALLOC_NO_FORCE); - if (ret < 0 && ret != -ENOSPC) { - btrfs_abort_transaction(trans, root, ret); - return ret; - } - } - WARN_ON(num_bytes < root->sectorsize); ret = find_free_extent(trans, root, num_bytes, empty_size, hint_byte, ins, data); @@ -5967,12 +5947,6 @@ again: num_bytes = num_bytes >> 1; num_bytes = num_bytes & ~(root->sectorsize - 1); num_bytes = max(num_bytes, min_alloc_size); - ret = do_chunk_alloc(trans, root->fs_info->extent_root, - num_bytes, data, CHUNK_ALLOC_FORCE); - if (ret < 0 && ret != -ENOSPC) { - btrfs_abort_transaction(trans, root, ret); - return ret; - } if (num_bytes == min_alloc_size) final_tried = true; goto again; @@ -6314,7 +6288,7 @@ use_block_rsv(struct btrfs_trans_handle *trans, ret = block_rsv_use_bytes(block_rsv, blocksize); if (!ret) return block_rsv; - if (ret) { + if (ret && !block_rsv->failfast) { static DEFINE_RATELIMIT_STATE(_rs, DEFAULT_RATELIMIT_INTERVAL, /*DEFAULT_RATELIMIT_BURST*/ 2); @@ -7279,7 +7253,7 @@ int btrfs_set_block_group_ro(struct btrfs_root *root, alloc_flags = update_block_group_flags(root, cache->flags); if (alloc_flags != cache->flags) { - ret = do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, + ret = do_chunk_alloc(trans, root, alloc_flags, CHUNK_ALLOC_FORCE); if (ret < 0) goto out; @@ -7289,7 +7263,7 @@ int btrfs_set_block_group_ro(struct btrfs_root *root, if (!ret) goto out; alloc_flags = get_alloc_profile(root, cache->space_info->flags); - ret = do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, + ret = do_chunk_alloc(trans, root, alloc_flags, CHUNK_ALLOC_FORCE); if (ret < 0) goto out; @@ -7303,7 +7277,7 @@ int btrfs_force_chunk_alloc(struct btrfs_trans_handle *trans, struct btrfs_root *root, u64 type) { u64 alloc_flags = get_alloc_profile(root, type); - return do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, + return do_chunk_alloc(trans, root, alloc_flags, CHUNK_ALLOC_FORCE); } @@ -7810,6 +7784,34 @@ error: return ret; } +void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans, + struct btrfs_root *root) +{ + struct btrfs_block_group_cache *block_group, *tmp; + struct btrfs_root *extent_root = root->fs_info->extent_root; + struct btrfs_block_group_item item; + struct btrfs_key key; + int ret = 0; + + list_for_each_entry_safe(block_group, tmp, &trans->new_bgs, + new_bg_list) { + list_del_init(&block_group->new_bg_list); + + if (ret) + continue; + + spin_lock(&block_group->lock); + memcpy(&item, &block_group->item, sizeof(item)); + memcpy(&key, &block_group->key, sizeof(key)); + spin_unlock(&block_group->lock); + + ret = btrfs_insert_item(trans, extent_root, &key, &item, + sizeof(item)); + if (ret) + btrfs_abort_transaction(trans, extent_root, ret); + } +} + int btrfs_make_block_group(struct btrfs_trans_handle *trans, struct btrfs_root *root, u64 bytes_used, u64 type, u64 chunk_objectid, u64 chunk_offset, @@ -7843,6 +7845,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, spin_lock_init(&cache->lock); INIT_LIST_HEAD(&cache->list); INIT_LIST_HEAD(&cache->cluster_list); + INIT_LIST_HEAD(&cache->new_bg_list); btrfs_init_free_space_ctl(cache); @@ -7874,12 +7877,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, ret = btrfs_add_block_group_cache(root->fs_info, cache); BUG_ON(ret); /* Logic error */ - ret = btrfs_insert_item(trans, extent_root, &cache->key, &cache->item, - sizeof(cache->item)); - if (ret) { - btrfs_abort_transaction(trans, extent_root, ret); - return ret; - } + list_add_tail(&cache->new_bg_list, &trans->new_bgs); set_avail_alloc_bits(extent_root->fs_info, type); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index b08ea4717e9d..472873a94d96 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -45,6 +45,7 @@ struct extent_page_data { struct bio *bio; struct extent_io_tree *tree; get_extent_t *get_extent; + unsigned long bio_flags; /* tells writepage not to lock the state bits for this range * it still does the unlocking @@ -64,13 +65,13 @@ tree_fs_info(struct extent_io_tree *tree) int __init extent_io_init(void) { - extent_state_cache = kmem_cache_create("extent_state", + extent_state_cache = kmem_cache_create("btrfs_extent_state", sizeof(struct extent_state), 0, SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!extent_state_cache) return -ENOMEM; - extent_buffer_cache = kmem_cache_create("extent_buffers", + extent_buffer_cache = kmem_cache_create("btrfs_extent_buffer", sizeof(struct extent_buffer), 0, SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!extent_buffer_cache) @@ -942,6 +943,7 @@ int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, int bits, * @end: the end offset in bytes (inclusive) * @bits: the bits to set in this range * @clear_bits: the bits to clear in this range + * @cached_state: state that we're going to cache * @mask: the allocation mask * * This will go through and set bits for the given range. If any states exist @@ -951,7 +953,8 @@ int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, int bits, * boundary bits like LOCK. */ int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, - int bits, int clear_bits, gfp_t mask) + int bits, int clear_bits, + struct extent_state **cached_state, gfp_t mask) { struct extent_state *state; struct extent_state *prealloc = NULL; @@ -968,6 +971,15 @@ again: } spin_lock(&tree->lock); + if (cached_state && *cached_state) { + state = *cached_state; + if (state->start <= start && state->end > start && + state->tree) { + node = &state->rb_node; + goto hit_next; + } + } + /* * this search will find all the extents that end after * our range starts. @@ -998,6 +1010,7 @@ hit_next: */ if (state->start == start && state->end <= end) { set_state_bits(tree, state, &bits); + cache_state(state, cached_state); state = clear_state_bit(tree, state, &clear_bits, 0); if (last_end == (u64)-1) goto out; @@ -1038,6 +1051,7 @@ hit_next: goto out; if (state->end <= end) { set_state_bits(tree, state, &bits); + cache_state(state, cached_state); state = clear_state_bit(tree, state, &clear_bits, 0); if (last_end == (u64)-1) goto out; @@ -1076,6 +1090,7 @@ hit_next: &bits); if (err) extent_io_tree_panic(tree, err); + cache_state(prealloc, cached_state); prealloc = NULL; start = this_end + 1; goto search_again; @@ -1098,6 +1113,7 @@ hit_next: extent_io_tree_panic(tree, err); set_state_bits(tree, prealloc, &bits); + cache_state(prealloc, cached_state); clear_state_bit(tree, prealloc, &clear_bits, 0); prealloc = NULL; goto out; @@ -1150,6 +1166,14 @@ int set_extent_delalloc(struct extent_io_tree *tree, u64 start, u64 end, NULL, cached_state, mask); } +int set_extent_defrag(struct extent_io_tree *tree, u64 start, u64 end, + struct extent_state **cached_state, gfp_t mask) +{ + return set_extent_bit(tree, start, end, + EXTENT_DELALLOC | EXTENT_UPTODATE | EXTENT_DEFRAG, + NULL, cached_state, mask); +} + int clear_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end, gfp_t mask) { @@ -1294,18 +1318,42 @@ out: * If nothing was found, 1 is returned. If found something, return 0. */ int find_first_extent_bit(struct extent_io_tree *tree, u64 start, - u64 *start_ret, u64 *end_ret, int bits) + u64 *start_ret, u64 *end_ret, int bits, + struct extent_state **cached_state) { struct extent_state *state; + struct rb_node *n; int ret = 1; spin_lock(&tree->lock); + if (cached_state && *cached_state) { + state = *cached_state; + if (state->end == start - 1 && state->tree) { + n = rb_next(&state->rb_node); + while (n) { + state = rb_entry(n, struct extent_state, + rb_node); + if (state->state & bits) + goto got_it; + n = rb_next(n); + } + free_extent_state(*cached_state); + *cached_state = NULL; + goto out; + } + free_extent_state(*cached_state); + *cached_state = NULL; + } + state = find_first_extent_bit_state(tree, start, bits); +got_it: if (state) { + cache_state(state, cached_state); *start_ret = state->start; *end_ret = state->end; ret = 0; } +out: spin_unlock(&tree->lock); return ret; } @@ -2068,7 +2116,7 @@ static int bio_readpage_error(struct bio *failed_bio, struct page *page, } read_unlock(&em_tree->lock); - if (!em || IS_ERR(em)) { + if (!em) { kfree(failrec); return -EIO; } @@ -2304,8 +2352,8 @@ static void end_bio_extent_readpage(struct bio *bio, int err) struct extent_state *cached = NULL; struct extent_state *state; - pr_debug("end_bio_extent_readpage: bi_vcnt=%d, idx=%d, err=%d, " - "mirror=%ld\n", bio->bi_vcnt, bio->bi_idx, err, + pr_debug("end_bio_extent_readpage: bi_sector=%llu, err=%d, " + "mirror=%ld\n", (u64)bio->bi_sector, err, (long int)bio->bi_bdev); tree = &BTRFS_I(page->mapping->host)->io_tree; @@ -2709,12 +2757,15 @@ static int __extent_read_full_page(struct extent_io_tree *tree, end_bio_extent_readpage, mirror_num, *bio_flags, this_bio_flag); - BUG_ON(ret == -ENOMEM); - nr++; - *bio_flags = this_bio_flag; + if (!ret) { + nr++; + *bio_flags = this_bio_flag; + } } - if (ret) + if (ret) { SetPageError(page); + unlock_extent(tree, cur, cur + iosize - 1); + } cur = cur + iosize; pg_offset += iosize; } @@ -3161,12 +3212,16 @@ static int write_one_eb(struct extent_buffer *eb, struct block_device *bdev = fs_info->fs_devices->latest_bdev; u64 offset = eb->start; unsigned long i, num_pages; + unsigned long bio_flags = 0; int rw = (epd->sync_io ? WRITE_SYNC : WRITE); int ret = 0; clear_bit(EXTENT_BUFFER_IOERR, &eb->bflags); num_pages = num_extent_pages(eb->start, eb->len); atomic_set(&eb->io_pages, num_pages); + if (btrfs_header_owner(eb) == BTRFS_TREE_LOG_OBJECTID) + bio_flags = EXTENT_BIO_TREE_LOG; + for (i = 0; i < num_pages; i++) { struct page *p = extent_buffer_page(eb, i); @@ -3175,7 +3230,8 @@ static int write_one_eb(struct extent_buffer *eb, ret = submit_extent_page(rw, eb->tree, p, offset >> 9, PAGE_CACHE_SIZE, 0, bdev, &epd->bio, -1, end_bio_extent_buffer_writepage, - 0, 0, 0); + 0, epd->bio_flags, bio_flags); + epd->bio_flags = bio_flags; if (ret) { set_bit(EXTENT_BUFFER_IOERR, &eb->bflags); SetPageError(p); @@ -3210,6 +3266,7 @@ int btree_write_cache_pages(struct address_space *mapping, .tree = tree, .extent_locked = 0, .sync_io = wbc->sync_mode == WB_SYNC_ALL, + .bio_flags = 0, }; int ret = 0; int done = 0; @@ -3254,19 +3311,34 @@ retry: break; } + spin_lock(&mapping->private_lock); + if (!PagePrivate(page)) { + spin_unlock(&mapping->private_lock); + continue; + } + eb = (struct extent_buffer *)page->private; + + /* + * Shouldn't happen and normally this would be a BUG_ON + * but no sense in crashing the users box for something + * we can survive anyway. + */ if (!eb) { + spin_unlock(&mapping->private_lock); WARN_ON(1); continue; } - if (eb == prev_eb) + if (eb == prev_eb) { + spin_unlock(&mapping->private_lock); continue; + } - if (!atomic_inc_not_zero(&eb->refs)) { - WARN_ON(1); + ret = atomic_inc_not_zero(&eb->refs); + spin_unlock(&mapping->private_lock); + if (!ret) continue; - } prev_eb = eb; ret = lock_extent_buffer_for_io(eb, fs_info, &epd); @@ -3457,7 +3529,7 @@ static void flush_epd_write_bio(struct extent_page_data *epd) if (epd->sync_io) rw = WRITE_SYNC; - ret = submit_one_bio(rw, epd->bio, 0, 0); + ret = submit_one_bio(rw, epd->bio, 0, epd->bio_flags); BUG_ON(ret < 0); /* -ENOMEM */ epd->bio = NULL; } @@ -3480,6 +3552,7 @@ int extent_write_full_page(struct extent_io_tree *tree, struct page *page, .get_extent = get_extent, .extent_locked = 0, .sync_io = wbc->sync_mode == WB_SYNC_ALL, + .bio_flags = 0, }; ret = __extent_writepage(page, wbc, &epd); @@ -3504,6 +3577,7 @@ int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode, .get_extent = get_extent, .extent_locked = 1, .sync_io = mode == WB_SYNC_ALL, + .bio_flags = 0, }; struct writeback_control wbc_writepages = { .sync_mode = mode, @@ -3543,6 +3617,7 @@ int extent_writepages(struct extent_io_tree *tree, .get_extent = get_extent, .extent_locked = 0, .sync_io = wbc->sync_mode == WB_SYNC_ALL, + .bio_flags = 0, }; ret = extent_write_cache_pages(tree, mapping, wbc, @@ -3920,18 +3995,6 @@ out: return ret; } -inline struct page *extent_buffer_page(struct extent_buffer *eb, - unsigned long i) -{ - return eb->pages[i]; -} - -inline unsigned long num_extent_pages(u64 start, u64 len) -{ - return ((start + len + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT) - - (start >> PAGE_CACHE_SHIFT); -} - static void __free_extent_buffer(struct extent_buffer *eb) { #if LEAK_DEBUG @@ -4047,8 +4110,8 @@ struct extent_buffer *alloc_dummy_extent_buffer(u64 start, unsigned long len) return eb; err: - for (i--; i > 0; i--) - __free_page(eb->pages[i]); + for (; i > 0; i--) + __free_page(eb->pages[i - 1]); __free_extent_buffer(eb); return NULL; } @@ -4192,10 +4255,8 @@ struct extent_buffer *alloc_extent_buffer(struct extent_io_tree *tree, for (i = 0; i < num_pages; i++, index++) { p = find_or_create_page(mapping, index, GFP_NOFS); - if (!p) { - WARN_ON(1); + if (!p) goto free_eb; - } spin_lock(&mapping->private_lock); if (PagePrivate(p)) { @@ -4338,7 +4399,6 @@ static int release_extent_buffer(struct extent_buffer *eb, gfp_t mask) /* Should be safe to release our pages at this point */ btrfs_release_extent_buffer_page(eb, 0); - call_rcu(&eb->rcu_head, btrfs_release_extent_buffer_rcu); return 1; } diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index 25900af5b15d..711d12b80028 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -27,6 +27,7 @@ * type for this bio */ #define EXTENT_BIO_COMPRESSED 1 +#define EXTENT_BIO_TREE_LOG 2 #define EXTENT_BIO_FLAG_SHIFT 16 /* these are bit numbers for test/set bit */ @@ -232,11 +233,15 @@ int set_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end, int clear_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end, gfp_t mask); int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, - int bits, int clear_bits, gfp_t mask); + int bits, int clear_bits, + struct extent_state **cached_state, gfp_t mask); int set_extent_delalloc(struct extent_io_tree *tree, u64 start, u64 end, struct extent_state **cached_state, gfp_t mask); +int set_extent_defrag(struct extent_io_tree *tree, u64 start, u64 end, + struct extent_state **cached_state, gfp_t mask); int find_first_extent_bit(struct extent_io_tree *tree, u64 start, - u64 *start_ret, u64 *end_ret, int bits); + u64 *start_ret, u64 *end_ret, int bits, + struct extent_state **cached_state); struct extent_state *find_first_extent_bit_state(struct extent_io_tree *tree, u64 start, int bits); int extent_invalidatepage(struct extent_io_tree *tree, @@ -277,8 +282,18 @@ void free_extent_buffer_stale(struct extent_buffer *eb); int read_extent_buffer_pages(struct extent_io_tree *tree, struct extent_buffer *eb, u64 start, int wait, get_extent_t *get_extent, int mirror_num); -unsigned long num_extent_pages(u64 start, u64 len); -struct page *extent_buffer_page(struct extent_buffer *eb, unsigned long i); + +static inline unsigned long num_extent_pages(u64 start, u64 len) +{ + return ((start + len + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT) - + (start >> PAGE_CACHE_SHIFT); +} + +static inline struct page *extent_buffer_page(struct extent_buffer *eb, + unsigned long i) +{ + return eb->pages[i]; +} static inline void extent_buffer_get(struct extent_buffer *eb) { diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 7c97b3301459..b8cbc8d5c7f7 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -11,7 +11,7 @@ static struct kmem_cache *extent_map_cache; int __init extent_map_init(void) { - extent_map_cache = kmem_cache_create("extent_map", + extent_map_cache = kmem_cache_create("btrfs_extent_map", sizeof(struct extent_map), 0, SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!extent_map_cache) @@ -35,6 +35,7 @@ void extent_map_exit(void) void extent_map_tree_init(struct extent_map_tree *tree) { tree->map = RB_ROOT; + INIT_LIST_HEAD(&tree->modified_extents); rwlock_init(&tree->lock); } @@ -54,7 +55,9 @@ struct extent_map *alloc_extent_map(void) em->in_tree = 0; em->flags = 0; em->compress_type = BTRFS_COMPRESS_NONE; + em->generation = 0; atomic_set(&em->refs, 1); + INIT_LIST_HEAD(&em->list); return em; } @@ -72,6 +75,7 @@ void free_extent_map(struct extent_map *em) WARN_ON(atomic_read(&em->refs) == 0); if (atomic_dec_and_test(&em->refs)) { WARN_ON(em->in_tree); + WARN_ON(!list_empty(&em->list)); kmem_cache_free(extent_map_cache, em); } } @@ -198,6 +202,14 @@ static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em) em->block_len += merge->block_len; em->block_start = merge->block_start; merge->in_tree = 0; + if (merge->generation > em->generation) { + em->mod_start = em->start; + em->mod_len = em->len; + em->generation = merge->generation; + list_move(&em->list, &tree->modified_extents); + } + + list_del_init(&merge->list); rb_erase(&merge->rb_node, &tree->map); free_extent_map(merge); } @@ -211,14 +223,34 @@ static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em) em->block_len += merge->len; rb_erase(&merge->rb_node, &tree->map); merge->in_tree = 0; + if (merge->generation > em->generation) { + em->mod_len = em->len; + em->generation = merge->generation; + list_move(&em->list, &tree->modified_extents); + } + list_del_init(&merge->list); free_extent_map(merge); } } -int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len) +/** + * unpint_extent_cache - unpin an extent from the cache + * @tree: tree to unpin the extent in + * @start: logical offset in the file + * @len: length of the extent + * @gen: generation that this extent has been modified in + * @prealloc: if this is set we need to clear the prealloc flag + * + * Called after an extent has been written to disk properly. Set the generation + * to the generation that actually added the file item to the inode so we know + * we need to sync this extent when we call fsync(). + */ +int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len, + u64 gen) { int ret = 0; struct extent_map *em; + bool prealloc = false; write_lock(&tree->lock); em = lookup_extent_mapping(tree, start, len); @@ -228,10 +260,24 @@ int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len) if (!em) goto out; + list_move(&em->list, &tree->modified_extents); + em->generation = gen; clear_bit(EXTENT_FLAG_PINNED, &em->flags); + em->mod_start = em->start; + em->mod_len = em->len; + + if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) { + prealloc = true; + clear_bit(EXTENT_FLAG_PREALLOC, &em->flags); + } try_merge_map(tree, em); + if (prealloc) { + em->mod_start = em->start; + em->mod_len = em->len; + } + free_extent_map(em); out: write_unlock(&tree->lock); @@ -269,6 +315,9 @@ int add_extent_mapping(struct extent_map_tree *tree, } atomic_inc(&em->refs); + em->mod_start = em->start; + em->mod_len = em->len; + try_merge_map(tree, em); out: return ret; @@ -358,6 +407,8 @@ int remove_extent_mapping(struct extent_map_tree *tree, struct extent_map *em) WARN_ON(test_bit(EXTENT_FLAG_PINNED, &em->flags)); rb_erase(&em->rb_node, &tree->map); + if (!test_bit(EXTENT_FLAG_LOGGING, &em->flags)) + list_del_init(&em->list); em->in_tree = 0; return ret; } diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h index 1195f09761fe..679225555f7b 100644 --- a/fs/btrfs/extent_map.h +++ b/fs/btrfs/extent_map.h @@ -13,6 +13,7 @@ #define EXTENT_FLAG_COMPRESSED 1 #define EXTENT_FLAG_VACANCY 2 /* no file extent item found */ #define EXTENT_FLAG_PREALLOC 3 /* pre-allocated extent */ +#define EXTENT_FLAG_LOGGING 4 /* Logging this extent */ struct extent_map { struct rb_node rb_node; @@ -20,18 +21,23 @@ struct extent_map { /* all of these are in bytes */ u64 start; u64 len; + u64 mod_start; + u64 mod_len; u64 orig_start; u64 block_start; u64 block_len; + u64 generation; unsigned long flags; struct block_device *bdev; atomic_t refs; unsigned int in_tree; unsigned int compress_type; + struct list_head list; }; struct extent_map_tree { struct rb_root map; + struct list_head modified_extents; rwlock_t lock; }; @@ -60,7 +66,7 @@ struct extent_map *alloc_extent_map(void); void free_extent_map(struct extent_map *em); int __init extent_map_init(void); void extent_map_exit(void); -int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len); +int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len, u64 gen); struct extent_map *search_extent_mapping(struct extent_map_tree *tree, u64 start, u64 len); #endif diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index 857d93cd01dc..1ad08e4e4a15 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -25,11 +25,12 @@ #include "transaction.h" #include "print-tree.h" -#define __MAX_CSUM_ITEMS(r, size) ((((BTRFS_LEAF_DATA_SIZE(r) - \ +#define __MAX_CSUM_ITEMS(r, size) ((unsigned long)(((BTRFS_LEAF_DATA_SIZE(r) - \ sizeof(struct btrfs_item) * 2) / \ size) - 1)) -#define MAX_CSUM_ITEMS(r, size) (min(__MAX_CSUM_ITEMS(r, size), PAGE_CACHE_SIZE)) +#define MAX_CSUM_ITEMS(r, size) (min_t(u32, __MAX_CSUM_ITEMS(r, size), \ + PAGE_CACHE_SIZE)) #define MAX_ORDERED_SUM_BYTES(r) ((PAGE_SIZE - \ sizeof(struct btrfs_ordered_sum)) / \ diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 5caf285c6e4d..9ab1bed88116 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -39,6 +39,7 @@ #include "tree-log.h" #include "locking.h" #include "compat.h" +#include "volumes.h" /* * when auto defrag is enabled we @@ -458,14 +459,15 @@ int btrfs_dirty_pages(struct btrfs_root *root, struct inode *inode, * this drops all the extents in the cache that intersect the range * [start, end]. Existing extents are split as required. */ -int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, - int skip_pinned) +void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, + int skip_pinned) { struct extent_map *em; struct extent_map *split = NULL; struct extent_map *split2 = NULL; struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; u64 len = end - start + 1; + u64 gen; int ret; int testend = 1; unsigned long flags; @@ -477,11 +479,14 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, testend = 0; } while (1) { + int no_splits = 0; + if (!split) split = alloc_extent_map(); if (!split2) split2 = alloc_extent_map(); - BUG_ON(!split || !split2); /* -ENOMEM */ + if (!split || !split2) + no_splits = 1; write_lock(&em_tree->lock); em = lookup_extent_mapping(em_tree, start, len); @@ -490,6 +495,7 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, break; } flags = em->flags; + gen = em->generation; if (skip_pinned && test_bit(EXTENT_FLAG_PINNED, &em->flags)) { if (testend && em->start + em->len >= start + len) { free_extent_map(em); @@ -506,6 +512,8 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, compressed = test_bit(EXTENT_FLAG_COMPRESSED, &em->flags); clear_bit(EXTENT_FLAG_PINNED, &em->flags); remove_extent_mapping(em_tree, em); + if (no_splits) + goto next; if (em->block_start < EXTENT_MAP_LAST_BYTE && em->start < start) { @@ -518,12 +526,13 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, split->block_len = em->block_len; else split->block_len = split->len; - + split->generation = gen; split->bdev = em->bdev; split->flags = flags; split->compress_type = em->compress_type; ret = add_extent_mapping(em_tree, split); BUG_ON(ret); /* Logic error */ + list_move(&split->list, &em_tree->modified_extents); free_extent_map(split); split = split2; split2 = NULL; @@ -537,6 +546,7 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, split->bdev = em->bdev; split->flags = flags; split->compress_type = em->compress_type; + split->generation = gen; if (compressed) { split->block_len = em->block_len; @@ -550,9 +560,11 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, ret = add_extent_mapping(em_tree, split); BUG_ON(ret); /* Logic error */ + list_move(&split->list, &em_tree->modified_extents); free_extent_map(split); split = NULL; } +next: write_unlock(&em_tree->lock); /* once for us */ @@ -564,7 +576,6 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, free_extent_map(split); if (split2) free_extent_map(split2); - return 0; } /* @@ -576,13 +587,13 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, * it is either truncated or split. Anything entirely inside the range * is deleted from the tree. */ -int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode, - u64 start, u64 end, u64 *hint_byte, int drop_cache) +int __btrfs_drop_extents(struct btrfs_trans_handle *trans, + struct btrfs_root *root, struct inode *inode, + struct btrfs_path *path, u64 start, u64 end, + u64 *drop_end, int drop_cache) { - struct btrfs_root *root = BTRFS_I(inode)->root; struct extent_buffer *leaf; struct btrfs_file_extent_item *fi; - struct btrfs_path *path; struct btrfs_key key; struct btrfs_key new_key; u64 ino = btrfs_ino(inode); @@ -597,14 +608,12 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode, int recow; int ret; int modify_tree = -1; + int update_refs = (root->ref_cows || root == root->fs_info->tree_root); + int found = 0; if (drop_cache) btrfs_drop_extent_cache(inode, start, end - 1, 0); - path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; - if (start >= BTRFS_I(inode)->disk_i_size) modify_tree = 0; @@ -666,6 +675,7 @@ next_slot: goto next_slot; } + found = 1; search_start = max(key.offset, start); if (recow || !modify_tree) { modify_tree = -1; @@ -707,14 +717,13 @@ next_slot: extent_end - start); btrfs_mark_buffer_dirty(leaf); - if (disk_bytenr > 0) { + if (update_refs && disk_bytenr > 0) { ret = btrfs_inc_extent_ref(trans, root, disk_bytenr, num_bytes, 0, root->root_key.objectid, new_key.objectid, start - extent_offset, 0); BUG_ON(ret); /* -ENOMEM */ - *hint_byte = disk_bytenr; } key.offset = start; } @@ -734,10 +743,8 @@ next_slot: btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - end); btrfs_mark_buffer_dirty(leaf); - if (disk_bytenr > 0) { + if (update_refs && disk_bytenr > 0) inode_sub_bytes(inode, end - key.offset); - *hint_byte = disk_bytenr; - } break; } @@ -753,10 +760,8 @@ next_slot: btrfs_set_file_extent_num_bytes(leaf, fi, start - key.offset); btrfs_mark_buffer_dirty(leaf); - if (disk_bytenr > 0) { + if (update_refs && disk_bytenr > 0) inode_sub_bytes(inode, extent_end - start); - *hint_byte = disk_bytenr; - } if (end == extent_end) break; @@ -777,12 +782,13 @@ next_slot: del_nr++; } - if (extent_type == BTRFS_FILE_EXTENT_INLINE) { + if (update_refs && + extent_type == BTRFS_FILE_EXTENT_INLINE) { inode_sub_bytes(inode, extent_end - key.offset); extent_end = ALIGN(extent_end, root->sectorsize); - } else if (disk_bytenr > 0) { + } else if (update_refs && disk_bytenr > 0) { ret = btrfs_free_extent(trans, root, disk_bytenr, num_bytes, 0, root->root_key.objectid, @@ -791,7 +797,6 @@ next_slot: BUG_ON(ret); /* -ENOMEM */ inode_sub_bytes(inode, extent_end - key.offset); - *hint_byte = disk_bytenr; } if (end == extent_end) @@ -806,7 +811,7 @@ next_slot: del_nr); if (ret) { btrfs_abort_transaction(trans, root, ret); - goto out; + break; } del_nr = 0; @@ -825,7 +830,24 @@ next_slot: btrfs_abort_transaction(trans, root, ret); } -out: + if (drop_end) + *drop_end = found ? min(end, extent_end) : end; + btrfs_release_path(path); + return ret; +} + +int btrfs_drop_extents(struct btrfs_trans_handle *trans, + struct btrfs_root *root, struct inode *inode, u64 start, + u64 end, int drop_cache) +{ + struct btrfs_path *path; + int ret; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + ret = __btrfs_drop_extents(trans, root, inode, path, start, end, NULL, + drop_cache); btrfs_free_path(path); return ret; } @@ -892,8 +914,6 @@ int btrfs_mark_extent_written(struct btrfs_trans_handle *trans, int ret; u64 ino = btrfs_ino(inode); - btrfs_drop_extent_cache(inode, start, end - 1, 0); - path = btrfs_alloc_path(); if (!path) return -ENOMEM; @@ -935,12 +955,16 @@ again: btrfs_set_item_key_safe(trans, root, path, &new_key); fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); + btrfs_set_file_extent_generation(leaf, fi, + trans->transid); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - end); btrfs_set_file_extent_offset(leaf, fi, end - orig_offset); fi = btrfs_item_ptr(leaf, path->slots[0] - 1, struct btrfs_file_extent_item); + btrfs_set_file_extent_generation(leaf, fi, + trans->transid); btrfs_set_file_extent_num_bytes(leaf, fi, end - other_start); btrfs_mark_buffer_dirty(leaf); @@ -958,12 +982,16 @@ again: struct btrfs_file_extent_item); btrfs_set_file_extent_num_bytes(leaf, fi, start - key.offset); + btrfs_set_file_extent_generation(leaf, fi, + trans->transid); path->slots[0]++; new_key.offset = start; btrfs_set_item_key_safe(trans, root, path, &new_key); fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); + btrfs_set_file_extent_generation(leaf, fi, + trans->transid); btrfs_set_file_extent_num_bytes(leaf, fi, other_end - start); btrfs_set_file_extent_offset(leaf, fi, @@ -991,12 +1019,14 @@ again: leaf = path->nodes[0]; fi = btrfs_item_ptr(leaf, path->slots[0] - 1, struct btrfs_file_extent_item); + btrfs_set_file_extent_generation(leaf, fi, trans->transid); btrfs_set_file_extent_num_bytes(leaf, fi, split - key.offset); fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); + btrfs_set_file_extent_generation(leaf, fi, trans->transid); btrfs_set_file_extent_offset(leaf, fi, split - orig_offset); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - split); @@ -1056,12 +1086,14 @@ again: struct btrfs_file_extent_item); btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG); + btrfs_set_file_extent_generation(leaf, fi, trans->transid); btrfs_mark_buffer_dirty(leaf); } else { fi = btrfs_item_ptr(leaf, del_slot - 1, struct btrfs_file_extent_item); btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG); + btrfs_set_file_extent_generation(leaf, fi, trans->transid); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - key.offset); btrfs_mark_buffer_dirty(leaf); @@ -1173,8 +1205,8 @@ again: clear_extent_bit(&BTRFS_I(inode)->io_tree, start_pos, last_pos - 1, EXTENT_DIRTY | EXTENT_DELALLOC | - EXTENT_DO_ACCOUNTING, 0, 0, &cached_state, - GFP_NOFS); + EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, + 0, 0, &cached_state, GFP_NOFS); unlock_extent_cached(&BTRFS_I(inode)->io_tree, start_pos, last_pos - 1, &cached_state, GFP_NOFS); @@ -1514,16 +1546,24 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) trace_btrfs_sync_file(file, datasync); + /* + * We write the dirty pages in the range and wait until they complete + * out of the ->i_mutex. If so, we can flush the dirty pages by + * multi-task, and make the performance up. + */ + ret = filemap_write_and_wait_range(inode->i_mapping, start, end); + if (ret) + return ret; + mutex_lock(&inode->i_mutex); /* - * we wait first, since the writeback may change the inode, also wait - * ordered range does a filemape_write_and_wait_range which is why we - * don't do it above like other file systems. + * We flush the dirty pages again to avoid some dirty pages in the + * range being left. */ - root->log_batch++; + atomic_inc(&root->log_batch); btrfs_wait_ordered_range(inode, start, end); - root->log_batch++; + atomic_inc(&root->log_batch); /* * check the transaction that last modified this inode @@ -1544,6 +1584,14 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) BTRFS_I(inode)->last_trans <= root->fs_info->last_trans_committed) { BTRFS_I(inode)->last_trans = 0; + + /* + * We'v had everything committed since the last time we were + * modified so clear this flag in case it was set for whatever + * reason, it's no longer relevant. + */ + clear_bit(BTRFS_INODE_NEEDS_FULL_SYNC, + &BTRFS_I(inode)->runtime_flags); mutex_unlock(&inode->i_mutex); goto out; } @@ -1599,6 +1647,7 @@ out: static const struct vm_operations_struct btrfs_file_vm_ops = { .fault = filemap_fault, .page_mkwrite = btrfs_page_mkwrite, + .remap_pages = generic_file_remap_pages, }; static int btrfs_file_mmap(struct file *filp, struct vm_area_struct *vma) @@ -1610,11 +1659,328 @@ static int btrfs_file_mmap(struct file *filp, struct vm_area_struct *vma) file_accessed(filp); vma->vm_ops = &btrfs_file_vm_ops; - vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } +static int hole_mergeable(struct inode *inode, struct extent_buffer *leaf, + int slot, u64 start, u64 end) +{ + struct btrfs_file_extent_item *fi; + struct btrfs_key key; + + if (slot < 0 || slot >= btrfs_header_nritems(leaf)) + return 0; + + btrfs_item_key_to_cpu(leaf, &key, slot); + if (key.objectid != btrfs_ino(inode) || + key.type != BTRFS_EXTENT_DATA_KEY) + return 0; + + fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item); + + if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_REG) + return 0; + + if (btrfs_file_extent_disk_bytenr(leaf, fi)) + return 0; + + if (key.offset == end) + return 1; + if (key.offset + btrfs_file_extent_num_bytes(leaf, fi) == start) + return 1; + return 0; +} + +static int fill_holes(struct btrfs_trans_handle *trans, struct inode *inode, + struct btrfs_path *path, u64 offset, u64 end) +{ + struct btrfs_root *root = BTRFS_I(inode)->root; + struct extent_buffer *leaf; + struct btrfs_file_extent_item *fi; + struct extent_map *hole_em; + struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; + struct btrfs_key key; + int ret; + + key.objectid = btrfs_ino(inode); + key.type = BTRFS_EXTENT_DATA_KEY; + key.offset = offset; + + + ret = btrfs_search_slot(trans, root, &key, path, 0, 1); + if (ret < 0) + return ret; + BUG_ON(!ret); + + leaf = path->nodes[0]; + if (hole_mergeable(inode, leaf, path->slots[0]-1, offset, end)) { + u64 num_bytes; + + path->slots[0]--; + fi = btrfs_item_ptr(leaf, path->slots[0], + struct btrfs_file_extent_item); + num_bytes = btrfs_file_extent_num_bytes(leaf, fi) + + end - offset; + btrfs_set_file_extent_num_bytes(leaf, fi, num_bytes); + btrfs_set_file_extent_ram_bytes(leaf, fi, num_bytes); + btrfs_set_file_extent_offset(leaf, fi, 0); + btrfs_mark_buffer_dirty(leaf); + goto out; + } + + if (hole_mergeable(inode, leaf, path->slots[0]+1, offset, end)) { + u64 num_bytes; + + path->slots[0]++; + key.offset = offset; + btrfs_set_item_key_safe(trans, root, path, &key); + fi = btrfs_item_ptr(leaf, path->slots[0], + struct btrfs_file_extent_item); + num_bytes = btrfs_file_extent_num_bytes(leaf, fi) + end - + offset; + btrfs_set_file_extent_num_bytes(leaf, fi, num_bytes); + btrfs_set_file_extent_ram_bytes(leaf, fi, num_bytes); + btrfs_set_file_extent_offset(leaf, fi, 0); + btrfs_mark_buffer_dirty(leaf); + goto out; + } + btrfs_release_path(path); + + ret = btrfs_insert_file_extent(trans, root, btrfs_ino(inode), offset, + 0, 0, end - offset, 0, end - offset, + 0, 0, 0); + if (ret) + return ret; + +out: + btrfs_release_path(path); + + hole_em = alloc_extent_map(); + if (!hole_em) { + btrfs_drop_extent_cache(inode, offset, end - 1, 0); + set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, + &BTRFS_I(inode)->runtime_flags); + } else { + hole_em->start = offset; + hole_em->len = end - offset; + hole_em->orig_start = offset; + + hole_em->block_start = EXTENT_MAP_HOLE; + hole_em->block_len = 0; + hole_em->bdev = root->fs_info->fs_devices->latest_bdev; + hole_em->compress_type = BTRFS_COMPRESS_NONE; + hole_em->generation = trans->transid; + + do { + btrfs_drop_extent_cache(inode, offset, end - 1, 0); + write_lock(&em_tree->lock); + ret = add_extent_mapping(em_tree, hole_em); + if (!ret) + list_move(&hole_em->list, + &em_tree->modified_extents); + write_unlock(&em_tree->lock); + } while (ret == -EEXIST); + free_extent_map(hole_em); + if (ret) + set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, + &BTRFS_I(inode)->runtime_flags); + } + + return 0; +} + +static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) +{ + struct btrfs_root *root = BTRFS_I(inode)->root; + struct extent_state *cached_state = NULL; + struct btrfs_path *path; + struct btrfs_block_rsv *rsv; + struct btrfs_trans_handle *trans; + u64 mask = BTRFS_I(inode)->root->sectorsize - 1; + u64 lockstart = (offset + mask) & ~mask; + u64 lockend = ((offset + len) & ~mask) - 1; + u64 cur_offset = lockstart; + u64 min_size = btrfs_calc_trunc_metadata_size(root, 1); + u64 drop_end; + unsigned long nr; + int ret = 0; + int err = 0; + bool same_page = (offset >> PAGE_CACHE_SHIFT) == + ((offset + len) >> PAGE_CACHE_SHIFT); + + btrfs_wait_ordered_range(inode, offset, len); + + mutex_lock(&inode->i_mutex); + if (offset >= inode->i_size) { + mutex_unlock(&inode->i_mutex); + return 0; + } + + /* + * Only do this if we are in the same page and we aren't doing the + * entire page. + */ + if (same_page && len < PAGE_CACHE_SIZE) { + ret = btrfs_truncate_page(inode, offset, len, 0); + mutex_unlock(&inode->i_mutex); + return ret; + } + + /* zero back part of the first page */ + ret = btrfs_truncate_page(inode, offset, 0, 0); + if (ret) { + mutex_unlock(&inode->i_mutex); + return ret; + } + + /* zero the front end of the last page */ + ret = btrfs_truncate_page(inode, offset + len, 0, 1); + if (ret) { + mutex_unlock(&inode->i_mutex); + return ret; + } + + if (lockend < lockstart) { + mutex_unlock(&inode->i_mutex); + return 0; + } + + while (1) { + struct btrfs_ordered_extent *ordered; + + truncate_pagecache_range(inode, lockstart, lockend); + + lock_extent_bits(&BTRFS_I(inode)->io_tree, lockstart, lockend, + 0, &cached_state); + ordered = btrfs_lookup_first_ordered_extent(inode, lockend); + + /* + * We need to make sure we have no ordered extents in this range + * and nobody raced in and read a page in this range, if we did + * we need to try again. + */ + if ((!ordered || + (ordered->file_offset + ordered->len < lockstart || + ordered->file_offset > lockend)) && + !test_range_bit(&BTRFS_I(inode)->io_tree, lockstart, + lockend, EXTENT_UPTODATE, 0, + cached_state)) { + if (ordered) + btrfs_put_ordered_extent(ordered); + break; + } + if (ordered) + btrfs_put_ordered_extent(ordered); + unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, + lockend, &cached_state, GFP_NOFS); + btrfs_wait_ordered_range(inode, lockstart, + lockend - lockstart + 1); + } + + path = btrfs_alloc_path(); + if (!path) { + ret = -ENOMEM; + goto out; + } + + rsv = btrfs_alloc_block_rsv(root, BTRFS_BLOCK_RSV_TEMP); + if (!rsv) { + ret = -ENOMEM; + goto out_free; + } + rsv->size = btrfs_calc_trunc_metadata_size(root, 1); + rsv->failfast = 1; + + /* + * 1 - update the inode + * 1 - removing the extents in the range + * 1 - adding the hole extent + */ + trans = btrfs_start_transaction(root, 3); + if (IS_ERR(trans)) { + err = PTR_ERR(trans); + goto out_free; + } + + ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv, rsv, + min_size); + BUG_ON(ret); + trans->block_rsv = rsv; + + while (cur_offset < lockend) { + ret = __btrfs_drop_extents(trans, root, inode, path, + cur_offset, lockend + 1, + &drop_end, 1); + if (ret != -ENOSPC) + break; + + trans->block_rsv = &root->fs_info->trans_block_rsv; + + ret = fill_holes(trans, inode, path, cur_offset, drop_end); + if (ret) { + err = ret; + break; + } + + cur_offset = drop_end; + + ret = btrfs_update_inode(trans, root, inode); + if (ret) { + err = ret; + break; + } + + nr = trans->blocks_used; + btrfs_end_transaction(trans, root); + btrfs_btree_balance_dirty(root, nr); + + trans = btrfs_start_transaction(root, 3); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + trans = NULL; + break; + } + + ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv, + rsv, min_size); + BUG_ON(ret); /* shouldn't happen */ + trans->block_rsv = rsv; + } + + if (ret) { + err = ret; + goto out_trans; + } + + trans->block_rsv = &root->fs_info->trans_block_rsv; + ret = fill_holes(trans, inode, path, cur_offset, drop_end); + if (ret) { + err = ret; + goto out_trans; + } + +out_trans: + if (!trans) + goto out_free; + + trans->block_rsv = &root->fs_info->trans_block_rsv; + ret = btrfs_update_inode(trans, root, inode); + nr = trans->blocks_used; + btrfs_end_transaction(trans, root); + btrfs_btree_balance_dirty(root, nr); +out_free: + btrfs_free_path(path); + btrfs_free_block_rsv(root, rsv); +out: + unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend, + &cached_state, GFP_NOFS); + mutex_unlock(&inode->i_mutex); + if (ret && !err) + err = ret; + return err; +} + static long btrfs_fallocate(struct file *file, int mode, loff_t offset, loff_t len) { @@ -1633,15 +1999,18 @@ static long btrfs_fallocate(struct file *file, int mode, alloc_start = offset & ~mask; alloc_end = (offset + len + mask) & ~mask; - /* We only support the FALLOC_FL_KEEP_SIZE mode */ - if (mode & ~FALLOC_FL_KEEP_SIZE) + /* Make sure we aren't being give some crap mode */ + if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE)) return -EOPNOTSUPP; + if (mode & FALLOC_FL_PUNCH_HOLE) + return btrfs_punch_hole(inode, offset, len); + /* * Make sure we have enough space before we do the * allocation. */ - ret = btrfs_check_data_free_space(inode, len); + ret = btrfs_check_data_free_space(inode, alloc_end - alloc_start + 1); if (ret) return ret; @@ -1748,7 +2117,7 @@ static long btrfs_fallocate(struct file *file, int mode, out: mutex_unlock(&inode->i_mutex); /* Let go of our reservation. */ - btrfs_free_reserved_data_space(inode, len); + btrfs_free_reserved_data_space(inode, alloc_end - alloc_start + 1); return ret; } diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index 6b10acfc2f5c..1027b854b90c 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -966,7 +966,7 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, block_group->key.offset)) { ret = find_first_extent_bit(unpin, start, &extent_start, &extent_end, - EXTENT_DIRTY); + EXTENT_DIRTY, NULL); if (ret) { ret = 0; break; @@ -1454,9 +1454,7 @@ static int search_bitmap(struct btrfs_free_space_ctl *ctl, max_t(u64, *offset, bitmap_info->offset)); bits = bytes_to_bits(*bytes, ctl->unit); - for (i = find_next_bit(bitmap_info->bitmap, BITS_PER_BITMAP, i); - i < BITS_PER_BITMAP; - i = find_next_bit(bitmap_info->bitmap, BITS_PER_BITMAP, i + 1)) { + for_each_set_bit_from(i, bitmap_info->bitmap, BITS_PER_BITMAP) { next_zero = find_next_zero_bit(bitmap_info->bitmap, BITS_PER_BITMAP, i); if ((next_zero - i) >= bits) { @@ -2307,9 +2305,7 @@ static int btrfs_bitmap_cluster(struct btrfs_block_group_cache *block_group, again: found_bits = 0; - for (i = find_next_bit(entry->bitmap, BITS_PER_BITMAP, i); - i < BITS_PER_BITMAP; - i = find_next_bit(entry->bitmap, BITS_PER_BITMAP, i + 1)) { + for_each_set_bit_from(i, entry->bitmap, BITS_PER_BITMAP) { next_zero = find_next_zero_bit(entry->bitmap, BITS_PER_BITMAP, i); if (next_zero - i >= min_bits) { diff --git a/fs/btrfs/hash.h b/fs/btrfs/hash.h index db2ff9773b99..1d982812ab67 100644 --- a/fs/btrfs/hash.h +++ b/fs/btrfs/hash.h @@ -24,4 +24,14 @@ static inline u64 btrfs_name_hash(const char *name, int len) { return crc32c((u32)~1, name, len); } + +/* + * Figure the key offset of an extended inode ref + */ +static inline u64 btrfs_extref_hash(u64 parent_objectid, const char *name, + int len) +{ + return (u64) crc32c(parent_objectid, name, len); +} + #endif diff --git a/fs/btrfs/inode-item.c b/fs/btrfs/inode-item.c index a13cf1a96c73..48b8fda93132 100644 --- a/fs/btrfs/inode-item.c +++ b/fs/btrfs/inode-item.c @@ -18,6 +18,7 @@ #include "ctree.h" #include "disk-io.h" +#include "hash.h" #include "transaction.h" #include "print-tree.h" @@ -50,18 +51,57 @@ static int find_name_in_backref(struct btrfs_path *path, const char *name, return 0; } -struct btrfs_inode_ref * +int btrfs_find_name_in_ext_backref(struct btrfs_path *path, u64 ref_objectid, + const char *name, int name_len, + struct btrfs_inode_extref **extref_ret) +{ + struct extent_buffer *leaf; + struct btrfs_inode_extref *extref; + unsigned long ptr; + unsigned long name_ptr; + u32 item_size; + u32 cur_offset = 0; + int ref_name_len; + + leaf = path->nodes[0]; + item_size = btrfs_item_size_nr(leaf, path->slots[0]); + ptr = btrfs_item_ptr_offset(leaf, path->slots[0]); + + /* + * Search all extended backrefs in this item. We're only + * looking through any collisions so most of the time this is + * just going to compare against one buffer. If all is well, + * we'll return success and the inode ref object. + */ + while (cur_offset < item_size) { + extref = (struct btrfs_inode_extref *) (ptr + cur_offset); + name_ptr = (unsigned long)(&extref->name); + ref_name_len = btrfs_inode_extref_name_len(leaf, extref); + + if (ref_name_len == name_len && + btrfs_inode_extref_parent(leaf, extref) == ref_objectid && + (memcmp_extent_buffer(leaf, name, name_ptr, name_len) == 0)) { + if (extref_ret) + *extref_ret = extref; + return 1; + } + + cur_offset += ref_name_len + sizeof(*extref); + } + return 0; +} + +static struct btrfs_inode_ref * btrfs_lookup_inode_ref(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - struct btrfs_path *path, - const char *name, int name_len, - u64 inode_objectid, u64 ref_objectid, int mod) + struct btrfs_root *root, + struct btrfs_path *path, + const char *name, int name_len, + u64 inode_objectid, u64 ref_objectid, int ins_len, + int cow) { + int ret; struct btrfs_key key; struct btrfs_inode_ref *ref; - int ins_len = mod < 0 ? -1 : 0; - int cow = mod != 0; - int ret; key.objectid = inode_objectid; key.type = BTRFS_INODE_REF_KEY; @@ -77,13 +117,150 @@ btrfs_lookup_inode_ref(struct btrfs_trans_handle *trans, return ref; } -int btrfs_del_inode_ref(struct btrfs_trans_handle *trans, +/* Returns NULL if no extref found */ +struct btrfs_inode_extref * +btrfs_lookup_inode_extref(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + struct btrfs_path *path, + const char *name, int name_len, + u64 inode_objectid, u64 ref_objectid, int ins_len, + int cow) +{ + int ret; + struct btrfs_key key; + struct btrfs_inode_extref *extref; + + key.objectid = inode_objectid; + key.type = BTRFS_INODE_EXTREF_KEY; + key.offset = btrfs_extref_hash(ref_objectid, name, name_len); + + ret = btrfs_search_slot(trans, root, &key, path, ins_len, cow); + if (ret < 0) + return ERR_PTR(ret); + if (ret > 0) + return NULL; + if (!btrfs_find_name_in_ext_backref(path, ref_objectid, name, name_len, &extref)) + return NULL; + return extref; +} + +int btrfs_get_inode_ref_index(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + struct btrfs_path *path, + const char *name, int name_len, + u64 inode_objectid, u64 ref_objectid, int mod, + u64 *ret_index) +{ + struct btrfs_inode_ref *ref; + struct btrfs_inode_extref *extref; + int ins_len = mod < 0 ? -1 : 0; + int cow = mod != 0; + + ref = btrfs_lookup_inode_ref(trans, root, path, name, name_len, + inode_objectid, ref_objectid, ins_len, + cow); + if (IS_ERR(ref)) + return PTR_ERR(ref); + + if (ref != NULL) { + *ret_index = btrfs_inode_ref_index(path->nodes[0], ref); + return 0; + } + + btrfs_release_path(path); + + extref = btrfs_lookup_inode_extref(trans, root, path, name, + name_len, inode_objectid, + ref_objectid, ins_len, cow); + if (IS_ERR(extref)) + return PTR_ERR(extref); + + if (extref) { + *ret_index = btrfs_inode_extref_index(path->nodes[0], extref); + return 0; + } + + return -ENOENT; +} + +int btrfs_del_inode_extref(struct btrfs_trans_handle *trans, struct btrfs_root *root, const char *name, int name_len, u64 inode_objectid, u64 ref_objectid, u64 *index) { struct btrfs_path *path; struct btrfs_key key; + struct btrfs_inode_extref *extref; + struct extent_buffer *leaf; + int ret; + int del_len = name_len + sizeof(*extref); + unsigned long ptr; + unsigned long item_start; + u32 item_size; + + key.objectid = inode_objectid; + btrfs_set_key_type(&key, BTRFS_INODE_EXTREF_KEY); + key.offset = btrfs_extref_hash(ref_objectid, name, name_len); + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + path->leave_spinning = 1; + + ret = btrfs_search_slot(trans, root, &key, path, -1, 1); + if (ret > 0) + ret = -ENOENT; + if (ret < 0) + goto out; + + /* + * Sanity check - did we find the right item for this name? + * This should always succeed so error here will make the FS + * readonly. + */ + if (!btrfs_find_name_in_ext_backref(path, ref_objectid, + name, name_len, &extref)) { + btrfs_std_error(root->fs_info, -ENOENT); + ret = -EROFS; + goto out; + } + + leaf = path->nodes[0]; + item_size = btrfs_item_size_nr(leaf, path->slots[0]); + if (index) + *index = btrfs_inode_extref_index(leaf, extref); + + if (del_len == item_size) { + /* + * Common case only one ref in the item, remove the + * whole item. + */ + ret = btrfs_del_item(trans, root, path); + goto out; + } + + ptr = (unsigned long)extref; + item_start = btrfs_item_ptr_offset(leaf, path->slots[0]); + + memmove_extent_buffer(leaf, ptr, ptr + del_len, + item_size - (ptr + del_len - item_start)); + + btrfs_truncate_item(trans, root, path, item_size - del_len, 1); + +out: + btrfs_free_path(path); + + return ret; +} + +int btrfs_del_inode_ref(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + const char *name, int name_len, + u64 inode_objectid, u64 ref_objectid, u64 *index) +{ + struct btrfs_path *path; + struct btrfs_key key; struct btrfs_inode_ref *ref; struct extent_buffer *leaf; unsigned long ptr; @@ -91,6 +268,7 @@ int btrfs_del_inode_ref(struct btrfs_trans_handle *trans, u32 item_size; u32 sub_item_len; int ret; + int search_ext_refs = 0; int del_len = name_len + sizeof(*ref); key.objectid = inode_objectid; @@ -106,12 +284,14 @@ int btrfs_del_inode_ref(struct btrfs_trans_handle *trans, ret = btrfs_search_slot(trans, root, &key, path, -1, 1); if (ret > 0) { ret = -ENOENT; + search_ext_refs = 1; goto out; } else if (ret < 0) { goto out; } if (!find_name_in_backref(path, name, name_len, &ref)) { ret = -ENOENT; + search_ext_refs = 1; goto out; } leaf = path->nodes[0]; @@ -129,8 +309,78 @@ int btrfs_del_inode_ref(struct btrfs_trans_handle *trans, item_start = btrfs_item_ptr_offset(leaf, path->slots[0]); memmove_extent_buffer(leaf, ptr, ptr + sub_item_len, item_size - (ptr + sub_item_len - item_start)); - btrfs_truncate_item(trans, root, path, - item_size - sub_item_len, 1); + btrfs_truncate_item(trans, root, path, item_size - sub_item_len, 1); +out: + btrfs_free_path(path); + + if (search_ext_refs) { + /* + * No refs were found, or we could not find the + * name in our ref array. Find and remove the extended + * inode ref then. + */ + return btrfs_del_inode_extref(trans, root, name, name_len, + inode_objectid, ref_objectid, index); + } + + return ret; +} + +/* + * btrfs_insert_inode_extref() - Inserts an extended inode ref into a tree. + * + * The caller must have checked against BTRFS_LINK_MAX already. + */ +static int btrfs_insert_inode_extref(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + const char *name, int name_len, + u64 inode_objectid, u64 ref_objectid, u64 index) +{ + struct btrfs_inode_extref *extref; + int ret; + int ins_len = name_len + sizeof(*extref); + unsigned long ptr; + struct btrfs_path *path; + struct btrfs_key key; + struct extent_buffer *leaf; + struct btrfs_item *item; + + key.objectid = inode_objectid; + key.type = BTRFS_INODE_EXTREF_KEY; + key.offset = btrfs_extref_hash(ref_objectid, name, name_len); + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + path->leave_spinning = 1; + ret = btrfs_insert_empty_item(trans, root, path, &key, + ins_len); + if (ret == -EEXIST) { + if (btrfs_find_name_in_ext_backref(path, ref_objectid, + name, name_len, NULL)) + goto out; + + btrfs_extend_item(trans, root, path, ins_len); + ret = 0; + } + if (ret < 0) + goto out; + + leaf = path->nodes[0]; + item = btrfs_item_nr(leaf, path->slots[0]); + ptr = (unsigned long)btrfs_item_ptr(leaf, path->slots[0], char); + ptr += btrfs_item_size(leaf, item) - ins_len; + extref = (struct btrfs_inode_extref *)ptr; + + btrfs_set_inode_extref_name_len(path->nodes[0], extref, name_len); + btrfs_set_inode_extref_index(path->nodes[0], extref, index); + btrfs_set_inode_extref_parent(path->nodes[0], extref, ref_objectid); + + ptr = (unsigned long)&extref->name; + write_extent_buffer(path->nodes[0], name, ptr, name_len); + btrfs_mark_buffer_dirty(path->nodes[0]); + out: btrfs_free_path(path); return ret; @@ -191,6 +441,19 @@ int btrfs_insert_inode_ref(struct btrfs_trans_handle *trans, out: btrfs_free_path(path); + + if (ret == -EMLINK) { + struct btrfs_super_block *disk_super = root->fs_info->super_copy; + /* We ran out of space in the ref array. Need to + * add an extended ref. */ + if (btrfs_super_incompat_flags(disk_super) + & BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF) + ret = btrfs_insert_inode_extref(trans, root, name, + name_len, + inode_objectid, + ref_objectid, index); + } + return ret; } diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index a6ed6944e50c..95542a1b3dfc 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -94,8 +94,6 @@ static noinline int cow_file_range(struct inode *inode, struct page *locked_page, u64 start, u64 end, int *page_started, unsigned long *nr_written, int unlock); -static noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans, - struct btrfs_root *root, struct inode *inode); static int btrfs_init_inode_security(struct btrfs_trans_handle *trans, struct inode *inode, struct inode *dir, @@ -230,7 +228,6 @@ static noinline int cow_file_range_inline(struct btrfs_trans_handle *trans, u64 inline_len = actual_end - start; u64 aligned_end = (end + root->sectorsize - 1) & ~((u64)root->sectorsize - 1); - u64 hint_byte; u64 data_len = inline_len; int ret; @@ -247,8 +244,7 @@ static noinline int cow_file_range_inline(struct btrfs_trans_handle *trans, return 1; } - ret = btrfs_drop_extents(trans, inode, start, aligned_end, - &hint_byte, 1); + ret = btrfs_drop_extents(trans, root, inode, start, aligned_end, 1); if (ret) return ret; @@ -664,7 +660,7 @@ retry: async_extent->compressed_size, async_extent->compressed_size, 0, alloc_hint, &ins, 1); - if (ret) + if (ret && ret != -ENOSPC) btrfs_abort_transaction(trans, root, ret); btrfs_end_transaction(trans, root); } @@ -1308,6 +1304,7 @@ out_check: em->block_start = disk_bytenr; em->bdev = root->fs_info->fs_devices->latest_bdev; set_bit(EXTENT_FLAG_PINNED, &em->flags); + set_bit(EXTENT_FLAG_PREALLOC, &em->flags); while (1) { write_lock(&em_tree->lock); ret = add_extent_mapping(em_tree, em); @@ -1364,11 +1361,7 @@ out_check: } error: - if (nolock) { - err = btrfs_end_transaction_nolock(trans, root); - } else { - err = btrfs_end_transaction(trans, root); - } + err = btrfs_end_transaction(trans, root); if (!ret) ret = err; @@ -1785,7 +1778,6 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, struct btrfs_path *path; struct extent_buffer *leaf; struct btrfs_key ins; - u64 hint; int ret; path = btrfs_alloc_path(); @@ -1803,8 +1795,8 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, * the caller is expected to unpin it and allow it to be merged * with the others. */ - ret = btrfs_drop_extents(trans, inode, file_pos, file_pos + num_bytes, - &hint, 0); + ret = btrfs_drop_extents(trans, root, inode, file_pos, + file_pos + num_bytes, 0); if (ret) goto out; @@ -1828,10 +1820,8 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, btrfs_set_file_extent_encryption(leaf, fi, encryption); btrfs_set_file_extent_other_encoding(leaf, fi, other_encoding); - btrfs_unlock_up_safe(path, 1); - btrfs_set_lock_blocking(leaf); - btrfs_mark_buffer_dirty(leaf); + btrfs_release_path(path); inode_add_bytes(inode, num_bytes); @@ -1929,11 +1919,10 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) ordered_extent->len, compress_type, 0, 0, BTRFS_FILE_EXTENT_REG); - unpin_extent_cache(&BTRFS_I(inode)->extent_tree, - ordered_extent->file_offset, - ordered_extent->len); } - + unpin_extent_cache(&BTRFS_I(inode)->extent_tree, + ordered_extent->file_offset, ordered_extent->len, + trans->transid); if (ret < 0) { btrfs_abort_transaction(trans, root, ret); goto out_unlock; @@ -1949,6 +1938,8 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) btrfs_abort_transaction(trans, root, ret); goto out_unlock; } + } else { + btrfs_set_inode_last_trans(trans, inode); } ret = 0; out_unlock: @@ -1958,12 +1949,8 @@ out_unlock: out: if (root != root->fs_info->tree_root) btrfs_delalloc_release_metadata(inode, ordered_extent->len); - if (trans) { - if (nolock) - btrfs_end_transaction_nolock(trans, root); - else - btrfs_end_transaction(trans, root); - } + if (trans) + btrfs_end_transaction(trans, root); if (ret) clear_extent_uptodate(io_tree, ordered_extent->file_offset, @@ -2119,7 +2106,6 @@ void btrfs_run_delayed_iputs(struct btrfs_root *root) if (empty) return; - down_read(&root->fs_info->cleanup_work_sem); spin_lock(&fs_info->delayed_iput_lock); list_splice_init(&fs_info->delayed_iputs, &list); spin_unlock(&fs_info->delayed_iput_lock); @@ -2130,7 +2116,6 @@ void btrfs_run_delayed_iputs(struct btrfs_root *root) iput(delayed->inode); kfree(delayed); } - up_read(&root->fs_info->cleanup_work_sem); } enum btrfs_orphan_cleanup_state { @@ -2198,7 +2183,7 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode) int ret; if (!root->orphan_block_rsv) { - block_rsv = btrfs_alloc_block_rsv(root); + block_rsv = btrfs_alloc_block_rsv(root, BTRFS_BLOCK_RSV_TEMP); if (!block_rsv) return -ENOMEM; } @@ -2225,7 +2210,7 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode) insert = 1; #endif insert = 1; - atomic_dec(&root->orphan_inodes); + atomic_inc(&root->orphan_inodes); } if (!test_and_set_bit(BTRFS_INODE_ORPHAN_META_RESERVED, @@ -2590,6 +2575,18 @@ static void btrfs_read_locked_inode(struct inode *inode) inode_set_bytes(inode, btrfs_inode_nbytes(leaf, inode_item)); BTRFS_I(inode)->generation = btrfs_inode_generation(leaf, inode_item); + BTRFS_I(inode)->last_trans = btrfs_inode_transid(leaf, inode_item); + + /* + * If we were modified in the current generation and evicted from memory + * and then re-read we need to do a full sync since we don't have any + * idea about which extents were modified before we were evicted from + * cache. + */ + if (BTRFS_I(inode)->last_trans == root->fs_info->generation) + set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, + &BTRFS_I(inode)->runtime_flags); + inode->i_version = btrfs_inode_sequence(leaf, inode_item); inode->i_generation = BTRFS_I(inode)->generation; inode->i_rdev = 0; @@ -2747,8 +2744,9 @@ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans, return btrfs_update_inode_item(trans, root, inode); } -static noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans, - struct btrfs_root *root, struct inode *inode) +noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + struct inode *inode) { int ret; @@ -2894,7 +2892,6 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir, struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(dir)->root; struct btrfs_path *path; - struct btrfs_inode_ref *ref; struct btrfs_dir_item *di; struct inode *inode = dentry->d_inode; u64 index; @@ -3008,17 +3005,17 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir, } btrfs_release_path(path); - ref = btrfs_lookup_inode_ref(trans, root, path, - dentry->d_name.name, dentry->d_name.len, - ino, dir_ino, 0); - if (IS_ERR(ref)) { - err = PTR_ERR(ref); + ret = btrfs_get_inode_ref_index(trans, root, path, dentry->d_name.name, + dentry->d_name.len, ino, dir_ino, 0, + &index); + if (ret) { + err = ret; goto out; } - BUG_ON(!ref); /* Logic error */ + if (check_path_shared(root, path)) goto out; - index = btrfs_inode_ref_index(path->nodes[0], ref); + btrfs_release_path(path); /* @@ -3061,7 +3058,7 @@ out: static void __unlink_end_trans(struct btrfs_trans_handle *trans, struct btrfs_root *root) { - if (trans->block_rsv == &root->fs_info->global_block_rsv) { + if (trans->block_rsv->type == BTRFS_BLOCK_RSV_GLOBAL) { btrfs_block_rsv_release(root, trans->block_rsv, trans->bytes_reserved); trans->block_rsv = &root->fs_info->trans_block_rsv; @@ -3191,9 +3188,10 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) struct btrfs_trans_handle *trans; unsigned long nr = 0; - if (inode->i_size > BTRFS_EMPTY_DIR_SIZE || - btrfs_ino(inode) == BTRFS_FIRST_FREE_OBJECTID) + if (inode->i_size > BTRFS_EMPTY_DIR_SIZE) return -ENOTEMPTY; + if (btrfs_ino(inode) == BTRFS_FIRST_FREE_OBJECTID) + return -EPERM; trans = __unlink_start_trans(dir, dentry); if (IS_ERR(trans)) @@ -3267,8 +3265,13 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, return -ENOMEM; path->reada = -1; + /* + * We want to drop from the next block forward in case this new size is + * not block aligned since we will be keeping the last block of the + * extent just the way it is. + */ if (root->ref_cows || root == root->fs_info->tree_root) - btrfs_drop_extent_cache(inode, new_size & (~mask), (u64)-1, 0); + btrfs_drop_extent_cache(inode, (new_size + mask) & (~mask), (u64)-1, 0); /* * This function is also used to drop the items in the log tree before @@ -3429,12 +3432,6 @@ delete: if (path->slots[0] == 0 || path->slots[0] != pending_del_slot) { - if (root->ref_cows && - BTRFS_I(inode)->location.objectid != - BTRFS_FREE_INO_OBJECTID) { - err = -EAGAIN; - goto out; - } if (pending_del_nr) { ret = btrfs_del_items(trans, root, path, pending_del_slot, @@ -3465,12 +3462,20 @@ error: } /* - * taken from block_truncate_page, but does cow as it zeros out - * any bytes left in the last page in the file. + * btrfs_truncate_page - read, zero a chunk and write a page + * @inode - inode that we're zeroing + * @from - the offset to start zeroing + * @len - the length to zero, 0 to zero the entire range respective to the + * offset + * @front - zero up to the offset instead of from the offset on + * + * This will find the page for the "from" offset and cow the page and zero the + * part we want to zero. This is used with truncate and hole punching. */ -static int btrfs_truncate_page(struct address_space *mapping, loff_t from) +int btrfs_truncate_page(struct inode *inode, loff_t from, loff_t len, + int front) { - struct inode *inode = mapping->host; + struct address_space *mapping = inode->i_mapping; struct btrfs_root *root = BTRFS_I(inode)->root; struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; struct btrfs_ordered_extent *ordered; @@ -3485,7 +3490,8 @@ static int btrfs_truncate_page(struct address_space *mapping, loff_t from) u64 page_start; u64 page_end; - if ((offset & (blocksize - 1)) == 0) + if ((offset & (blocksize - 1)) == 0 && + (!len || ((len & (blocksize - 1)) == 0))) goto out; ret = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE); if (ret) @@ -3532,7 +3538,8 @@ again: } clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, page_end, - EXTENT_DIRTY | EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING, + EXTENT_DIRTY | EXTENT_DELALLOC | + EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 0, 0, &cached_state, GFP_NOFS); ret = btrfs_set_extent_delalloc(inode, page_start, page_end, @@ -3545,8 +3552,13 @@ again: ret = 0; if (offset != PAGE_CACHE_SIZE) { + if (!len) + len = PAGE_CACHE_SIZE - offset; kaddr = kmap(page); - memset(kaddr + offset, 0, PAGE_CACHE_SIZE - offset); + if (front) + memset(kaddr, 0, offset); + else + memset(kaddr + offset, 0, len); flush_dcache_page(page); kunmap(page); } @@ -3577,6 +3589,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; struct extent_map *em = NULL; struct extent_state *cached_state = NULL; + struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; u64 mask = root->sectorsize - 1; u64 hole_start = (oldsize + mask) & ~mask; u64 block_end = (size + mask) & ~mask; @@ -3613,7 +3626,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) last_byte = min(extent_map_end(em), block_end); last_byte = (last_byte + mask) & ~mask; if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) { - u64 hint_byte = 0; + struct extent_map *hole_em; hole_size = last_byte - cur_offset; trans = btrfs_start_transaction(root, 3); @@ -3622,9 +3635,9 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) break; } - err = btrfs_drop_extents(trans, inode, cur_offset, - cur_offset + hole_size, - &hint_byte, 1); + err = btrfs_drop_extents(trans, root, inode, + cur_offset, + cur_offset + hole_size, 1); if (err) { btrfs_abort_transaction(trans, root, err); btrfs_end_transaction(trans, root); @@ -3641,9 +3654,39 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) break; } - btrfs_drop_extent_cache(inode, hole_start, - last_byte - 1, 0); + btrfs_drop_extent_cache(inode, cur_offset, + cur_offset + hole_size - 1, 0); + hole_em = alloc_extent_map(); + if (!hole_em) { + set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, + &BTRFS_I(inode)->runtime_flags); + goto next; + } + hole_em->start = cur_offset; + hole_em->len = hole_size; + hole_em->orig_start = cur_offset; + + hole_em->block_start = EXTENT_MAP_HOLE; + hole_em->block_len = 0; + hole_em->bdev = root->fs_info->fs_devices->latest_bdev; + hole_em->compress_type = BTRFS_COMPRESS_NONE; + hole_em->generation = trans->transid; + while (1) { + write_lock(&em_tree->lock); + err = add_extent_mapping(em_tree, hole_em); + if (!err) + list_move(&hole_em->list, + &em_tree->modified_extents); + write_unlock(&em_tree->lock); + if (err != -EEXIST) + break; + btrfs_drop_extent_cache(inode, cur_offset, + cur_offset + + hole_size - 1, 0); + } + free_extent_map(hole_em); +next: btrfs_update_inode(trans, root, inode); btrfs_end_transaction(trans, root); } @@ -3768,26 +3811,22 @@ void btrfs_evict_inode(struct inode *inode) goto no_delete; } - rsv = btrfs_alloc_block_rsv(root); + rsv = btrfs_alloc_block_rsv(root, BTRFS_BLOCK_RSV_TEMP); if (!rsv) { btrfs_orphan_del(NULL, inode); goto no_delete; } rsv->size = min_size; + rsv->failfast = 1; global_rsv = &root->fs_info->global_block_rsv; btrfs_i_size_write(inode, 0); /* - * This is a bit simpler than btrfs_truncate since - * - * 1) We've already reserved our space for our orphan item in the - * unlink. - * 2) We're going to delete the inode item, so we don't need to update - * it at all. - * - * So we just need to reserve some slack space in case we add bytes when - * doing the truncate. + * This is a bit simpler than btrfs_truncate since we've already + * reserved our space for our orphan item in the unlink, so we just + * need to reserve some slack space in case we add bytes and update + * inode item when doing the truncate. */ while (1) { ret = btrfs_block_rsv_refill_noflush(root, rsv, min_size); @@ -3808,7 +3847,7 @@ void btrfs_evict_inode(struct inode *inode) goto no_delete; } - trans = btrfs_start_transaction(root, 0); + trans = btrfs_start_transaction_noflush(root, 1); if (IS_ERR(trans)) { btrfs_orphan_del(NULL, inode); btrfs_free_block_rsv(root, rsv); @@ -3818,9 +3857,13 @@ void btrfs_evict_inode(struct inode *inode) trans->block_rsv = rsv; ret = btrfs_truncate_inode_items(trans, root, inode, 0, 0); - if (ret != -EAGAIN) + if (ret != -ENOSPC) break; + trans->block_rsv = &root->fs_info->trans_block_rsv; + ret = btrfs_update_inode(trans, root, inode); + BUG_ON(ret); + nr = trans->blocks_used; btrfs_end_transaction(trans, root); trans = NULL; @@ -4470,10 +4513,7 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc) trans = btrfs_join_transaction(root); if (IS_ERR(trans)) return PTR_ERR(trans); - if (nolock) - ret = btrfs_end_transaction_nolock(trans, root); - else - ret = btrfs_commit_transaction(trans, root); + ret = btrfs_commit_transaction(trans, root); } return ret; } @@ -4671,6 +4711,14 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, BTRFS_I(inode)->generation = trans->transid; inode->i_generation = BTRFS_I(inode)->generation; + /* + * We could have gotten an inode number from somebody who was fsynced + * and then removed in this same transaction, so let's just set full + * sync since it will be a full sync anyway and this will blow away the + * old info in the log. + */ + set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(inode)->runtime_flags); + if (S_ISDIR(mode)) owner = 0; else @@ -4680,6 +4728,12 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, btrfs_set_key_type(&key[0], BTRFS_INODE_ITEM_KEY); key[0].offset = 0; + /* + * Start new inodes with an inode_ref. This is slightly more + * efficient for small numbers of hard links since they will + * be packed into one item. Extended refs will kick in if we + * add more hard links than can fit in the ref item. + */ key[1].objectid = objectid; btrfs_set_key_type(&key[1], BTRFS_INODE_REF_KEY); key[1].offset = ref_objectid; @@ -4986,7 +5040,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, if (root->objectid != BTRFS_I(inode)->root->objectid) return -EXDEV; - if (inode->i_nlink == ~0U) + if (inode->i_nlink >= BTRFS_LINK_MAX) return -EMLINK; err = btrfs_set_inode_index(dir, &index); @@ -5450,7 +5504,8 @@ insert: write_unlock(&em_tree->lock); out: - trace_btrfs_get_extent(root, em); + if (em) + trace_btrfs_get_extent(root, em); if (path) btrfs_free_path(path); @@ -5836,6 +5891,48 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend, return ret; } +static struct extent_map *create_pinned_em(struct inode *inode, u64 start, + u64 len, u64 orig_start, + u64 block_start, u64 block_len, + int type) +{ + struct extent_map_tree *em_tree; + struct extent_map *em; + struct btrfs_root *root = BTRFS_I(inode)->root; + int ret; + + em_tree = &BTRFS_I(inode)->extent_tree; + em = alloc_extent_map(); + if (!em) + return ERR_PTR(-ENOMEM); + + em->start = start; + em->orig_start = orig_start; + em->len = len; + em->block_len = block_len; + em->block_start = block_start; + em->bdev = root->fs_info->fs_devices->latest_bdev; + set_bit(EXTENT_FLAG_PINNED, &em->flags); + if (type == BTRFS_ORDERED_PREALLOC) + set_bit(EXTENT_FLAG_PREALLOC, &em->flags); + + do { + btrfs_drop_extent_cache(inode, em->start, + em->start + em->len - 1, 0); + write_lock(&em_tree->lock); + ret = add_extent_mapping(em_tree, em); + write_unlock(&em_tree->lock); + } while (ret == -EEXIST); + + if (ret) { + free_extent_map(em); + return ERR_PTR(ret); + } + + return em; +} + + static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { @@ -5950,6 +6047,19 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, goto must_cow; if (can_nocow_odirect(trans, inode, start, len) == 1) { + u64 orig_start = em->start; + + if (type == BTRFS_ORDERED_PREALLOC) { + free_extent_map(em); + em = create_pinned_em(inode, start, len, + orig_start, + block_start, len, type); + if (IS_ERR(em)) { + btrfs_end_transaction(trans, root); + goto unlock_err; + } + } + ret = btrfs_add_ordered_extent_dio(inode, start, block_start, len, len, type); btrfs_end_transaction(trans, root); @@ -5999,7 +6109,8 @@ unlock: if (lockstart < lockend) { if (create && len < lockend - lockstart) { clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart, - lockstart + len - 1, unlock_bits, 1, 0, + lockstart + len - 1, + unlock_bits | EXTENT_DEFRAG, 1, 0, &cached_state, GFP_NOFS); /* * Beside unlock, we also need to cleanup reserved space @@ -6007,8 +6118,8 @@ unlock: */ clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart + len, lockend, - unlock_bits | EXTENT_DO_ACCOUNTING, - 1, 0, NULL, GFP_NOFS); + unlock_bits | EXTENT_DO_ACCOUNTING | + EXTENT_DEFRAG, 1, 0, NULL, GFP_NOFS); } else { clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart, lockend, unlock_bits, 1, 0, @@ -6573,8 +6684,8 @@ static void btrfs_invalidatepage(struct page *page, unsigned long offset) */ clear_extent_bit(tree, page_start, page_end, EXTENT_DIRTY | EXTENT_DELALLOC | - EXTENT_LOCKED | EXTENT_DO_ACCOUNTING, 1, 0, - &cached_state, GFP_NOFS); + EXTENT_LOCKED | EXTENT_DO_ACCOUNTING | + EXTENT_DEFRAG, 1, 0, &cached_state, GFP_NOFS); /* * whoever cleared the private bit is responsible * for the finish_ordered_io @@ -6590,7 +6701,8 @@ static void btrfs_invalidatepage(struct page *page, unsigned long offset) } clear_extent_bit(tree, page_start, page_end, EXTENT_LOCKED | EXTENT_DIRTY | EXTENT_DELALLOC | - EXTENT_DO_ACCOUNTING, 1, 1, &cached_state, GFP_NOFS); + EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 1, 1, + &cached_state, GFP_NOFS); __btrfs_releasepage(page, GFP_NOFS); ClearPageChecked(page); @@ -6687,7 +6799,8 @@ again: * prepare_pages in the normal write path. */ clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, page_end, - EXTENT_DIRTY | EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING, + EXTENT_DIRTY | EXTENT_DELALLOC | + EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 0, 0, &cached_state, GFP_NOFS); ret = btrfs_set_extent_delalloc(inode, page_start, page_end, @@ -6718,6 +6831,7 @@ again: BTRFS_I(inode)->last_trans = root->fs_info->generation; BTRFS_I(inode)->last_sub_trans = BTRFS_I(inode)->root->log_transid; + BTRFS_I(inode)->last_log_commit = BTRFS_I(inode)->root->last_log_commit; unlock_extent_cached(io_tree, page_start, page_end, &cached_state, GFP_NOFS); @@ -6745,7 +6859,7 @@ static int btrfs_truncate(struct inode *inode) u64 mask = root->sectorsize - 1; u64 min_size = btrfs_calc_trunc_metadata_size(root, 1); - ret = btrfs_truncate_page(inode->i_mapping, inode->i_size); + ret = btrfs_truncate_page(inode, inode->i_size, 0, 0); if (ret) return ret; @@ -6788,10 +6902,11 @@ static int btrfs_truncate(struct inode *inode) * 3) fs_info->trans_block_rsv - this will have 1 items worth left for * updating the inode. */ - rsv = btrfs_alloc_block_rsv(root); + rsv = btrfs_alloc_block_rsv(root, BTRFS_BLOCK_RSV_TEMP); if (!rsv) return -ENOMEM; rsv->size = min_size; + rsv->failfast = 1; /* * 1 for the truncate slack space @@ -6837,36 +6952,21 @@ static int btrfs_truncate(struct inode *inode) &BTRFS_I(inode)->runtime_flags)) btrfs_add_ordered_operation(trans, root, inode); - while (1) { - ret = btrfs_block_rsv_refill(root, rsv, min_size); - if (ret) { - /* - * This can only happen with the original transaction we - * started above, every other time we shouldn't have a - * transaction started yet. - */ - if (ret == -EAGAIN) - goto end_trans; - err = ret; - break; - } - - if (!trans) { - /* Just need the 1 for updating the inode */ - trans = btrfs_start_transaction(root, 1); - if (IS_ERR(trans)) { - ret = err = PTR_ERR(trans); - trans = NULL; - break; - } - } - - trans->block_rsv = rsv; + /* + * So if we truncate and then write and fsync we normally would just + * write the extents that changed, which is a problem if we need to + * first truncate that entire inode. So set this flag so we write out + * all of the extents in the inode to the sync log so we're completely + * safe. + */ + set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(inode)->runtime_flags); + trans->block_rsv = rsv; + while (1) { ret = btrfs_truncate_inode_items(trans, root, inode, inode->i_size, BTRFS_EXTENT_DATA_KEY); - if (ret != -EAGAIN) { + if (ret != -ENOSPC) { err = ret; break; } @@ -6877,11 +6977,22 @@ static int btrfs_truncate(struct inode *inode) err = ret; break; } -end_trans: + nr = trans->blocks_used; btrfs_end_transaction(trans, root); - trans = NULL; btrfs_btree_balance_dirty(root, nr); + + trans = btrfs_start_transaction(root, 2); + if (IS_ERR(trans)) { + ret = err = PTR_ERR(trans); + trans = NULL; + break; + } + + ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv, + rsv, min_size); + BUG_ON(ret); /* shouldn't happen */ + trans->block_rsv = rsv; } if (ret == 0 && inode->i_nlink > 0) { @@ -6965,6 +7076,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) ei->csum_bytes = 0; ei->index_cnt = (u64)-1; ei->last_unlink_trans = 0; + ei->last_log_commit = 0; spin_lock_init(&ei->lock); ei->outstanding_extents = 0; @@ -7095,31 +7207,31 @@ void btrfs_destroy_cachep(void) int btrfs_init_cachep(void) { - btrfs_inode_cachep = kmem_cache_create("btrfs_inode_cache", + btrfs_inode_cachep = kmem_cache_create("btrfs_inode", sizeof(struct btrfs_inode), 0, SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, init_once); if (!btrfs_inode_cachep) goto fail; - btrfs_trans_handle_cachep = kmem_cache_create("btrfs_trans_handle_cache", + btrfs_trans_handle_cachep = kmem_cache_create("btrfs_trans_handle", sizeof(struct btrfs_trans_handle), 0, SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!btrfs_trans_handle_cachep) goto fail; - btrfs_transaction_cachep = kmem_cache_create("btrfs_transaction_cache", + btrfs_transaction_cachep = kmem_cache_create("btrfs_transaction", sizeof(struct btrfs_transaction), 0, SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!btrfs_transaction_cachep) goto fail; - btrfs_path_cachep = kmem_cache_create("btrfs_path_cache", + btrfs_path_cachep = kmem_cache_create("btrfs_path", sizeof(struct btrfs_path), 0, SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!btrfs_path_cachep) goto fail; - btrfs_free_space_cachep = kmem_cache_create("btrfs_free_space_cache", + btrfs_free_space_cachep = kmem_cache_create("btrfs_free_space", sizeof(struct btrfs_free_space), 0, SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL); if (!btrfs_free_space_cachep) @@ -7513,6 +7625,8 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, loff_t actual_len, u64 *alloc_hint, struct btrfs_trans_handle *trans) { + struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; + struct extent_map *em; struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_key ins; u64 cur_offset = start; @@ -7553,6 +7667,37 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, btrfs_drop_extent_cache(inode, cur_offset, cur_offset + ins.offset -1, 0); + em = alloc_extent_map(); + if (!em) { + set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, + &BTRFS_I(inode)->runtime_flags); + goto next; + } + + em->start = cur_offset; + em->orig_start = cur_offset; + em->len = ins.offset; + em->block_start = ins.objectid; + em->block_len = ins.offset; + em->bdev = root->fs_info->fs_devices->latest_bdev; + set_bit(EXTENT_FLAG_PREALLOC, &em->flags); + em->generation = trans->transid; + + while (1) { + write_lock(&em_tree->lock); + ret = add_extent_mapping(em_tree, em); + if (!ret) + list_move(&em->list, + &em_tree->modified_extents); + write_unlock(&em_tree->lock); + if (ret != -EEXIST) + break; + btrfs_drop_extent_cache(inode, cur_offset, + cur_offset + ins.offset - 1, + 0); + } + free_extent_map(em); +next: num_bytes -= ins.offset; cur_offset += ins.offset; *alloc_hint = ins.objectid + ins.offset; diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 47127c1bd290..8fcf9a59c28d 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -181,6 +181,7 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) int ret; u64 ip_oldflags; unsigned int i_oldflags; + umode_t mode; if (btrfs_root_readonly(root)) return -EROFS; @@ -203,6 +204,7 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) ip_oldflags = ip->flags; i_oldflags = inode->i_flags; + mode = inode->i_mode; flags = btrfs_mask_flags(inode->i_mode, flags); oldflags = btrfs_flags_to_ioctl(ip->flags); @@ -237,10 +239,31 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) ip->flags |= BTRFS_INODE_DIRSYNC; else ip->flags &= ~BTRFS_INODE_DIRSYNC; - if (flags & FS_NOCOW_FL) - ip->flags |= BTRFS_INODE_NODATACOW; - else - ip->flags &= ~BTRFS_INODE_NODATACOW; + if (flags & FS_NOCOW_FL) { + if (S_ISREG(mode)) { + /* + * It's safe to turn csums off here, no extents exist. + * Otherwise we want the flag to reflect the real COW + * status of the file and will not set it. + */ + if (inode->i_size == 0) + ip->flags |= BTRFS_INODE_NODATACOW + | BTRFS_INODE_NODATASUM; + } else { + ip->flags |= BTRFS_INODE_NODATACOW; + } + } else { + /* + * Revert back under same assuptions as above + */ + if (S_ISREG(mode)) { + if (inode->i_size == 0) + ip->flags &= ~(BTRFS_INODE_NODATACOW + | BTRFS_INODE_NODATASUM); + } else { + ip->flags &= ~BTRFS_INODE_NODATACOW; + } + } /* * The COMPRESS flag can only be changed by users, while the NOCOMPRESS @@ -320,7 +343,8 @@ static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg) return -EOPNOTSUPP; if (copy_from_user(&range, arg, sizeof(range))) return -EFAULT; - if (range.start > total_bytes) + if (range.start > total_bytes || + range.len < fs_info->sb->s_blocksize) return -EINVAL; range.len = min(range.len, total_bytes - range.start); @@ -516,7 +540,8 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry, if (!pending_snapshot) return -ENOMEM; - btrfs_init_block_rsv(&pending_snapshot->block_rsv); + btrfs_init_block_rsv(&pending_snapshot->block_rsv, + BTRFS_BLOCK_RSV_TEMP); pending_snapshot->dentry = dentry; pending_snapshot->root = root; pending_snapshot->readonly = readonly; @@ -525,7 +550,7 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry, *inherit = NULL; /* take responsibility to free it */ } - trans = btrfs_start_transaction(root->fs_info->extent_root, 5); + trans = btrfs_start_transaction(root->fs_info->extent_root, 6); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto fail; @@ -546,7 +571,8 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry, ret = btrfs_commit_transaction(trans, root->fs_info->extent_root); } - BUG_ON(ret); + if (ret) + goto fail; ret = pending_snapshot->error; if (ret) @@ -614,7 +640,7 @@ static int btrfs_may_delete(struct inode *dir,struct dentry *victim,int isdir) return -ENOENT; BUG_ON(victim->d_parent->d_inode != dir); - audit_inode_child(victim, dir); + audit_inode_child(dir, victim, AUDIT_TYPE_CHILD_DELETE); error = inode_permission(dir, MAY_WRITE | MAY_EXEC); if (error) @@ -1022,8 +1048,8 @@ again: page_start, page_end - 1, 0, &cached_state); clear_extent_bit(&BTRFS_I(inode)->io_tree, page_start, page_end - 1, EXTENT_DIRTY | EXTENT_DELALLOC | - EXTENT_DO_ACCOUNTING, 0, 0, &cached_state, - GFP_NOFS); + EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 0, 0, + &cached_state, GFP_NOFS); if (i_done != page_cnt) { spin_lock(&BTRFS_I(inode)->lock); @@ -1034,8 +1060,8 @@ again: } - btrfs_set_extent_delalloc(inode, page_start, page_end - 1, - &cached_state); + set_extent_defrag(&BTRFS_I(inode)->io_tree, page_start, page_end - 1, + &cached_state, GFP_NOFS); unlock_extent_cached(&BTRFS_I(inode)->io_tree, page_start, page_end - 1, &cached_state, @@ -2351,7 +2377,6 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, int ret; u64 len = olen; u64 bs = root->fs_info->sb->s_blocksize; - u64 hint_byte; /* * TODO: @@ -2456,13 +2481,13 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, another, and lock file content */ while (1) { struct btrfs_ordered_extent *ordered; - lock_extent(&BTRFS_I(src)->io_tree, off, off+len); - ordered = btrfs_lookup_first_ordered_extent(src, off+len); + lock_extent(&BTRFS_I(src)->io_tree, off, off + len - 1); + ordered = btrfs_lookup_first_ordered_extent(src, off + len - 1); if (!ordered && - !test_range_bit(&BTRFS_I(src)->io_tree, off, off+len, - EXTENT_DELALLOC, 0, NULL)) + !test_range_bit(&BTRFS_I(src)->io_tree, off, off + len - 1, + EXTENT_DELALLOC, 0, NULL)) break; - unlock_extent(&BTRFS_I(src)->io_tree, off, off+len); + unlock_extent(&BTRFS_I(src)->io_tree, off, off + len - 1); if (ordered) btrfs_put_ordered_extent(ordered); btrfs_wait_ordered_range(src, off, len); @@ -2536,7 +2561,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, btrfs_release_path(path); if (key.offset + datal <= off || - key.offset >= off+len) + key.offset >= off + len - 1) goto next; memcpy(&new_key, &key, sizeof(new_key)); @@ -2574,10 +2599,10 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, datal -= off - key.offset; } - ret = btrfs_drop_extents(trans, inode, + ret = btrfs_drop_extents(trans, root, inode, new_key.offset, new_key.offset + datal, - &hint_byte, 1); + 1); if (ret) { btrfs_abort_transaction(trans, root, ret); @@ -2637,8 +2662,8 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, new_key.offset += skip; } - if (key.offset + datal > off+len) - trim = key.offset + datal - (off+len); + if (key.offset + datal > off + len) + trim = key.offset + datal - (off + len); if (comp && (skip || trim)) { ret = -EINVAL; @@ -2648,10 +2673,10 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, size -= skip + trim; datal -= skip + trim; - ret = btrfs_drop_extents(trans, inode, + ret = btrfs_drop_extents(trans, root, inode, new_key.offset, new_key.offset + datal, - &hint_byte, 1); + 1); if (ret) { btrfs_abort_transaction(trans, root, ret); @@ -2715,7 +2740,7 @@ next: ret = 0; out: btrfs_release_path(path); - unlock_extent(&BTRFS_I(src)->io_tree, off, off+len); + unlock_extent(&BTRFS_I(src)->io_tree, off, off + len - 1); out_unlock: mutex_unlock(&src->i_mutex); mutex_unlock(&inode->i_mutex); @@ -2850,8 +2875,8 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp) return 0; } -static void get_block_group_info(struct list_head *groups_list, - struct btrfs_ioctl_space_info *space) +void btrfs_get_block_group_info(struct list_head *groups_list, + struct btrfs_ioctl_space_info *space) { struct btrfs_block_group_cache *block_group; @@ -2959,8 +2984,8 @@ long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) down_read(&info->groups_sem); for (c = 0; c < BTRFS_NR_RAID_TYPES; c++) { if (!list_empty(&info->block_groups[c])) { - get_block_group_info(&info->block_groups[c], - &space); + btrfs_get_block_group_info( + &info->block_groups[c], &space); memcpy(dest, &space, sizeof(space)); dest++; space_args.total_spaces++; @@ -3208,11 +3233,9 @@ static long btrfs_ioctl_logical_to_ino(struct btrfs_root *root, { int ret = 0; int size; - u64 extent_item_pos; struct btrfs_ioctl_logical_ino_args *loi; struct btrfs_data_container *inodes = NULL; struct btrfs_path *path = NULL; - struct btrfs_key key; if (!capable(CAP_SYS_ADMIN)) return -EPERM; @@ -3230,7 +3253,7 @@ static long btrfs_ioctl_logical_to_ino(struct btrfs_root *root, goto out; } - size = min_t(u32, loi->size, 4096); + size = min_t(u32, loi->size, 64 * 1024); inodes = init_data_container(size); if (IS_ERR(inodes)) { ret = PTR_ERR(inodes); @@ -3238,22 +3261,13 @@ static long btrfs_ioctl_logical_to_ino(struct btrfs_root *root, goto out; } - ret = extent_from_logical(root->fs_info, loi->logical, path, &key); - btrfs_release_path(path); - - if (ret & BTRFS_EXTENT_FLAG_TREE_BLOCK) + ret = iterate_inodes_from_logical(loi->logical, root->fs_info, path, + build_ino_list, inodes); + if (ret == -EINVAL) ret = -ENOENT; if (ret < 0) goto out; - extent_item_pos = loi->logical - key.objectid; - ret = iterate_extent_inodes(root->fs_info, key.objectid, - extent_item_pos, 0, build_ino_list, - inodes); - - if (ret < 0) - goto out; - ret = copy_to_user((void *)(unsigned long)loi->inodes, (void *)(unsigned long)inodes, size); if (ret) @@ -3261,7 +3275,7 @@ static long btrfs_ioctl_logical_to_ino(struct btrfs_root *root, out: btrfs_free_path(path); - kfree(inodes); + vfree(inodes); kfree(loi); return ret; diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index 051c7fe551dd..7772f02ba28e 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -25,6 +25,8 @@ #include "btrfs_inode.h" #include "extent_io.h" +static struct kmem_cache *btrfs_ordered_extent_cache; + static u64 entry_end(struct btrfs_ordered_extent *entry) { if (entry->file_offset + entry->len < entry->file_offset) @@ -187,7 +189,7 @@ static int __btrfs_add_ordered_extent(struct inode *inode, u64 file_offset, struct btrfs_ordered_extent *entry; tree = &BTRFS_I(inode)->ordered_tree; - entry = kzalloc(sizeof(*entry), GFP_NOFS); + entry = kmem_cache_zalloc(btrfs_ordered_extent_cache, GFP_NOFS); if (!entry) return -ENOMEM; @@ -421,7 +423,7 @@ void btrfs_put_ordered_extent(struct btrfs_ordered_extent *entry) list_del(&sum->list); kfree(sum); } - kfree(entry); + kmem_cache_free(btrfs_ordered_extent_cache, entry); } } @@ -466,8 +468,7 @@ void btrfs_remove_ordered_extent(struct inode *inode, * wait for all the ordered extents in a root. This is done when balancing * space between drives. */ -void btrfs_wait_ordered_extents(struct btrfs_root *root, - int nocow_only, int delay_iput) +void btrfs_wait_ordered_extents(struct btrfs_root *root, int delay_iput) { struct list_head splice; struct list_head *cur; @@ -482,15 +483,6 @@ void btrfs_wait_ordered_extents(struct btrfs_root *root, cur = splice.next; ordered = list_entry(cur, struct btrfs_ordered_extent, root_extent_list); - if (nocow_only && - !test_bit(BTRFS_ORDERED_NOCOW, &ordered->flags) && - !test_bit(BTRFS_ORDERED_PREALLOC, &ordered->flags)) { - list_move(&ordered->root_extent_list, - &root->fs_info->ordered_extents); - cond_resched_lock(&root->fs_info->ordered_extent_lock); - continue; - } - list_del_init(&ordered->root_extent_list); atomic_inc(&ordered->refs); @@ -775,7 +767,6 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset, struct btrfs_ordered_inode_tree *tree = &BTRFS_I(inode)->ordered_tree; u64 disk_i_size; u64 new_i_size; - u64 i_size_test; u64 i_size = i_size_read(inode); struct rb_node *node; struct rb_node *prev = NULL; @@ -835,55 +826,30 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset, break; if (test->file_offset >= i_size) break; - if (test->file_offset >= disk_i_size) + if (test->file_offset >= disk_i_size) { + /* + * we don't update disk_i_size now, so record this + * undealt i_size. Or we will not know the real + * i_size. + */ + if (test->outstanding_isize < offset) + test->outstanding_isize = offset; + if (ordered && + ordered->outstanding_isize > + test->outstanding_isize) + test->outstanding_isize = + ordered->outstanding_isize; goto out; - } - new_i_size = min_t(u64, offset, i_size); - - /* - * at this point, we know we can safely update i_size to at least - * the offset from this ordered extent. But, we need to - * walk forward and see if ios from higher up in the file have - * finished. - */ - if (ordered) { - node = rb_next(&ordered->rb_node); - } else { - if (prev) - node = rb_next(prev); - else - node = rb_first(&tree->tree); - } - - /* - * We are looking for an area between our current extent and the next - * ordered extent to update the i_size to. There are 3 cases here - * - * 1) We don't actually have anything and we can update to i_size. - * 2) We have stuff but they already did their i_size update so again we - * can just update to i_size. - * 3) We have an outstanding ordered extent so the most we can update - * our disk_i_size to is the start of the next offset. - */ - i_size_test = i_size; - for (; node; node = rb_next(node)) { - test = rb_entry(node, struct btrfs_ordered_extent, rb_node); - - if (test_bit(BTRFS_ORDERED_UPDATED_ISIZE, &test->flags)) - continue; - if (test->file_offset > offset) { - i_size_test = test->file_offset; - break; } } + new_i_size = min_t(u64, offset, i_size); /* - * i_size_test is the end of a region after this ordered - * extent where there are no ordered extents, we can safely set - * disk_i_size to this. + * Some ordered extents may completed before the current one, and + * we hold the real i_size in ->outstanding_isize. */ - if (i_size_test > offset) - new_i_size = min_t(u64, i_size_test, i_size); + if (ordered && ordered->outstanding_isize > new_i_size) + new_i_size = min_t(u64, ordered->outstanding_isize, i_size); BTRFS_I(inode)->disk_i_size = new_i_size; ret = 0; out: @@ -984,3 +950,20 @@ void btrfs_add_ordered_operation(struct btrfs_trans_handle *trans, } spin_unlock(&root->fs_info->ordered_extent_lock); } + +int __init ordered_data_init(void) +{ + btrfs_ordered_extent_cache = kmem_cache_create("btrfs_ordered_extent", + sizeof(struct btrfs_ordered_extent), 0, + SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, + NULL); + if (!btrfs_ordered_extent_cache) + return -ENOMEM; + return 0; +} + +void ordered_data_exit(void) +{ + if (btrfs_ordered_extent_cache) + kmem_cache_destroy(btrfs_ordered_extent_cache); +} diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h index e03c560d2997..dd27a0b46a37 100644 --- a/fs/btrfs/ordered-data.h +++ b/fs/btrfs/ordered-data.h @@ -96,6 +96,13 @@ struct btrfs_ordered_extent { /* number of bytes that still need writing */ u64 bytes_left; + /* + * the end of the ordered extent which is behind it but + * didn't update disk_i_size. Please see the comment of + * btrfs_ordered_update_i_size(); + */ + u64 outstanding_isize; + /* flags (described above) */ unsigned long flags; @@ -183,6 +190,7 @@ void btrfs_run_ordered_operations(struct btrfs_root *root, int wait); void btrfs_add_ordered_operation(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *inode); -void btrfs_wait_ordered_extents(struct btrfs_root *root, - int nocow_only, int delay_iput); +void btrfs_wait_ordered_extents(struct btrfs_root *root, int delay_iput); +int __init ordered_data_init(void); +void ordered_data_exit(void); #endif diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index b65015581744..fe9d02c45f8e 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -790,8 +790,10 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans, } path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; + if (!path) { + ret = -ENOMEM; + goto out_free_root; + } key.objectid = 0; key.type = BTRFS_QGROUP_STATUS_KEY; @@ -800,7 +802,7 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans, ret = btrfs_insert_empty_item(trans, quota_root, path, &key, sizeof(*ptr)); if (ret) - goto out; + goto out_free_path; leaf = path->nodes[0]; ptr = btrfs_item_ptr(leaf, path->slots[0], @@ -818,8 +820,15 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans, fs_info->quota_root = quota_root; fs_info->pending_quota_state = 1; spin_unlock(&fs_info->qgroup_lock); -out: +out_free_path: btrfs_free_path(path); +out_free_root: + if (ret) { + free_extent_buffer(quota_root->node); + free_extent_buffer(quota_root->commit_root); + kfree(quota_root); + } +out: return ret; } @@ -1145,12 +1154,12 @@ int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans, ulist_reinit(tmp); /* XXX id not needed */ - ulist_add(tmp, qg->qgroupid, (unsigned long)qg, GFP_ATOMIC); + ulist_add(tmp, qg->qgroupid, (u64)(uintptr_t)qg, GFP_ATOMIC); ULIST_ITER_INIT(&tmp_uiter); while ((tmp_unode = ulist_next(tmp, &tmp_uiter))) { struct btrfs_qgroup_list *glist; - qg = (struct btrfs_qgroup *)tmp_unode->aux; + qg = (struct btrfs_qgroup *)(uintptr_t)tmp_unode->aux; if (qg->refcnt < seq) qg->refcnt = seq + 1; else @@ -1158,7 +1167,7 @@ int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans, list_for_each_entry(glist, &qg->groups, next_group) { ulist_add(tmp, glist->group->qgroupid, - (unsigned long)glist->group, + (u64)(uintptr_t)glist->group, GFP_ATOMIC); } } @@ -1168,13 +1177,13 @@ int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans, * step 2: walk from the new root */ ulist_reinit(tmp); - ulist_add(tmp, qgroup->qgroupid, (unsigned long)qgroup, GFP_ATOMIC); + ulist_add(tmp, qgroup->qgroupid, (uintptr_t)qgroup, GFP_ATOMIC); ULIST_ITER_INIT(&uiter); while ((unode = ulist_next(tmp, &uiter))) { struct btrfs_qgroup *qg; struct btrfs_qgroup_list *glist; - qg = (struct btrfs_qgroup *)unode->aux; + qg = (struct btrfs_qgroup *)(uintptr_t)unode->aux; if (qg->refcnt < seq) { /* not visited by step 1 */ qg->rfer += sgn * node->num_bytes; @@ -1190,7 +1199,7 @@ int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans, list_for_each_entry(glist, &qg->groups, next_group) { ulist_add(tmp, glist->group->qgroupid, - (unsigned long)glist->group, GFP_ATOMIC); + (uintptr_t)glist->group, GFP_ATOMIC); } } @@ -1208,12 +1217,12 @@ int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans, continue; ulist_reinit(tmp); - ulist_add(tmp, qg->qgroupid, (unsigned long)qg, GFP_ATOMIC); + ulist_add(tmp, qg->qgroupid, (uintptr_t)qg, GFP_ATOMIC); ULIST_ITER_INIT(&tmp_uiter); while ((tmp_unode = ulist_next(tmp, &tmp_uiter))) { struct btrfs_qgroup_list *glist; - qg = (struct btrfs_qgroup *)tmp_unode->aux; + qg = (struct btrfs_qgroup *)(uintptr_t)tmp_unode->aux; if (qg->tag == seq) continue; @@ -1225,7 +1234,7 @@ int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans, list_for_each_entry(glist, &qg->groups, next_group) { ulist_add(tmp, glist->group->qgroupid, - (unsigned long)glist->group, + (uintptr_t)glist->group, GFP_ATOMIC); } } @@ -1469,13 +1478,17 @@ int btrfs_qgroup_reserve(struct btrfs_root *root, u64 num_bytes) * be exceeded */ ulist = ulist_alloc(GFP_ATOMIC); - ulist_add(ulist, qgroup->qgroupid, (unsigned long)qgroup, GFP_ATOMIC); + if (!ulist) { + ret = -ENOMEM; + goto out; + } + ulist_add(ulist, qgroup->qgroupid, (uintptr_t)qgroup, GFP_ATOMIC); ULIST_ITER_INIT(&uiter); while ((unode = ulist_next(ulist, &uiter))) { struct btrfs_qgroup *qg; struct btrfs_qgroup_list *glist; - qg = (struct btrfs_qgroup *)unode->aux; + qg = (struct btrfs_qgroup *)(uintptr_t)unode->aux; if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_RFER) && qg->reserved + qg->rfer + num_bytes > @@ -1489,7 +1502,7 @@ int btrfs_qgroup_reserve(struct btrfs_root *root, u64 num_bytes) list_for_each_entry(glist, &qg->groups, next_group) { ulist_add(ulist, glist->group->qgroupid, - (unsigned long)glist->group, GFP_ATOMIC); + (uintptr_t)glist->group, GFP_ATOMIC); } } if (ret) @@ -1502,7 +1515,7 @@ int btrfs_qgroup_reserve(struct btrfs_root *root, u64 num_bytes) while ((unode = ulist_next(ulist, &uiter))) { struct btrfs_qgroup *qg; - qg = (struct btrfs_qgroup *)unode->aux; + qg = (struct btrfs_qgroup *)(uintptr_t)unode->aux; qg->reserved += num_bytes; } @@ -1541,19 +1554,23 @@ void btrfs_qgroup_free(struct btrfs_root *root, u64 num_bytes) goto out; ulist = ulist_alloc(GFP_ATOMIC); - ulist_add(ulist, qgroup->qgroupid, (unsigned long)qgroup, GFP_ATOMIC); + if (!ulist) { + btrfs_std_error(fs_info, -ENOMEM); + goto out; + } + ulist_add(ulist, qgroup->qgroupid, (uintptr_t)qgroup, GFP_ATOMIC); ULIST_ITER_INIT(&uiter); while ((unode = ulist_next(ulist, &uiter))) { struct btrfs_qgroup *qg; struct btrfs_qgroup_list *glist; - qg = (struct btrfs_qgroup *)unode->aux; + qg = (struct btrfs_qgroup *)(uintptr_t)unode->aux; qg->reserved -= num_bytes; list_for_each_entry(glist, &qg->groups, next_group) { ulist_add(ulist, glist->group->qgroupid, - (unsigned long)glist->group, GFP_ATOMIC); + (uintptr_t)glist->group, GFP_ATOMIC); } } diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 4da08652004d..776f0aa128fc 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -3270,8 +3270,8 @@ static int delete_block_group_cache(struct btrfs_fs_info *fs_info, key.offset = 0; inode = btrfs_iget(fs_info->sb, &key, root, NULL); - if (IS_ERR_OR_NULL(inode) || is_bad_inode(inode)) { - if (inode && !IS_ERR(inode)) + if (IS_ERR(inode) || is_bad_inode(inode)) { + if (!IS_ERR(inode)) iput(inode); return -ENOENT; } @@ -3621,7 +3621,7 @@ next: ret = find_first_extent_bit(&rc->processed_blocks, key.objectid, &start, &end, - EXTENT_DIRTY); + EXTENT_DIRTY, NULL); if (ret == 0 && start <= key.objectid) { btrfs_release_path(path); @@ -3674,7 +3674,8 @@ int prepare_to_relocate(struct reloc_control *rc) struct btrfs_trans_handle *trans; int ret; - rc->block_rsv = btrfs_alloc_block_rsv(rc->extent_root); + rc->block_rsv = btrfs_alloc_block_rsv(rc->extent_root, + BTRFS_BLOCK_RSV_TEMP); if (!rc->block_rsv) return -ENOMEM; @@ -4057,7 +4058,7 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start) (unsigned long long)rc->block_group->flags); btrfs_start_delalloc_inodes(fs_info->tree_root, 0); - btrfs_wait_ordered_extents(fs_info->tree_root, 0, 0); + btrfs_wait_ordered_extents(fs_info->tree_root, 0); while (1) { mutex_lock(&fs_info->cleaner_mutex); diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c index 10d8e4d88071..eb923d087da7 100644 --- a/fs/btrfs/root-tree.c +++ b/fs/btrfs/root-tree.c @@ -141,8 +141,10 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root return -ENOMEM; ret = btrfs_search_slot(trans, root, key, path, 0, 1); - if (ret < 0) - goto out_abort; + if (ret < 0) { + btrfs_abort_transaction(trans, root, ret); + goto out; + } if (ret != 0) { btrfs_print_leaf(root, path->nodes[0]); @@ -166,16 +168,23 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root btrfs_release_path(path); ret = btrfs_search_slot(trans, root, key, path, -1, 1); - if (ret < 0) - goto out_abort; + if (ret < 0) { + btrfs_abort_transaction(trans, root, ret); + goto out; + } + ret = btrfs_del_item(trans, root, path); - if (ret < 0) - goto out_abort; + if (ret < 0) { + btrfs_abort_transaction(trans, root, ret); + goto out; + } btrfs_release_path(path); ret = btrfs_insert_empty_item(trans, root, path, key, sizeof(*item)); - if (ret < 0) - goto out_abort; + if (ret < 0) { + btrfs_abort_transaction(trans, root, ret); + goto out; + } l = path->nodes[0]; slot = path->slots[0]; ptr = btrfs_item_ptr_offset(l, slot); @@ -192,10 +201,6 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root out: btrfs_free_path(path); return ret; - -out_abort: - btrfs_abort_transaction(trans, root, ret); - goto out; } int btrfs_insert_root(struct btrfs_trans_handle *trans, struct btrfs_root *root, diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index b223620cd5a6..27892f67e69b 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -352,13 +352,14 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock) struct extent_buffer *eb; struct btrfs_extent_item *ei; struct scrub_warning swarn; - u32 item_size; - int ret; + unsigned long ptr = 0; + u64 extent_item_pos; + u64 flags = 0; u64 ref_root; + u32 item_size; u8 ref_level; - unsigned long ptr = 0; const int bufsize = 4096; - u64 extent_item_pos; + int ret; path = btrfs_alloc_path(); @@ -375,7 +376,8 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock) if (!path || !swarn.scratch_buf || !swarn.msg_buf) goto out; - ret = extent_from_logical(fs_info, swarn.logical, path, &found_key); + ret = extent_from_logical(fs_info, swarn.logical, path, &found_key, + &flags); if (ret < 0) goto out; @@ -387,7 +389,7 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock) item_size = btrfs_item_size_nr(eb, path->slots[0]); btrfs_release_path(path); - if (ret & BTRFS_EXTENT_FLAG_TREE_BLOCK) { + if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) { do { ret = tree_backref_for_extent(&ptr, eb, ei, item_size, &ref_root, &ref_level); @@ -1029,6 +1031,7 @@ static int scrub_setup_recheck_block(struct scrub_dev *sdev, spin_lock(&sdev->stat_lock); sdev->stat.malloc_errors++; spin_unlock(&sdev->stat_lock); + kfree(bbio); return -ENOMEM; } sblock->page_count++; @@ -1666,21 +1669,6 @@ static void scrub_bio_end_io_worker(struct btrfs_work *work) scrub_block_put(sblock); } - if (sbio->err) { - /* what is this good for??? */ - sbio->bio->bi_flags &= ~(BIO_POOL_MASK - 1); - sbio->bio->bi_flags |= 1 << BIO_UPTODATE; - sbio->bio->bi_phys_segments = 0; - sbio->bio->bi_idx = 0; - - for (i = 0; i < sbio->page_count; i++) { - struct bio_vec *bi; - bi = &sbio->bio->bi_io_vec[i]; - bi->bv_offset = 0; - bi->bv_len = PAGE_SIZE; - } - } - bio_put(sbio->bio); sbio->bio = NULL; spin_lock(&sdev->list_lock); diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index fb5ffe95f869..e78b297b0b00 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -107,7 +107,6 @@ struct send_ctx { int cur_inode_new; int cur_inode_new_gen; int cur_inode_deleted; - int cur_inode_first_ref_orphan; u64 cur_inode_size; u64 cur_inode_mode; @@ -126,7 +125,15 @@ struct send_ctx { struct name_cache_entry { struct list_head list; - struct list_head use_list; + /* + * radix_tree has only 32bit entries but we need to handle 64bit inums. + * We use the lower 32bit of the 64bit inum to store it in the tree. If + * more then one inum would fall into the same entry, we use radix_list + * to store the additional entries. radix_list is also used to store + * entries where two entries have the same inum but different + * generations. + */ + struct list_head radix_list; u64 ino; u64 gen; u64 parent_ino; @@ -328,6 +335,7 @@ out: return ret; } +#if 0 static void fs_path_remove(struct fs_path *p) { BUG_ON(p->reversed); @@ -335,6 +343,7 @@ static void fs_path_remove(struct fs_path *p) p->end--; *p->end = 0; } +#endif static int fs_path_copy(struct fs_path *p, struct fs_path *from) { @@ -377,7 +386,7 @@ static struct btrfs_path *alloc_path_for_send(void) return path; } -static int write_buf(struct send_ctx *sctx, const void *buf, u32 len) +int write_buf(struct file *filp, const void *buf, u32 len, loff_t *off) { int ret; mm_segment_t old_fs; @@ -387,8 +396,7 @@ static int write_buf(struct send_ctx *sctx, const void *buf, u32 len) set_fs(KERNEL_DS); while (pos < len) { - ret = vfs_write(sctx->send_filp, (char *)buf + pos, len - pos, - &sctx->send_off); + ret = vfs_write(filp, (char *)buf + pos, len - pos, off); /* TODO handle that correctly */ /*if (ret == -ERESTARTSYS) { continue; @@ -544,7 +552,8 @@ static int send_header(struct send_ctx *sctx) strcpy(hdr.magic, BTRFS_SEND_STREAM_MAGIC); hdr.version = cpu_to_le32(BTRFS_SEND_STREAM_VERSION); - return write_buf(sctx, &hdr, sizeof(hdr)); + return write_buf(sctx->send_filp, &hdr, sizeof(hdr), + &sctx->send_off); } /* @@ -581,7 +590,8 @@ static int send_cmd(struct send_ctx *sctx) crc = crc32c(0, (unsigned char *)sctx->send_buf, sctx->send_size); hdr->crc = cpu_to_le32(crc); - ret = write_buf(sctx, sctx->send_buf, sctx->send_size); + ret = write_buf(sctx->send_filp, sctx->send_buf, sctx->send_size, + &sctx->send_off); sctx->total_send_size += sctx->send_size; sctx->cmd_send_size[le16_to_cpu(hdr->cmd)] += sctx->send_size; @@ -687,7 +697,8 @@ out: */ static int get_inode_info(struct btrfs_root *root, u64 ino, u64 *size, u64 *gen, - u64 *mode, u64 *uid, u64 *gid) + u64 *mode, u64 *uid, u64 *gid, + u64 *rdev) { int ret; struct btrfs_inode_item *ii; @@ -721,6 +732,8 @@ static int get_inode_info(struct btrfs_root *root, *uid = btrfs_inode_uid(path->nodes[0], ii); if (gid) *gid = btrfs_inode_gid(path->nodes[0], ii); + if (rdev) + *rdev = btrfs_inode_rdev(path->nodes[0], ii); out: btrfs_free_path(path); @@ -732,31 +745,36 @@ typedef int (*iterate_inode_ref_t)(int num, u64 dir, int index, void *ctx); /* - * Helper function to iterate the entries in ONE btrfs_inode_ref. + * Helper function to iterate the entries in ONE btrfs_inode_ref or + * btrfs_inode_extref. * The iterate callback may return a non zero value to stop iteration. This can * be a negative value for error codes or 1 to simply stop it. * - * path must point to the INODE_REF when called. + * path must point to the INODE_REF or INODE_EXTREF when called. */ static int iterate_inode_ref(struct send_ctx *sctx, struct btrfs_root *root, struct btrfs_path *path, struct btrfs_key *found_key, int resolve, iterate_inode_ref_t iterate, void *ctx) { - struct extent_buffer *eb; + struct extent_buffer *eb = path->nodes[0]; struct btrfs_item *item; struct btrfs_inode_ref *iref; + struct btrfs_inode_extref *extref; struct btrfs_path *tmp_path; struct fs_path *p; - u32 cur; - u32 len; + u32 cur = 0; u32 total; - int slot; + int slot = path->slots[0]; u32 name_len; char *start; int ret = 0; - int num; + int num = 0; int index; + u64 dir; + unsigned long name_off; + unsigned long elem_size; + unsigned long ptr; p = fs_path_alloc_reversed(sctx); if (!p) @@ -768,24 +786,40 @@ static int iterate_inode_ref(struct send_ctx *sctx, return -ENOMEM; } - eb = path->nodes[0]; - slot = path->slots[0]; - item = btrfs_item_nr(eb, slot); - iref = btrfs_item_ptr(eb, slot, struct btrfs_inode_ref); - cur = 0; - len = 0; - total = btrfs_item_size(eb, item); - num = 0; + if (found_key->type == BTRFS_INODE_REF_KEY) { + ptr = (unsigned long)btrfs_item_ptr(eb, slot, + struct btrfs_inode_ref); + item = btrfs_item_nr(eb, slot); + total = btrfs_item_size(eb, item); + elem_size = sizeof(*iref); + } else { + ptr = btrfs_item_ptr_offset(eb, slot); + total = btrfs_item_size_nr(eb, slot); + elem_size = sizeof(*extref); + } + while (cur < total) { fs_path_reset(p); - name_len = btrfs_inode_ref_name_len(eb, iref); - index = btrfs_inode_ref_index(eb, iref); + if (found_key->type == BTRFS_INODE_REF_KEY) { + iref = (struct btrfs_inode_ref *)(ptr + cur); + name_len = btrfs_inode_ref_name_len(eb, iref); + name_off = (unsigned long)(iref + 1); + index = btrfs_inode_ref_index(eb, iref); + dir = found_key->offset; + } else { + extref = (struct btrfs_inode_extref *)(ptr + cur); + name_len = btrfs_inode_extref_name_len(eb, extref); + name_off = (unsigned long)&extref->name; + index = btrfs_inode_extref_index(eb, extref); + dir = btrfs_inode_extref_parent(eb, extref); + } + if (resolve) { - start = btrfs_iref_to_path(root, tmp_path, iref, eb, - found_key->offset, p->buf, - p->buf_len); + start = btrfs_ref_to_path(root, tmp_path, name_len, + name_off, eb, dir, + p->buf, p->buf_len); if (IS_ERR(start)) { ret = PTR_ERR(start); goto out; @@ -796,9 +830,10 @@ static int iterate_inode_ref(struct send_ctx *sctx, p->buf_len + p->buf - start); if (ret < 0) goto out; - start = btrfs_iref_to_path(root, tmp_path, iref, - eb, found_key->offset, p->buf, - p->buf_len); + start = btrfs_ref_to_path(root, tmp_path, + name_len, name_off, + eb, dir, + p->buf, p->buf_len); if (IS_ERR(start)) { ret = PTR_ERR(start); goto out; @@ -807,21 +842,16 @@ static int iterate_inode_ref(struct send_ctx *sctx, } p->start = start; } else { - ret = fs_path_add_from_extent_buffer(p, eb, - (unsigned long)(iref + 1), name_len); + ret = fs_path_add_from_extent_buffer(p, eb, name_off, + name_len); if (ret < 0) goto out; } - - len = sizeof(*iref) + name_len; - iref = (struct btrfs_inode_ref *)((char *)iref + len); - cur += len; - - ret = iterate(num, found_key->offset, index, p, ctx); + cur += elem_size + name_len; + ret = iterate(num, dir, index, p, ctx); if (ret) goto out; - num++; } @@ -852,7 +882,6 @@ static int iterate_dir_item(struct send_ctx *sctx, struct extent_buffer *eb; struct btrfs_item *item; struct btrfs_dir_item *di; - struct btrfs_path *tmp_path = NULL; struct btrfs_key di_key; char *buf = NULL; char *buf2 = NULL; @@ -874,12 +903,6 @@ static int iterate_dir_item(struct send_ctx *sctx, goto out; } - tmp_path = alloc_path_for_send(); - if (!tmp_path) { - ret = -ENOMEM; - goto out; - } - eb = path->nodes[0]; slot = path->slots[0]; item = btrfs_item_nr(eb, slot); @@ -941,7 +964,6 @@ static int iterate_dir_item(struct send_ctx *sctx, } out: - btrfs_free_path(tmp_path); if (buf_virtual) vfree(buf); else @@ -993,7 +1015,8 @@ static int get_inode_path(struct send_ctx *sctx, struct btrfs_root *root, } btrfs_item_key_to_cpu(p->nodes[0], &found_key, p->slots[0]); if (found_key.objectid != ino || - found_key.type != BTRFS_INODE_REF_KEY) { + (found_key.type != BTRFS_INODE_REF_KEY && + found_key.type != BTRFS_INODE_EXTREF_KEY)) { ret = -ENOENT; goto out; } @@ -1026,12 +1049,12 @@ struct backref_ctx { u64 extent_len; /* Just to check for bugs in backref resolving */ - int found_in_send_root; + int found_itself; }; static int __clone_root_cmp_bsearch(const void *key, const void *elt) { - u64 root = (u64)key; + u64 root = (u64)(uintptr_t)key; struct clone_root *cr = (struct clone_root *)elt; if (root < cr->root->objectid) @@ -1055,6 +1078,7 @@ static int __clone_root_cmp_sort(const void *e1, const void *e2) /* * Called for every backref that is found for the current extent. + * Results are collected in sctx->clone_roots->ino/offset/found_refs */ static int __iterate_backrefs(u64 ino, u64 offset, u64 root, void *ctx_) { @@ -1064,7 +1088,7 @@ static int __iterate_backrefs(u64 ino, u64 offset, u64 root, void *ctx_) u64 i_size; /* First check if the root is in the list of accepted clone sources */ - found = bsearch((void *)root, bctx->sctx->clone_roots, + found = bsearch((void *)(uintptr_t)root, bctx->sctx->clone_roots, bctx->sctx->clone_roots_cnt, sizeof(struct clone_root), __clone_root_cmp_bsearch); @@ -1074,14 +1098,15 @@ static int __iterate_backrefs(u64 ino, u64 offset, u64 root, void *ctx_) if (found->root == bctx->sctx->send_root && ino == bctx->cur_objectid && offset == bctx->cur_offset) { - bctx->found_in_send_root = 1; + bctx->found_itself = 1; } /* - * There are inodes that have extents that lie behind it's i_size. Don't + * There are inodes that have extents that lie behind its i_size. Don't * accept clones from these extents. */ - ret = get_inode_info(found->root, ino, &i_size, NULL, NULL, NULL, NULL); + ret = get_inode_info(found->root, ino, &i_size, NULL, NULL, NULL, NULL, + NULL); if (ret < 0) return ret; @@ -1101,16 +1126,12 @@ static int __iterate_backrefs(u64 ino, u64 offset, u64 root, void *ctx_) */ if (ino >= bctx->cur_objectid) return 0; - /*if (ino > ctx->cur_objectid) +#if 0 + if (ino > bctx->cur_objectid) return 0; - if (offset + ctx->extent_len > ctx->cur_offset) - return 0;*/ - - bctx->found++; - found->found_refs++; - found->ino = ino; - found->offset = offset; - return 0; + if (offset + bctx->extent_len > bctx->cur_offset) + return 0; +#endif } bctx->found++; @@ -1130,6 +1151,12 @@ static int __iterate_backrefs(u64 ino, u64 offset, u64 root, void *ctx_) } /* + * Given an inode, offset and extent item, it finds a good clone for a clone + * instruction. Returns -ENOENT when none could be found. The function makes + * sure that the returned clone is usable at the point where sending is at the + * moment. This means, that no clones are accepted which lie behind the current + * inode+offset. + * * path must point to the extent item when called. */ static int find_extent_clone(struct send_ctx *sctx, @@ -1141,20 +1168,29 @@ static int find_extent_clone(struct send_ctx *sctx, int ret; int extent_type; u64 logical; + u64 disk_byte; u64 num_bytes; u64 extent_item_pos; + u64 flags = 0; struct btrfs_file_extent_item *fi; struct extent_buffer *eb = path->nodes[0]; - struct backref_ctx backref_ctx; + struct backref_ctx *backref_ctx = NULL; struct clone_root *cur_clone_root; struct btrfs_key found_key; struct btrfs_path *tmp_path; + int compressed; u32 i; tmp_path = alloc_path_for_send(); if (!tmp_path) return -ENOMEM; + backref_ctx = kmalloc(sizeof(*backref_ctx), GFP_NOFS); + if (!backref_ctx) { + ret = -ENOMEM; + goto out; + } + if (data_offset >= ino_size) { /* * There may be extents that lie behind the file's size. @@ -1172,22 +1208,23 @@ static int find_extent_clone(struct send_ctx *sctx, ret = -ENOENT; goto out; } + compressed = btrfs_file_extent_compression(eb, fi); num_bytes = btrfs_file_extent_num_bytes(eb, fi); - logical = btrfs_file_extent_disk_bytenr(eb, fi); - if (logical == 0) { + disk_byte = btrfs_file_extent_disk_bytenr(eb, fi); + if (disk_byte == 0) { ret = -ENOENT; goto out; } - logical += btrfs_file_extent_offset(eb, fi); + logical = disk_byte + btrfs_file_extent_offset(eb, fi); - ret = extent_from_logical(sctx->send_root->fs_info, - logical, tmp_path, &found_key); + ret = extent_from_logical(sctx->send_root->fs_info, disk_byte, tmp_path, + &found_key, &flags); btrfs_release_path(tmp_path); if (ret < 0) goto out; - if (ret & BTRFS_EXTENT_FLAG_TREE_BLOCK) { + if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) { ret = -EIO; goto out; } @@ -1202,12 +1239,12 @@ static int find_extent_clone(struct send_ctx *sctx, cur_clone_root->found_refs = 0; } - backref_ctx.sctx = sctx; - backref_ctx.found = 0; - backref_ctx.cur_objectid = ino; - backref_ctx.cur_offset = data_offset; - backref_ctx.found_in_send_root = 0; - backref_ctx.extent_len = num_bytes; + backref_ctx->sctx = sctx; + backref_ctx->found = 0; + backref_ctx->cur_objectid = ino; + backref_ctx->cur_offset = data_offset; + backref_ctx->found_itself = 0; + backref_ctx->extent_len = num_bytes; /* * The last extent of a file may be too large due to page alignment. @@ -1215,25 +1252,31 @@ static int find_extent_clone(struct send_ctx *sctx, * __iterate_backrefs work. */ if (data_offset + num_bytes >= ino_size) - backref_ctx.extent_len = ino_size - data_offset; + backref_ctx->extent_len = ino_size - data_offset; /* * Now collect all backrefs. */ + if (compressed == BTRFS_COMPRESS_NONE) + extent_item_pos = logical - found_key.objectid; + else + extent_item_pos = 0; + extent_item_pos = logical - found_key.objectid; ret = iterate_extent_inodes(sctx->send_root->fs_info, found_key.objectid, extent_item_pos, 1, - __iterate_backrefs, &backref_ctx); + __iterate_backrefs, backref_ctx); + if (ret < 0) goto out; - if (!backref_ctx.found_in_send_root) { + if (!backref_ctx->found_itself) { /* found a bug in backref code? */ ret = -EIO; printk(KERN_ERR "btrfs: ERROR did not find backref in " "send_root. inode=%llu, offset=%llu, " - "logical=%llu\n", - ino, data_offset, logical); + "disk_byte=%llu found extent=%llu\n", + ino, data_offset, disk_byte, found_key.objectid); goto out; } @@ -1242,7 +1285,7 @@ verbose_printk(KERN_DEBUG "btrfs: find_extent_clone: data_offset=%llu, " "num_bytes=%llu, logical=%llu\n", data_offset, ino, num_bytes, logical); - if (!backref_ctx.found) + if (!backref_ctx->found) verbose_printk("btrfs: no clones found\n"); cur_clone_root = NULL; @@ -1253,7 +1296,6 @@ verbose_printk(KERN_DEBUG "btrfs: find_extent_clone: data_offset=%llu, " else if (sctx->clone_roots[i].root == sctx->send_root) /* prefer clones from send_root over others */ cur_clone_root = sctx->clone_roots + i; - break; } } @@ -1267,6 +1309,7 @@ verbose_printk(KERN_DEBUG "btrfs: find_extent_clone: data_offset=%llu, " out: btrfs_free_path(tmp_path); + kfree(backref_ctx); return ret; } @@ -1307,8 +1350,6 @@ static int read_symlink(struct send_ctx *sctx, len = btrfs_file_extent_inline_len(path->nodes[0], ei); ret = fs_path_add_from_extent_buffer(dest, path->nodes[0], off, len); - if (ret < 0) - goto out; out: btrfs_free_path(path); @@ -1404,7 +1445,7 @@ static int get_cur_inode_state(struct send_ctx *sctx, u64 ino, u64 gen) u64 right_gen; ret = get_inode_info(sctx->send_root, ino, NULL, &left_gen, NULL, NULL, - NULL); + NULL, NULL); if (ret < 0 && ret != -ENOENT) goto out; left_ret = ret; @@ -1413,16 +1454,16 @@ static int get_cur_inode_state(struct send_ctx *sctx, u64 ino, u64 gen) right_ret = -ENOENT; } else { ret = get_inode_info(sctx->parent_root, ino, NULL, &right_gen, - NULL, NULL, NULL); + NULL, NULL, NULL, NULL); if (ret < 0 && ret != -ENOENT) goto out; right_ret = ret; } if (!left_ret && !right_ret) { - if (left_gen == gen && right_gen == gen) + if (left_gen == gen && right_gen == gen) { ret = inode_state_no_change; - else if (left_gen == gen) { + } else if (left_gen == gen) { if (ino < sctx->send_progress) ret = inode_state_did_create; else @@ -1516,6 +1557,10 @@ out: return ret; } +/* + * Looks up the first btrfs_inode_ref of a given ino. It returns the parent dir, + * generation of the parent dir and the name of the dir entry. + */ static int get_first_ref(struct send_ctx *sctx, struct btrfs_root *root, u64 ino, u64 *dir, u64 *dir_gen, struct fs_path *name) @@ -1524,8 +1569,8 @@ static int get_first_ref(struct send_ctx *sctx, struct btrfs_key key; struct btrfs_key found_key; struct btrfs_path *path; - struct btrfs_inode_ref *iref; int len; + u64 parent_dir; path = alloc_path_for_send(); if (!path) @@ -1541,27 +1586,41 @@ static int get_first_ref(struct send_ctx *sctx, if (!ret) btrfs_item_key_to_cpu(path->nodes[0], &found_key, path->slots[0]); - if (ret || found_key.objectid != key.objectid || - found_key.type != key.type) { + if (ret || found_key.objectid != ino || + (found_key.type != BTRFS_INODE_REF_KEY && + found_key.type != BTRFS_INODE_EXTREF_KEY)) { ret = -ENOENT; goto out; } - iref = btrfs_item_ptr(path->nodes[0], path->slots[0], - struct btrfs_inode_ref); - len = btrfs_inode_ref_name_len(path->nodes[0], iref); - ret = fs_path_add_from_extent_buffer(name, path->nodes[0], - (unsigned long)(iref + 1), len); + if (key.type == BTRFS_INODE_REF_KEY) { + struct btrfs_inode_ref *iref; + iref = btrfs_item_ptr(path->nodes[0], path->slots[0], + struct btrfs_inode_ref); + len = btrfs_inode_ref_name_len(path->nodes[0], iref); + ret = fs_path_add_from_extent_buffer(name, path->nodes[0], + (unsigned long)(iref + 1), + len); + parent_dir = found_key.offset; + } else { + struct btrfs_inode_extref *extref; + extref = btrfs_item_ptr(path->nodes[0], path->slots[0], + struct btrfs_inode_extref); + len = btrfs_inode_extref_name_len(path->nodes[0], extref); + ret = fs_path_add_from_extent_buffer(name, path->nodes[0], + (unsigned long)&extref->name, len); + parent_dir = btrfs_inode_extref_parent(path->nodes[0], extref); + } if (ret < 0) goto out; btrfs_release_path(path); - ret = get_inode_info(root, found_key.offset, NULL, dir_gen, NULL, NULL, - NULL); + ret = get_inode_info(root, parent_dir, NULL, dir_gen, NULL, NULL, + NULL, NULL); if (ret < 0) goto out; - *dir = found_key.offset; + *dir = parent_dir; out: btrfs_free_path(path); @@ -1586,22 +1645,28 @@ static int is_first_ref(struct send_ctx *sctx, if (ret < 0) goto out; - if (name_len != fs_path_len(tmp_name)) { + if (dir != tmp_dir || name_len != fs_path_len(tmp_name)) { ret = 0; goto out; } - ret = memcmp(tmp_name->start, name, name_len); - if (ret) - ret = 0; - else - ret = 1; + ret = !memcmp(tmp_name->start, name, name_len); out: fs_path_free(sctx, tmp_name); return ret; } +/* + * Used by process_recorded_refs to determine if a new ref would overwrite an + * already existing ref. In case it detects an overwrite, it returns the + * inode/gen in who_ino/who_gen. + * When an overwrite is detected, process_recorded_refs does proper orphanizing + * to make sure later references to the overwritten inode are possible. + * Orphanizing is however only required for the first ref of an inode. + * process_recorded_refs does an additional is_first_ref check to see if + * orphanizing is really required. + */ static int will_overwrite_ref(struct send_ctx *sctx, u64 dir, u64 dir_gen, const char *name, int name_len, u64 *who_ino, u64 *who_gen) @@ -1626,9 +1691,14 @@ static int will_overwrite_ref(struct send_ctx *sctx, u64 dir, u64 dir_gen, goto out; } + /* + * Check if the overwritten ref was already processed. If yes, the ref + * was already unlinked/moved, so we can safely assume that we will not + * overwrite anything at this point in time. + */ if (other_inode > sctx->send_progress) { ret = get_inode_info(sctx->parent_root, other_inode, NULL, - who_gen, NULL, NULL, NULL); + who_gen, NULL, NULL, NULL, NULL); if (ret < 0) goto out; @@ -1642,6 +1712,13 @@ out: return ret; } +/* + * Checks if the ref was overwritten by an already processed inode. This is + * used by __get_cur_name_and_parent to find out if the ref was orphanized and + * thus the orphan name needs be used. + * process_recorded_refs also uses it to avoid unlinking of refs that were + * overwritten. + */ static int did_overwrite_ref(struct send_ctx *sctx, u64 dir, u64 dir_gen, u64 ino, u64 ino_gen, @@ -1671,7 +1748,7 @@ static int did_overwrite_ref(struct send_ctx *sctx, } ret = get_inode_info(sctx->send_root, ow_inode, NULL, &gen, NULL, NULL, - NULL); + NULL, NULL); if (ret < 0) goto out; @@ -1690,6 +1767,11 @@ out: return ret; } +/* + * Same as did_overwrite_ref, but also checks if it is the first ref of an inode + * that got overwritten. This is used by process_recorded_refs to determine + * if it has to use the path as returned by get_cur_path or the orphan name. + */ static int did_overwrite_first_ref(struct send_ctx *sctx, u64 ino, u64 gen) { int ret = 0; @@ -1710,39 +1792,40 @@ static int did_overwrite_first_ref(struct send_ctx *sctx, u64 ino, u64 gen) ret = did_overwrite_ref(sctx, dir, dir_gen, ino, gen, name->start, fs_path_len(name)); - if (ret < 0) - goto out; out: fs_path_free(sctx, name); return ret; } +/* + * Insert a name cache entry. On 32bit kernels the radix tree index is 32bit, + * so we need to do some special handling in case we have clashes. This function + * takes care of this with the help of name_cache_entry::radix_list. + * In case of error, nce is kfreed. + */ static int name_cache_insert(struct send_ctx *sctx, struct name_cache_entry *nce) { int ret = 0; - struct name_cache_entry **ncea; - - ncea = radix_tree_lookup(&sctx->name_cache, nce->ino); - if (ncea) { - if (!ncea[0]) - ncea[0] = nce; - else if (!ncea[1]) - ncea[1] = nce; - else - BUG(); - } else { - ncea = kmalloc(sizeof(void *) * 2, GFP_NOFS); - if (!ncea) + struct list_head *nce_head; + + nce_head = radix_tree_lookup(&sctx->name_cache, + (unsigned long)nce->ino); + if (!nce_head) { + nce_head = kmalloc(sizeof(*nce_head), GFP_NOFS); + if (!nce_head) return -ENOMEM; + INIT_LIST_HEAD(nce_head); - ncea[0] = nce; - ncea[1] = NULL; - ret = radix_tree_insert(&sctx->name_cache, nce->ino, ncea); - if (ret < 0) + ret = radix_tree_insert(&sctx->name_cache, nce->ino, nce_head); + if (ret < 0) { + kfree(nce_head); + kfree(nce); return ret; + } } + list_add_tail(&nce->radix_list, nce_head); list_add_tail(&nce->list, &sctx->name_cache_list); sctx->name_cache_size++; @@ -1752,50 +1835,52 @@ static int name_cache_insert(struct send_ctx *sctx, static void name_cache_delete(struct send_ctx *sctx, struct name_cache_entry *nce) { - struct name_cache_entry **ncea; - - ncea = radix_tree_lookup(&sctx->name_cache, nce->ino); - BUG_ON(!ncea); - - if (ncea[0] == nce) - ncea[0] = NULL; - else if (ncea[1] == nce) - ncea[1] = NULL; - else - BUG(); + struct list_head *nce_head; - if (!ncea[0] && !ncea[1]) { - radix_tree_delete(&sctx->name_cache, nce->ino); - kfree(ncea); - } + nce_head = radix_tree_lookup(&sctx->name_cache, + (unsigned long)nce->ino); + BUG_ON(!nce_head); + list_del(&nce->radix_list); list_del(&nce->list); - sctx->name_cache_size--; + + if (list_empty(nce_head)) { + radix_tree_delete(&sctx->name_cache, (unsigned long)nce->ino); + kfree(nce_head); + } } static struct name_cache_entry *name_cache_search(struct send_ctx *sctx, u64 ino, u64 gen) { - struct name_cache_entry **ncea; + struct list_head *nce_head; + struct name_cache_entry *cur; - ncea = radix_tree_lookup(&sctx->name_cache, ino); - if (!ncea) + nce_head = radix_tree_lookup(&sctx->name_cache, (unsigned long)ino); + if (!nce_head) return NULL; - if (ncea[0] && ncea[0]->gen == gen) - return ncea[0]; - else if (ncea[1] && ncea[1]->gen == gen) - return ncea[1]; + list_for_each_entry(cur, nce_head, radix_list) { + if (cur->ino == ino && cur->gen == gen) + return cur; + } return NULL; } +/* + * Removes the entry from the list and adds it back to the end. This marks the + * entry as recently used so that name_cache_clean_unused does not remove it. + */ static void name_cache_used(struct send_ctx *sctx, struct name_cache_entry *nce) { list_del(&nce->list); list_add_tail(&nce->list, &sctx->name_cache_list); } +/* + * Remove some entries from the beginning of name_cache_list. + */ static void name_cache_clean_unused(struct send_ctx *sctx) { struct name_cache_entry *nce; @@ -1814,13 +1899,23 @@ static void name_cache_clean_unused(struct send_ctx *sctx) static void name_cache_free(struct send_ctx *sctx) { struct name_cache_entry *nce; - struct name_cache_entry *tmp; - list_for_each_entry_safe(nce, tmp, &sctx->name_cache_list, list) { + while (!list_empty(&sctx->name_cache_list)) { + nce = list_entry(sctx->name_cache_list.next, + struct name_cache_entry, list); name_cache_delete(sctx, nce); + kfree(nce); } } +/* + * Used by get_cur_path for each ref up to the root. + * Returns 0 if it succeeded. + * Returns 1 if the inode is not existent or got overwritten. In that case, the + * name is an orphan name. This instructs get_cur_path to stop iterating. If 1 + * is returned, parent_ino/parent_gen are not guaranteed to be valid. + * Returns <0 in case of error. + */ static int __get_cur_name_and_parent(struct send_ctx *sctx, u64 ino, u64 gen, u64 *parent_ino, @@ -1832,6 +1927,11 @@ static int __get_cur_name_and_parent(struct send_ctx *sctx, struct btrfs_path *path = NULL; struct name_cache_entry *nce = NULL; + /* + * First check if we already did a call to this function with the same + * ino/gen. If yes, check if the cache entry is still up-to-date. If yes + * return the cached result. + */ nce = name_cache_search(sctx, ino, gen); if (nce) { if (ino < sctx->send_progress && nce->need_later_update) { @@ -1854,6 +1954,11 @@ static int __get_cur_name_and_parent(struct send_ctx *sctx, if (!path) return -ENOMEM; + /* + * If the inode is not existent yet, add the orphan name and return 1. + * This should only happen for the parent dir that we determine in + * __record_new_ref + */ ret = is_inode_existent(sctx, ino, gen); if (ret < 0) goto out; @@ -1866,6 +1971,10 @@ static int __get_cur_name_and_parent(struct send_ctx *sctx, goto out_cache; } + /* + * Depending on whether the inode was already processed or not, use + * send_root or parent_root for ref lookup. + */ if (ino < sctx->send_progress) ret = get_first_ref(sctx, sctx->send_root, ino, parent_ino, parent_gen, dest); @@ -1875,6 +1984,10 @@ static int __get_cur_name_and_parent(struct send_ctx *sctx, if (ret < 0) goto out; + /* + * Check if the ref was overwritten by an inode's ref that was processed + * earlier. If yes, treat as orphan and return 1. + */ ret = did_overwrite_ref(sctx, *parent_ino, *parent_gen, ino, gen, dest->start, dest->end - dest->start); if (ret < 0) @@ -1888,6 +2001,9 @@ static int __get_cur_name_and_parent(struct send_ctx *sctx, } out_cache: + /* + * Store the result of the lookup in the name cache. + */ nce = kmalloc(sizeof(*nce) + fs_path_len(dest) + 1, GFP_NOFS); if (!nce) { ret = -ENOMEM; @@ -1901,7 +2017,6 @@ out_cache: nce->name_len = fs_path_len(dest); nce->ret = ret; strcpy(nce->name, dest->start); - memset(&nce->use_list, 0, sizeof(nce->use_list)); if (ino < sctx->send_progress) nce->need_later_update = 0; @@ -2107,9 +2222,6 @@ static int send_subvol_begin(struct send_ctx *sctx) read_extent_buffer(leaf, name, (unsigned long)(ref + 1), namelen); btrfs_release_path(path); - if (ret < 0) - goto out; - if (parent_root) { ret = begin_cmd(sctx, BTRFS_SEND_C_SNAPSHOT); if (ret < 0) @@ -2276,7 +2388,7 @@ verbose_printk("btrfs: send_utimes %llu\n", ino); btrfs_inode_mtime(ii)); TLV_PUT_BTRFS_TIMESPEC(sctx, BTRFS_SEND_A_CTIME, eb, btrfs_inode_ctime(ii)); - /* TODO otime? */ + /* TODO Add otime support when the otime patches get into upstream */ ret = send_cmd(sctx); @@ -2292,39 +2404,39 @@ out: * a valid path yet because we did not process the refs yet. So, the inode * is created as orphan. */ -static int send_create_inode(struct send_ctx *sctx, struct btrfs_path *path, - struct btrfs_key *key) +static int send_create_inode(struct send_ctx *sctx, u64 ino) { int ret = 0; - struct extent_buffer *eb = path->nodes[0]; - struct btrfs_inode_item *ii; struct fs_path *p; - int slot = path->slots[0]; int cmd; + u64 gen; u64 mode; + u64 rdev; -verbose_printk("btrfs: send_create_inode %llu\n", sctx->cur_ino); +verbose_printk("btrfs: send_create_inode %llu\n", ino); p = fs_path_alloc(sctx); if (!p) return -ENOMEM; - ii = btrfs_item_ptr(eb, slot, struct btrfs_inode_item); - mode = btrfs_inode_mode(eb, ii); + ret = get_inode_info(sctx->send_root, ino, NULL, &gen, &mode, NULL, + NULL, &rdev); + if (ret < 0) + goto out; - if (S_ISREG(mode)) + if (S_ISREG(mode)) { cmd = BTRFS_SEND_C_MKFILE; - else if (S_ISDIR(mode)) + } else if (S_ISDIR(mode)) { cmd = BTRFS_SEND_C_MKDIR; - else if (S_ISLNK(mode)) + } else if (S_ISLNK(mode)) { cmd = BTRFS_SEND_C_SYMLINK; - else if (S_ISCHR(mode) || S_ISBLK(mode)) + } else if (S_ISCHR(mode) || S_ISBLK(mode)) { cmd = BTRFS_SEND_C_MKNOD; - else if (S_ISFIFO(mode)) + } else if (S_ISFIFO(mode)) { cmd = BTRFS_SEND_C_MKFIFO; - else if (S_ISSOCK(mode)) + } else if (S_ISSOCK(mode)) { cmd = BTRFS_SEND_C_MKSOCK; - else { + } else { printk(KERN_WARNING "btrfs: unexpected inode type %o", (int)(mode & S_IFMT)); ret = -ENOTSUPP; @@ -2335,22 +2447,23 @@ verbose_printk("btrfs: send_create_inode %llu\n", sctx->cur_ino); if (ret < 0) goto out; - ret = gen_unique_name(sctx, sctx->cur_ino, sctx->cur_inode_gen, p); + ret = gen_unique_name(sctx, ino, gen, p); if (ret < 0) goto out; TLV_PUT_PATH(sctx, BTRFS_SEND_A_PATH, p); - TLV_PUT_U64(sctx, BTRFS_SEND_A_INO, sctx->cur_ino); + TLV_PUT_U64(sctx, BTRFS_SEND_A_INO, ino); if (S_ISLNK(mode)) { fs_path_reset(p); - ret = read_symlink(sctx, sctx->send_root, sctx->cur_ino, p); + ret = read_symlink(sctx, sctx->send_root, ino, p); if (ret < 0) goto out; TLV_PUT_PATH(sctx, BTRFS_SEND_A_PATH_LINK, p); } else if (S_ISCHR(mode) || S_ISBLK(mode) || S_ISFIFO(mode) || S_ISSOCK(mode)) { - TLV_PUT_U64(sctx, BTRFS_SEND_A_RDEV, btrfs_inode_rdev(eb, ii)); + TLV_PUT_U64(sctx, BTRFS_SEND_A_RDEV, new_encode_dev(rdev)); + TLV_PUT_U64(sctx, BTRFS_SEND_A_MODE, mode); } ret = send_cmd(sctx); @@ -2364,6 +2477,92 @@ out: return ret; } +/* + * We need some special handling for inodes that get processed before the parent + * directory got created. See process_recorded_refs for details. + * This function does the check if we already created the dir out of order. + */ +static int did_create_dir(struct send_ctx *sctx, u64 dir) +{ + int ret = 0; + struct btrfs_path *path = NULL; + struct btrfs_key key; + struct btrfs_key found_key; + struct btrfs_key di_key; + struct extent_buffer *eb; + struct btrfs_dir_item *di; + int slot; + + path = alloc_path_for_send(); + if (!path) { + ret = -ENOMEM; + goto out; + } + + key.objectid = dir; + key.type = BTRFS_DIR_INDEX_KEY; + key.offset = 0; + while (1) { + ret = btrfs_search_slot_for_read(sctx->send_root, &key, path, + 1, 0); + if (ret < 0) + goto out; + if (!ret) { + eb = path->nodes[0]; + slot = path->slots[0]; + btrfs_item_key_to_cpu(eb, &found_key, slot); + } + if (ret || found_key.objectid != key.objectid || + found_key.type != key.type) { + ret = 0; + goto out; + } + + di = btrfs_item_ptr(eb, slot, struct btrfs_dir_item); + btrfs_dir_item_key_to_cpu(eb, di, &di_key); + + if (di_key.objectid < sctx->send_progress) { + ret = 1; + goto out; + } + + key.offset = found_key.offset + 1; + btrfs_release_path(path); + } + +out: + btrfs_free_path(path); + return ret; +} + +/* + * Only creates the inode if it is: + * 1. Not a directory + * 2. Or a directory which was not created already due to out of order + * directories. See did_create_dir and process_recorded_refs for details. + */ +static int send_create_inode_if_needed(struct send_ctx *sctx) +{ + int ret; + + if (S_ISDIR(sctx->cur_inode_mode)) { + ret = did_create_dir(sctx, sctx->cur_ino); + if (ret < 0) + goto out; + if (ret) { + ret = 0; + goto out; + } + } + + ret = send_create_inode(sctx, sctx->cur_ino); + if (ret < 0) + goto out; + +out: + return ret; +} + struct recorded_ref { struct list_head list; char *dir_path; @@ -2416,13 +2615,13 @@ static int record_ref(struct list_head *head, u64 dir, static void __free_recorded_refs(struct send_ctx *sctx, struct list_head *head) { struct recorded_ref *cur; - struct recorded_ref *tmp; - list_for_each_entry_safe(cur, tmp, head, list) { + while (!list_empty(head)) { + cur = list_entry(head->next, struct recorded_ref, list); fs_path_free(sctx, cur->full_path); + list_del(&cur->list); kfree(cur); } - INIT_LIST_HEAD(head); } static void free_recorded_refs(struct send_ctx *sctx) @@ -2432,7 +2631,7 @@ static void free_recorded_refs(struct send_ctx *sctx) } /* - * Renames/moves a file/dir to it's orphan name. Used when the first + * Renames/moves a file/dir to its orphan name. Used when the first * ref of an unprocessed inode gets overwritten and for all non empty * directories. */ @@ -2472,6 +2671,12 @@ static int can_rmdir(struct send_ctx *sctx, u64 dir, u64 send_progress) struct btrfs_key loc; struct btrfs_dir_item *di; + /* + * Don't try to rmdir the top/root subvolume dir. + */ + if (dir == BTRFS_FIRST_FREE_OBJECTID) + return 0; + path = alloc_path_for_send(); if (!path) return -ENOMEM; @@ -2513,160 +2718,6 @@ out: return ret; } -struct finish_unordered_dir_ctx { - struct send_ctx *sctx; - struct fs_path *cur_path; - struct fs_path *dir_path; - u64 dir_ino; - int need_delete; - int delete_pass; -}; - -int __finish_unordered_dir(int num, struct btrfs_key *di_key, - const char *name, int name_len, - const char *data, int data_len, - u8 type, void *ctx) -{ - int ret = 0; - struct finish_unordered_dir_ctx *fctx = ctx; - struct send_ctx *sctx = fctx->sctx; - u64 di_gen; - u64 di_mode; - int is_orphan = 0; - - if (di_key->objectid >= fctx->dir_ino) - goto out; - - fs_path_reset(fctx->cur_path); - - ret = get_inode_info(sctx->send_root, di_key->objectid, - NULL, &di_gen, &di_mode, NULL, NULL); - if (ret < 0) - goto out; - - ret = is_first_ref(sctx, sctx->send_root, di_key->objectid, - fctx->dir_ino, name, name_len); - if (ret < 0) - goto out; - if (ret) { - is_orphan = 1; - ret = gen_unique_name(sctx, di_key->objectid, di_gen, - fctx->cur_path); - } else { - ret = get_cur_path(sctx, di_key->objectid, di_gen, - fctx->cur_path); - } - if (ret < 0) - goto out; - - ret = fs_path_add(fctx->dir_path, name, name_len); - if (ret < 0) - goto out; - - if (!fctx->delete_pass) { - if (S_ISDIR(di_mode)) { - ret = send_rename(sctx, fctx->cur_path, - fctx->dir_path); - } else { - ret = send_link(sctx, fctx->dir_path, - fctx->cur_path); - if (is_orphan) - fctx->need_delete = 1; - } - } else if (!S_ISDIR(di_mode)) { - ret = send_unlink(sctx, fctx->cur_path); - } else { - ret = 0; - } - - fs_path_remove(fctx->dir_path); - -out: - return ret; -} - -/* - * Go through all dir items and see if we find refs which could not be created - * in the past because the dir did not exist at that time. - */ -static int finish_outoforder_dir(struct send_ctx *sctx, u64 dir, u64 dir_gen) -{ - int ret = 0; - struct btrfs_path *path = NULL; - struct btrfs_key key; - struct btrfs_key found_key; - struct extent_buffer *eb; - struct finish_unordered_dir_ctx fctx; - int slot; - - path = alloc_path_for_send(); - if (!path) { - ret = -ENOMEM; - goto out; - } - - memset(&fctx, 0, sizeof(fctx)); - fctx.sctx = sctx; - fctx.cur_path = fs_path_alloc(sctx); - fctx.dir_path = fs_path_alloc(sctx); - if (!fctx.cur_path || !fctx.dir_path) { - ret = -ENOMEM; - goto out; - } - fctx.dir_ino = dir; - - ret = get_cur_path(sctx, dir, dir_gen, fctx.dir_path); - if (ret < 0) - goto out; - - /* - * We do two passes. The first links in the new refs and the second - * deletes orphans if required. Deletion of orphans is not required for - * directory inodes, as we always have only one ref and use rename - * instead of link for those. - */ - -again: - key.objectid = dir; - key.type = BTRFS_DIR_ITEM_KEY; - key.offset = 0; - while (1) { - ret = btrfs_search_slot_for_read(sctx->send_root, &key, path, - 1, 0); - if (ret < 0) - goto out; - eb = path->nodes[0]; - slot = path->slots[0]; - btrfs_item_key_to_cpu(eb, &found_key, slot); - - if (found_key.objectid != key.objectid || - found_key.type != key.type) { - btrfs_release_path(path); - break; - } - - ret = iterate_dir_item(sctx, sctx->send_root, path, - &found_key, __finish_unordered_dir, - &fctx); - if (ret < 0) - goto out; - - key.offset = found_key.offset + 1; - btrfs_release_path(path); - } - - if (!fctx.delete_pass && fctx.need_delete) { - fctx.delete_pass = 1; - goto again; - } - -out: - btrfs_free_path(path); - fs_path_free(sctx, fctx.cur_path); - fs_path_free(sctx, fctx.dir_path); - return ret; -} - /* * This does all the move/link/unlink/rmdir magic. */ @@ -2674,6 +2725,7 @@ static int process_recorded_refs(struct send_ctx *sctx) { int ret = 0; struct recorded_ref *cur; + struct recorded_ref *cur2; struct ulist *check_dirs = NULL; struct ulist_iterator uit; struct ulist_node *un; @@ -2685,6 +2737,12 @@ static int process_recorded_refs(struct send_ctx *sctx) verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino); + /* + * This should never happen as the root dir always has the same ref + * which is always '..' + */ + BUG_ON(sctx->cur_ino <= BTRFS_FIRST_FREE_OBJECTID); + valid_path = fs_path_alloc(sctx); if (!valid_path) { ret = -ENOMEM; @@ -2731,6 +2789,46 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino); list_for_each_entry(cur, &sctx->new_refs, list) { /* + * We may have refs where the parent directory does not exist + * yet. This happens if the parent directories inum is higher + * the the current inum. To handle this case, we create the + * parent directory out of order. But we need to check if this + * did already happen before due to other refs in the same dir. + */ + ret = get_cur_inode_state(sctx, cur->dir, cur->dir_gen); + if (ret < 0) + goto out; + if (ret == inode_state_will_create) { + ret = 0; + /* + * First check if any of the current inodes refs did + * already create the dir. + */ + list_for_each_entry(cur2, &sctx->new_refs, list) { + if (cur == cur2) + break; + if (cur2->dir == cur->dir) { + ret = 1; + break; + } + } + + /* + * If that did not happen, check if a previous inode + * did already create the dir. + */ + if (!ret) + ret = did_create_dir(sctx, cur->dir); + if (ret < 0) + goto out; + if (!ret) { + ret = send_create_inode(sctx, cur->dir); + if (ret < 0) + goto out; + } + } + + /* * Check if this new ref would overwrite the first ref of * another unprocessed inode. If yes, orphanize the * overwritten inode. If we find an overwritten ref that is @@ -2764,7 +2862,7 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino); * inode, move it and update valid_path. If not, link or move * it depending on the inode mode. */ - if (is_orphan && !sctx->cur_inode_first_ref_orphan) { + if (is_orphan) { ret = send_rename(sctx, valid_path, cur->full_path); if (ret < 0) goto out; @@ -2827,6 +2925,17 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino); if (ret < 0) goto out; } + } else if (S_ISDIR(sctx->cur_inode_mode) && + !list_empty(&sctx->deleted_refs)) { + /* + * We have a moved dir. Add the old parent to check_dirs + */ + cur = list_entry(sctx->deleted_refs.next, struct recorded_ref, + list); + ret = ulist_add(check_dirs, cur->dir, cur->dir_gen, + GFP_NOFS); + if (ret < 0) + goto out; } else if (!S_ISDIR(sctx->cur_inode_mode)) { /* * We have a non dir inode. Go through all deleted refs and @@ -2840,35 +2949,9 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino); if (ret < 0) goto out; if (!ret) { - /* - * In case the inode was moved to a directory - * that was not created yet (see - * __record_new_ref), we can not unlink the ref - * as it will be needed later when the parent - * directory is created, so that we can move in - * the inode to the new dir. - */ - if (!is_orphan && - sctx->cur_inode_first_ref_orphan) { - ret = orphanize_inode(sctx, - sctx->cur_ino, - sctx->cur_inode_gen, - cur->full_path); - if (ret < 0) - goto out; - ret = gen_unique_name(sctx, - sctx->cur_ino, - sctx->cur_inode_gen, - valid_path); - if (ret < 0) - goto out; - is_orphan = 1; - - } else { - ret = send_unlink(sctx, cur->full_path); - if (ret < 0) - goto out; - } + ret = send_unlink(sctx, cur->full_path); + if (ret < 0) + goto out; } ret = ulist_add(check_dirs, cur->dir, cur->dir_gen, GFP_NOFS); @@ -2880,12 +2963,11 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino); * If the inode is still orphan, unlink the orphan. This may * happen when a previous inode did overwrite the first ref * of this inode and no new refs were added for the current - * inode. - * We can however not delete the orphan in case the inode relies - * in a directory that was not created yet (see - * __record_new_ref) + * inode. Unlinking does not mean that the inode is deleted in + * all cases. There may still be links to this inode in other + * places. */ - if (is_orphan && !sctx->cur_inode_first_ref_orphan) { + if (is_orphan) { ret = send_unlink(sctx, valid_path); if (ret < 0) goto out; @@ -2900,6 +2982,11 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino); */ ULIST_ITER_INIT(&uit); while ((un = ulist_next(check_dirs, &uit))) { + /* + * In case we had refs into dirs that were not processed yet, + * we don't need to do the utime and rmdir logic for these dirs. + * The dir will be processed later. + */ if (un->val > sctx->cur_ino) continue; @@ -2929,25 +3016,6 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino); } } - /* - * Current inode is now at it's new position, so we must increase - * send_progress - */ - sctx->send_progress = sctx->cur_ino + 1; - - /* - * We may have a directory here that has pending refs which could not - * be created before (because the dir did not exist before, see - * __record_new_ref). finish_outoforder_dir will link/move the pending - * refs. - */ - if (S_ISDIR(sctx->cur_inode_mode) && sctx->cur_inode_new) { - ret = finish_outoforder_dir(sctx, sctx->cur_ino, - sctx->cur_inode_gen); - if (ret < 0) - goto out; - } - ret = 0; out: @@ -2971,34 +3039,9 @@ static int __record_new_ref(int num, u64 dir, int index, return -ENOMEM; ret = get_inode_info(sctx->send_root, dir, NULL, &gen, NULL, NULL, - NULL); - if (ret < 0) - goto out; - - /* - * The parent may be non-existent at this point in time. This happens - * if the ino of the parent dir is higher then the current ino. In this - * case, we can not process this ref until the parent dir is finally - * created. If we reach the parent dir later, process_recorded_refs - * will go through all dir items and process the refs that could not be - * processed before. In case this is the first ref, we set - * cur_inode_first_ref_orphan to 1 to inform process_recorded_refs to - * keep an orphan of the inode so that it later can be used for - * link/move - */ - ret = is_inode_existent(sctx, dir, gen); + NULL, NULL); if (ret < 0) goto out; - if (!ret) { - ret = is_first_ref(sctx, sctx->send_root, sctx->cur_ino, dir, - name->start, fs_path_len(name)); - if (ret < 0) - goto out; - if (ret) - sctx->cur_inode_first_ref_orphan = 1; - ret = 0; - goto out; - } ret = get_cur_path(sctx, dir, gen, p); if (ret < 0) @@ -3029,7 +3072,7 @@ static int __record_deleted_ref(int num, u64 dir, int index, return -ENOMEM; ret = get_inode_info(sctx->parent_root, dir, NULL, &gen, NULL, NULL, - NULL); + NULL, NULL); if (ret < 0) goto out; @@ -3206,33 +3249,29 @@ static int process_all_refs(struct send_ctx *sctx, key.offset = 0; while (1) { ret = btrfs_search_slot_for_read(root, &key, path, 1, 0); - if (ret < 0) { - btrfs_release_path(path); + if (ret < 0) goto out; - } - if (ret) { - btrfs_release_path(path); + if (ret) break; - } eb = path->nodes[0]; slot = path->slots[0]; btrfs_item_key_to_cpu(eb, &found_key, slot); if (found_key.objectid != key.objectid || - found_key.type != key.type) { - btrfs_release_path(path); + (found_key.type != BTRFS_INODE_REF_KEY && + found_key.type != BTRFS_INODE_EXTREF_KEY)) break; - } - ret = iterate_inode_ref(sctx, sctx->parent_root, path, - &found_key, 0, cb, sctx); + ret = iterate_inode_ref(sctx, root, path, &found_key, 0, cb, + sctx); btrfs_release_path(path); if (ret < 0) goto out; key.offset = found_key.offset + 1; } + btrfs_release_path(path); ret = process_recorded_refs(sctx); @@ -3555,7 +3594,7 @@ static int send_write(struct send_ctx *sctx, u64 offset, u32 len) int ret = 0; struct fs_path *p; loff_t pos = offset; - int readed = 0; + int num_read = 0; mm_segment_t old_fs; p = fs_path_alloc(sctx); @@ -3580,8 +3619,8 @@ verbose_printk("btrfs: send_write offset=%llu, len=%d\n", offset, len); ret = vfs_read(sctx->cur_inode_filp, sctx->read_buf, len, &pos); if (ret < 0) goto out; - readed = ret; - if (!readed) + num_read = ret; + if (!num_read) goto out; ret = begin_cmd(sctx, BTRFS_SEND_C_WRITE); @@ -3594,7 +3633,7 @@ verbose_printk("btrfs: send_write offset=%llu, len=%d\n", offset, len); TLV_PUT_PATH(sctx, BTRFS_SEND_A_PATH, p); TLV_PUT_U64(sctx, BTRFS_SEND_A_FILE_OFFSET, offset); - TLV_PUT(sctx, BTRFS_SEND_A_DATA, sctx->read_buf, readed); + TLV_PUT(sctx, BTRFS_SEND_A_DATA, sctx->read_buf, num_read); ret = send_cmd(sctx); @@ -3604,7 +3643,7 @@ out: set_fs(old_fs); if (ret < 0) return ret; - return readed; + return num_read; } /* @@ -3615,7 +3654,6 @@ static int send_clone(struct send_ctx *sctx, struct clone_root *clone_root) { int ret = 0; - struct btrfs_root *clone_root2 = clone_root->root; struct fs_path *p; u64 gen; @@ -3640,22 +3678,23 @@ verbose_printk("btrfs: send_clone offset=%llu, len=%d, clone_root=%llu, " TLV_PUT_U64(sctx, BTRFS_SEND_A_CLONE_LEN, len); TLV_PUT_PATH(sctx, BTRFS_SEND_A_PATH, p); - if (clone_root2 == sctx->send_root) { + if (clone_root->root == sctx->send_root) { ret = get_inode_info(sctx->send_root, clone_root->ino, NULL, - &gen, NULL, NULL, NULL); + &gen, NULL, NULL, NULL, NULL); if (ret < 0) goto out; ret = get_cur_path(sctx, clone_root->ino, gen, p); } else { - ret = get_inode_path(sctx, clone_root2, clone_root->ino, p); + ret = get_inode_path(sctx, clone_root->root, + clone_root->ino, p); } if (ret < 0) goto out; TLV_PUT_UUID(sctx, BTRFS_SEND_A_CLONE_UUID, - clone_root2->root_item.uuid); + clone_root->root->root_item.uuid); TLV_PUT_U64(sctx, BTRFS_SEND_A_CLONE_CTRANSID, - clone_root2->root_item.ctransid); + clone_root->root->root_item.ctransid); TLV_PUT_PATH(sctx, BTRFS_SEND_A_CLONE_PATH, p); TLV_PUT_U64(sctx, BTRFS_SEND_A_CLONE_OFFSET, clone_root->offset); @@ -3684,10 +3723,17 @@ static int send_write_or_clone(struct send_ctx *sctx, ei = btrfs_item_ptr(path->nodes[0], path->slots[0], struct btrfs_file_extent_item); type = btrfs_file_extent_type(path->nodes[0], ei); - if (type == BTRFS_FILE_EXTENT_INLINE) + if (type == BTRFS_FILE_EXTENT_INLINE) { len = btrfs_file_extent_inline_len(path->nodes[0], ei); - else + /* + * it is possible the inline item won't cover the whole page, + * but there may be items after this page. Make + * sure to send the whole thing + */ + len = PAGE_CACHE_ALIGN(len); + } else { len = btrfs_file_extent_num_bytes(path->nodes[0], ei); + } if (offset + len > sctx->cur_inode_size) len = sctx->cur_inode_size - offset; @@ -3735,6 +3781,8 @@ static int is_extent_unchanged(struct send_ctx *sctx, u64 left_offset_fixed; u64 left_len; u64 right_len; + u64 left_gen; + u64 right_gen; u8 left_type; u8 right_type; @@ -3744,17 +3792,17 @@ static int is_extent_unchanged(struct send_ctx *sctx, eb = left_path->nodes[0]; slot = left_path->slots[0]; - ei = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item); left_type = btrfs_file_extent_type(eb, ei); - left_disknr = btrfs_file_extent_disk_bytenr(eb, ei); - left_len = btrfs_file_extent_num_bytes(eb, ei); - left_offset = btrfs_file_extent_offset(eb, ei); if (left_type != BTRFS_FILE_EXTENT_REG) { ret = 0; goto out; } + left_disknr = btrfs_file_extent_disk_bytenr(eb, ei); + left_len = btrfs_file_extent_num_bytes(eb, ei); + left_offset = btrfs_file_extent_offset(eb, ei); + left_gen = btrfs_file_extent_generation(eb, ei); /* * Following comments will refer to these graphics. L is the left @@ -3810,6 +3858,7 @@ static int is_extent_unchanged(struct send_ctx *sctx, right_disknr = btrfs_file_extent_disk_bytenr(eb, ei); right_len = btrfs_file_extent_num_bytes(eb, ei); right_offset = btrfs_file_extent_offset(eb, ei); + right_gen = btrfs_file_extent_generation(eb, ei); if (right_type != BTRFS_FILE_EXTENT_REG) { ret = 0; @@ -3820,7 +3869,7 @@ static int is_extent_unchanged(struct send_ctx *sctx, * Are we at extent 8? If yes, we know the extent is changed. * This may only happen on the first iteration. */ - if (found_key.offset + right_len < ekey->offset) { + if (found_key.offset + right_len <= ekey->offset) { ret = 0; goto out; } @@ -3837,8 +3886,9 @@ static int is_extent_unchanged(struct send_ctx *sctx, /* * Check if we have the same extent. */ - if (left_disknr + left_offset_fixed != - right_disknr + right_offset) { + if (left_disknr != right_disknr || + left_offset_fixed != right_offset || + left_gen != right_gen) { ret = 0; goto out; } @@ -3971,12 +4021,21 @@ static int process_recorded_refs_if_needed(struct send_ctx *sctx, int at_end) if (sctx->cur_ino == 0) goto out; if (!at_end && sctx->cur_ino == sctx->cmp_key->objectid && - sctx->cmp_key->type <= BTRFS_INODE_REF_KEY) + sctx->cmp_key->type <= BTRFS_INODE_EXTREF_KEY) goto out; if (list_empty(&sctx->new_refs) && list_empty(&sctx->deleted_refs)) goto out; ret = process_recorded_refs(sctx); + if (ret < 0) + goto out; + + /* + * We have processed the refs and thus need to advance send_progress. + * Now, calls to get_cur_xxx will take the updated refs of the current + * inode into account. + */ + sctx->send_progress = sctx->cur_ino + 1; out: return ret; @@ -4004,26 +4063,25 @@ static int finish_inode_if_needed(struct send_ctx *sctx, int at_end) goto out; ret = get_inode_info(sctx->send_root, sctx->cur_ino, NULL, NULL, - &left_mode, &left_uid, &left_gid); + &left_mode, &left_uid, &left_gid, NULL); if (ret < 0) goto out; - if (!S_ISLNK(sctx->cur_inode_mode)) { - if (!sctx->parent_root || sctx->cur_inode_new) { + if (!sctx->parent_root || sctx->cur_inode_new) { + need_chown = 1; + if (!S_ISLNK(sctx->cur_inode_mode)) need_chmod = 1; - need_chown = 1; - } else { - ret = get_inode_info(sctx->parent_root, sctx->cur_ino, - NULL, NULL, &right_mode, &right_uid, - &right_gid); - if (ret < 0) - goto out; + } else { + ret = get_inode_info(sctx->parent_root, sctx->cur_ino, + NULL, NULL, &right_mode, &right_uid, + &right_gid, NULL); + if (ret < 0) + goto out; - if (left_uid != right_uid || left_gid != right_gid) - need_chown = 1; - if (left_mode != right_mode) - need_chmod = 1; - } + if (left_uid != right_uid || left_gid != right_gid) + need_chown = 1; + if (!S_ISLNK(sctx->cur_inode_mode) && left_mode != right_mode) + need_chmod = 1; } if (S_ISREG(sctx->cur_inode_mode)) { @@ -4074,7 +4132,12 @@ static int changed_inode(struct send_ctx *sctx, sctx->cur_ino = key->objectid; sctx->cur_inode_new_gen = 0; - sctx->cur_inode_first_ref_orphan = 0; + + /* + * Set send_progress to current inode. This will tell all get_cur_xxx + * functions that the current inode's refs are not updated yet. Later, + * when process_recorded_refs is finished, it is set to cur_ino + 1. + */ sctx->send_progress = sctx->cur_ino; if (result == BTRFS_COMPARE_TREE_NEW || @@ -4098,7 +4161,14 @@ static int changed_inode(struct send_ctx *sctx, right_gen = btrfs_inode_generation(sctx->right_path->nodes[0], right_ii); - if (left_gen != right_gen) + + /* + * The cur_ino = root dir case is special here. We can't treat + * the inode as deleted+reused because it would generate a + * stream that tries to delete/mkdir the root dir. + */ + if (left_gen != right_gen && + sctx->cur_ino != BTRFS_FIRST_FREE_OBJECTID) sctx->cur_inode_new_gen = 1; } @@ -4111,8 +4181,7 @@ static int changed_inode(struct send_ctx *sctx, sctx->cur_inode_mode = btrfs_inode_mode( sctx->left_path->nodes[0], left_ii); if (sctx->cur_ino != BTRFS_FIRST_FREE_OBJECTID) - ret = send_create_inode(sctx, sctx->left_path, - sctx->cmp_key); + ret = send_create_inode_if_needed(sctx); } else if (result == BTRFS_COMPARE_TREE_DELETED) { sctx->cur_inode_gen = right_gen; sctx->cur_inode_new = 0; @@ -4122,7 +4191,17 @@ static int changed_inode(struct send_ctx *sctx, sctx->cur_inode_mode = btrfs_inode_mode( sctx->right_path->nodes[0], right_ii); } else if (result == BTRFS_COMPARE_TREE_CHANGED) { + /* + * We need to do some special handling in case the inode was + * reported as changed with a changed generation number. This + * means that the original inode was deleted and new inode + * reused the same inum. So we have to treat the old inode as + * deleted and the new one as new. + */ if (sctx->cur_inode_new_gen) { + /* + * First, process the inode as if it was deleted. + */ sctx->cur_inode_gen = right_gen; sctx->cur_inode_new = 0; sctx->cur_inode_deleted = 1; @@ -4135,6 +4214,9 @@ static int changed_inode(struct send_ctx *sctx, if (ret < 0) goto out; + /* + * Now process the inode as if it was new. + */ sctx->cur_inode_gen = left_gen; sctx->cur_inode_new = 1; sctx->cur_inode_deleted = 0; @@ -4142,14 +4224,23 @@ static int changed_inode(struct send_ctx *sctx, sctx->left_path->nodes[0], left_ii); sctx->cur_inode_mode = btrfs_inode_mode( sctx->left_path->nodes[0], left_ii); - ret = send_create_inode(sctx, sctx->left_path, - sctx->cmp_key); + ret = send_create_inode_if_needed(sctx); if (ret < 0) goto out; ret = process_all_refs(sctx, BTRFS_COMPARE_TREE_NEW); if (ret < 0) goto out; + /* + * Advance send_progress now as we did not get into + * process_recorded_refs_if_needed in the new_gen case. + */ + sctx->send_progress = sctx->cur_ino + 1; + + /* + * Now process all extents and xattrs of the inode as if + * they were all new. + */ ret = process_all_extents(sctx); if (ret < 0) goto out; @@ -4172,6 +4263,16 @@ out: return ret; } +/* + * We have to process new refs before deleted refs, but compare_trees gives us + * the new and deleted refs mixed. To fix this, we record the new/deleted refs + * first and later process them in process_recorded_refs. + * For the cur_inode_new_gen case, we skip recording completely because + * changed_inode did already initiate processing of refs. The reason for this is + * that in this case, compare_tree actually compares the refs of 2 different + * inodes. To fix this, process_all_refs is used in changed_inode to handle all + * refs of the right tree as deleted and all refs of the left tree as new. + */ static int changed_ref(struct send_ctx *sctx, enum btrfs_compare_tree_result result) { @@ -4192,6 +4293,11 @@ static int changed_ref(struct send_ctx *sctx, return ret; } +/* + * Process new/deleted/changed xattrs. We skip processing in the + * cur_inode_new_gen case because changed_inode did already initiate processing + * of xattrs. The reason is the same as in changed_ref + */ static int changed_xattr(struct send_ctx *sctx, enum btrfs_compare_tree_result result) { @@ -4211,6 +4317,11 @@ static int changed_xattr(struct send_ctx *sctx, return ret; } +/* + * Process new/deleted/changed extents. We skip processing in the + * cur_inode_new_gen case because changed_inode did already initiate processing + * of extents. The reason is the same as in changed_ref + */ static int changed_extent(struct send_ctx *sctx, enum btrfs_compare_tree_result result) { @@ -4227,7 +4338,10 @@ static int changed_extent(struct send_ctx *sctx, return ret; } - +/* + * Updates compare related fields in sctx and simply forwards to the actual + * changed_xxx functions. + */ static int changed_cb(struct btrfs_root *left_root, struct btrfs_root *right_root, struct btrfs_path *left_path, @@ -4247,9 +4361,15 @@ static int changed_cb(struct btrfs_root *left_root, if (ret < 0) goto out; + /* Ignore non-FS objects */ + if (key->objectid == BTRFS_FREE_INO_OBJECTID || + key->objectid == BTRFS_FREE_SPACE_OBJECTID) + goto out; + if (key->type == BTRFS_INODE_ITEM_KEY) ret = changed_inode(sctx, result); - else if (key->type == BTRFS_INODE_REF_KEY) + else if (key->type == BTRFS_INODE_REF_KEY || + key->type == BTRFS_INODE_EXTREF_KEY) ret = changed_ref(sctx, result); else if (key->type == BTRFS_XATTR_ITEM_KEY) ret = changed_xattr(sctx, result); @@ -4299,7 +4419,8 @@ join_trans: } /* - * Make sure the tree has not changed + * Make sure the tree has not changed after re-joining. We detect this + * by comparing start_ctransid and ctransid. They should always match. */ spin_lock(&send_root->root_times_lock); ctransid = btrfs_root_ctransid(&send_root->root_item); diff --git a/fs/btrfs/send.h b/fs/btrfs/send.h index 9934e948e57f..1bf4f32fd4ef 100644 --- a/fs/btrfs/send.h +++ b/fs/btrfs/send.h @@ -130,4 +130,5 @@ enum { #ifdef __KERNEL__ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg); +int write_buf(struct file *filp, const void *buf, u32 len, loff_t *off); #endif diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 83d6f9f9c220..915ac14c2064 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -243,12 +243,18 @@ void __btrfs_abort_transaction(struct btrfs_trans_handle *trans, struct btrfs_root *root, const char *function, unsigned int line, int errno) { - WARN_ONCE(1, KERN_DEBUG "btrfs: Transaction aborted"); + WARN_ONCE(1, KERN_DEBUG "btrfs: Transaction aborted\n"); trans->aborted = errno; /* Nothing used. The other threads that have joined this * transaction may be able to continue. */ if (!trans->blocks_used) { - btrfs_printk(root->fs_info, "Aborting unused transaction.\n"); + char nbuf[16]; + const char *errstr; + + errstr = btrfs_decode_error(root->fs_info, errno, nbuf); + btrfs_printk(root->fs_info, + "%s:%d: Aborting unused transaction(%s).\n", + function, line, errstr); return; } trans->transaction->aborted = errno; @@ -407,7 +413,15 @@ int btrfs_parse_options(struct btrfs_root *root, char *options) btrfs_set_opt(info->mount_opt, NODATASUM); break; case Opt_nodatacow: - printk(KERN_INFO "btrfs: setting nodatacow\n"); + if (!btrfs_test_opt(root, COMPRESS) || + !btrfs_test_opt(root, FORCE_COMPRESS)) { + printk(KERN_INFO "btrfs: setting nodatacow, compression disabled\n"); + } else { + printk(KERN_INFO "btrfs: setting nodatacow\n"); + } + info->compress_type = BTRFS_COMPRESS_NONE; + btrfs_clear_opt(info->mount_opt, COMPRESS); + btrfs_clear_opt(info->mount_opt, FORCE_COMPRESS); btrfs_set_opt(info->mount_opt, NODATACOW); btrfs_set_opt(info->mount_opt, NODATASUM); break; @@ -422,10 +436,14 @@ int btrfs_parse_options(struct btrfs_root *root, char *options) compress_type = "zlib"; info->compress_type = BTRFS_COMPRESS_ZLIB; btrfs_set_opt(info->mount_opt, COMPRESS); + btrfs_clear_opt(info->mount_opt, NODATACOW); + btrfs_clear_opt(info->mount_opt, NODATASUM); } else if (strcmp(args[0].from, "lzo") == 0) { compress_type = "lzo"; info->compress_type = BTRFS_COMPRESS_LZO; btrfs_set_opt(info->mount_opt, COMPRESS); + btrfs_clear_opt(info->mount_opt, NODATACOW); + btrfs_clear_opt(info->mount_opt, NODATASUM); btrfs_set_fs_incompat(info, COMPRESS_LZO); } else if (strncmp(args[0].from, "no", 2) == 0) { compress_type = "no"; @@ -543,11 +561,11 @@ int btrfs_parse_options(struct btrfs_root *root, char *options) btrfs_set_opt(info->mount_opt, ENOSPC_DEBUG); break; case Opt_defrag: - printk(KERN_INFO "btrfs: enabling auto defrag"); + printk(KERN_INFO "btrfs: enabling auto defrag\n"); btrfs_set_opt(info->mount_opt, AUTO_DEFRAG); break; case Opt_recovery: - printk(KERN_INFO "btrfs: enabling auto recovery"); + printk(KERN_INFO "btrfs: enabling auto recovery\n"); btrfs_set_opt(info->mount_opt, RECOVERY); break; case Opt_skip_balance: @@ -846,18 +864,15 @@ int btrfs_sync_fs(struct super_block *sb, int wait) return 0; } - btrfs_wait_ordered_extents(root, 0, 0); - - spin_lock(&fs_info->trans_lock); - if (!fs_info->running_transaction) { - spin_unlock(&fs_info->trans_lock); - return 0; - } - spin_unlock(&fs_info->trans_lock); + btrfs_wait_ordered_extents(root, 0); - trans = btrfs_join_transaction(root); - if (IS_ERR(trans)) + trans = btrfs_attach_transaction(root); + if (IS_ERR(trans)) { + /* no transaction, don't bother */ + if (PTR_ERR(trans) == -ENOENT) + return 0; return PTR_ERR(trans); + } return btrfs_commit_transaction(trans, root); } @@ -1508,17 +1523,21 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd, static int btrfs_freeze(struct super_block *sb) { - struct btrfs_fs_info *fs_info = btrfs_sb(sb); - mutex_lock(&fs_info->transaction_kthread_mutex); - mutex_lock(&fs_info->cleaner_mutex); - return 0; + struct btrfs_trans_handle *trans; + struct btrfs_root *root = btrfs_sb(sb)->tree_root; + + trans = btrfs_attach_transaction(root); + if (IS_ERR(trans)) { + /* no transaction, don't bother */ + if (PTR_ERR(trans) == -ENOENT) + return 0; + return PTR_ERR(trans); + } + return btrfs_commit_transaction(trans, root); } static int btrfs_unfreeze(struct super_block *sb) { - struct btrfs_fs_info *fs_info = btrfs_sb(sb); - mutex_unlock(&fs_info->cleaner_mutex); - mutex_unlock(&fs_info->transaction_kthread_mutex); return 0; } @@ -1595,7 +1614,7 @@ static int btrfs_interface_init(void) static void btrfs_interface_exit(void) { if (misc_deregister(&btrfs_misc) < 0) - printk(KERN_INFO "misc_deregister failed for control device"); + printk(KERN_INFO "btrfs: misc_deregister failed for control device\n"); } static int __init init_btrfs_fs(void) @@ -1620,10 +1639,14 @@ static int __init init_btrfs_fs(void) if (err) goto free_extent_io; - err = btrfs_delayed_inode_init(); + err = ordered_data_init(); if (err) goto free_extent_map; + err = btrfs_delayed_inode_init(); + if (err) + goto free_ordered_data; + err = btrfs_interface_init(); if (err) goto free_delayed_inode; @@ -1641,6 +1664,8 @@ unregister_ioctl: btrfs_interface_exit(); free_delayed_inode: btrfs_delayed_inode_exit(); +free_ordered_data: + ordered_data_exit(); free_extent_map: extent_map_exit(); free_extent_io: @@ -1657,6 +1682,7 @@ static void __exit exit_btrfs_fs(void) { btrfs_destroy_cachep(); btrfs_delayed_inode_exit(); + ordered_data_exit(); extent_map_exit(); extent_io_exit(); btrfs_interface_exit(); diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 27c26004e050..04bbfb1052eb 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -53,7 +53,7 @@ static noinline void switch_commit_root(struct btrfs_root *root) /* * either allocate a new transaction or hop into the existing one */ -static noinline int join_transaction(struct btrfs_root *root, int nofail) +static noinline int join_transaction(struct btrfs_root *root, int type) { struct btrfs_transaction *cur_trans; struct btrfs_fs_info *fs_info = root->fs_info; @@ -67,7 +67,13 @@ loop: } if (fs_info->trans_no_join) { - if (!nofail) { + /* + * If we are JOIN_NOLOCK we're already committing a current + * transaction, we just need a handle to deal with something + * when committing the transaction, such as inode cache and + * space cache. It is a special case. + */ + if (type != TRANS_JOIN_NOLOCK) { spin_unlock(&fs_info->trans_lock); return -EBUSY; } @@ -87,6 +93,13 @@ loop: } spin_unlock(&fs_info->trans_lock); + /* + * If we are ATTACH, we just want to catch the current transaction, + * and commit it. If there is no transaction, just return ENOENT. + */ + if (type == TRANS_ATTACH) + return -ENOENT; + cur_trans = kmem_cache_alloc(btrfs_transaction_cachep, GFP_NOFS); if (!cur_trans) return -ENOMEM; @@ -267,13 +280,6 @@ static void wait_current_trans(struct btrfs_root *root) } } -enum btrfs_trans_type { - TRANS_START, - TRANS_JOIN, - TRANS_USERSPACE, - TRANS_JOIN_NOLOCK, -}; - static int may_wait_transaction(struct btrfs_root *root, int type) { if (root->fs_info->log_root_recovering) @@ -290,7 +296,8 @@ static int may_wait_transaction(struct btrfs_root *root, int type) } static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root, - u64 num_items, int type) + u64 num_items, int type, + int noflush) { struct btrfs_trans_handle *h; struct btrfs_transaction *cur_trans; @@ -324,9 +331,14 @@ static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root, } num_bytes = btrfs_calc_trans_metadata_size(root, num_items); - ret = btrfs_block_rsv_add(root, - &root->fs_info->trans_block_rsv, - num_bytes); + if (noflush) + ret = btrfs_block_rsv_add_noflush(root, + &root->fs_info->trans_block_rsv, + num_bytes); + else + ret = btrfs_block_rsv_add(root, + &root->fs_info->trans_block_rsv, + num_bytes); if (ret) return ERR_PTR(ret); } @@ -335,19 +347,34 @@ again: if (!h) return ERR_PTR(-ENOMEM); - sb_start_intwrite(root->fs_info->sb); + /* + * If we are JOIN_NOLOCK we're already committing a transaction and + * waiting on this guy, so we don't need to do the sb_start_intwrite + * because we're already holding a ref. We need this because we could + * have raced in and did an fsync() on a file which can kick a commit + * and then we deadlock with somebody doing a freeze. + * + * If we are ATTACH, it means we just want to catch the current + * transaction and commit it, so we needn't do sb_start_intwrite(). + */ + if (type < TRANS_JOIN_NOLOCK) + sb_start_intwrite(root->fs_info->sb); if (may_wait_transaction(root, type)) wait_current_trans(root); do { - ret = join_transaction(root, type == TRANS_JOIN_NOLOCK); + ret = join_transaction(root, type); if (ret == -EBUSY) wait_current_trans(root); } while (ret == -EBUSY); if (ret < 0) { - sb_end_intwrite(root->fs_info->sb); + /* We must get the transaction if we are JOIN_NOLOCK. */ + BUG_ON(type == TRANS_JOIN_NOLOCK); + + if (type < TRANS_JOIN_NOLOCK) + sb_end_intwrite(root->fs_info->sb); kmem_cache_free(btrfs_trans_handle_cachep, h); return ERR_PTR(ret); } @@ -367,7 +394,9 @@ again: h->aborted = 0; h->qgroup_reserved = qgroup_reserved; h->delayed_ref_elem.seq = 0; + h->type = type; INIT_LIST_HEAD(&h->qgroup_ref_list); + INIT_LIST_HEAD(&h->new_bgs); smp_mb(); if (cur_trans->blocked && may_wait_transaction(root, type)) { @@ -393,21 +422,33 @@ got_it: struct btrfs_trans_handle *btrfs_start_transaction(struct btrfs_root *root, int num_items) { - return start_transaction(root, num_items, TRANS_START); + return start_transaction(root, num_items, TRANS_START, 0); +} + +struct btrfs_trans_handle *btrfs_start_transaction_noflush( + struct btrfs_root *root, int num_items) +{ + return start_transaction(root, num_items, TRANS_START, 1); } + struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root) { - return start_transaction(root, 0, TRANS_JOIN); + return start_transaction(root, 0, TRANS_JOIN, 0); } struct btrfs_trans_handle *btrfs_join_transaction_nolock(struct btrfs_root *root) { - return start_transaction(root, 0, TRANS_JOIN_NOLOCK); + return start_transaction(root, 0, TRANS_JOIN_NOLOCK, 0); } struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *root) { - return start_transaction(root, 0, TRANS_USERSPACE); + return start_transaction(root, 0, TRANS_USERSPACE, 0); +} + +struct btrfs_trans_handle *btrfs_attach_transaction(struct btrfs_root *root) +{ + return start_transaction(root, 0, TRANS_ATTACH, 0); } /* wait for a transaction commit to be fully complete */ @@ -506,11 +547,12 @@ int btrfs_should_end_transaction(struct btrfs_trans_handle *trans, } static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, - struct btrfs_root *root, int throttle, int lock) + struct btrfs_root *root, int throttle) { struct btrfs_transaction *cur_trans = trans->transaction; struct btrfs_fs_info *info = root->fs_info; int count = 0; + int lock = (trans->type != TRANS_JOIN_NOLOCK); int err = 0; if (--trans->use_count) { @@ -536,6 +578,9 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, trans->qgroup_reserved = 0; } + if (!list_empty(&trans->new_bgs)) + btrfs_create_pending_block_groups(trans, root); + while (count < 2) { unsigned long cur = trans->delayed_ref_updates; trans->delayed_ref_updates = 0; @@ -551,7 +596,8 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, btrfs_trans_release_metadata(trans, root); trans->block_rsv = NULL; - sb_end_intwrite(root->fs_info->sb); + if (!list_empty(&trans->new_bgs)) + btrfs_create_pending_block_groups(trans, root); if (lock && !atomic_read(&root->fs_info->open_ioctl_trans) && should_end_transaction(trans, root)) { @@ -573,6 +619,9 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, } } + if (trans->type < TRANS_JOIN_NOLOCK) + sb_end_intwrite(root->fs_info->sb); + WARN_ON(cur_trans != info->running_transaction); WARN_ON(atomic_read(&cur_trans->num_writers) < 1); atomic_dec(&cur_trans->num_writers); @@ -604,7 +653,7 @@ int btrfs_end_transaction(struct btrfs_trans_handle *trans, { int ret; - ret = __btrfs_end_transaction(trans, root, 0, 1); + ret = __btrfs_end_transaction(trans, root, 0); if (ret) return ret; return 0; @@ -615,18 +664,7 @@ int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans, { int ret; - ret = __btrfs_end_transaction(trans, root, 1, 1); - if (ret) - return ret; - return 0; -} - -int btrfs_end_transaction_nolock(struct btrfs_trans_handle *trans, - struct btrfs_root *root) -{ - int ret; - - ret = __btrfs_end_transaction(trans, root, 0, 0); + ret = __btrfs_end_transaction(trans, root, 1); if (ret) return ret; return 0; @@ -635,7 +673,7 @@ int btrfs_end_transaction_nolock(struct btrfs_trans_handle *trans, int btrfs_end_transaction_dmeta(struct btrfs_trans_handle *trans, struct btrfs_root *root) { - return __btrfs_end_transaction(trans, root, 1, 1); + return __btrfs_end_transaction(trans, root, 1); } /* @@ -649,13 +687,15 @@ int btrfs_write_marked_extents(struct btrfs_root *root, int err = 0; int werr = 0; struct address_space *mapping = root->fs_info->btree_inode->i_mapping; + struct extent_state *cached_state = NULL; u64 start = 0; u64 end; while (!find_first_extent_bit(dirty_pages, start, &start, &end, - mark)) { - convert_extent_bit(dirty_pages, start, end, EXTENT_NEED_WAIT, mark, - GFP_NOFS); + mark, &cached_state)) { + convert_extent_bit(dirty_pages, start, end, EXTENT_NEED_WAIT, + mark, &cached_state, GFP_NOFS); + cached_state = NULL; err = filemap_fdatawrite_range(mapping, start, end); if (err) werr = err; @@ -679,12 +719,14 @@ int btrfs_wait_marked_extents(struct btrfs_root *root, int err = 0; int werr = 0; struct address_space *mapping = root->fs_info->btree_inode->i_mapping; + struct extent_state *cached_state = NULL; u64 start = 0; u64 end; while (!find_first_extent_bit(dirty_pages, start, &start, &end, - EXTENT_NEED_WAIT)) { - clear_extent_bits(dirty_pages, start, end, EXTENT_NEED_WAIT, GFP_NOFS); + EXTENT_NEED_WAIT, &cached_state)) { + clear_extent_bit(dirty_pages, start, end, EXTENT_NEED_WAIT, + 0, 0, &cached_state, GFP_NOFS); err = filemap_fdatawait_range(mapping, start, end); if (err) werr = err; @@ -955,6 +997,8 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, struct btrfs_root *parent_root; struct btrfs_block_rsv *rsv; struct inode *parent_inode; + struct btrfs_path *path; + struct btrfs_dir_item *dir_item; struct dentry *parent; struct dentry *dentry; struct extent_buffer *tmp; @@ -967,18 +1011,22 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, u64 root_flags; uuid_le new_uuid; - rsv = trans->block_rsv; + path = btrfs_alloc_path(); + if (!path) { + ret = pending->error = -ENOMEM; + goto path_alloc_fail; + } new_root_item = kmalloc(sizeof(*new_root_item), GFP_NOFS); if (!new_root_item) { ret = pending->error = -ENOMEM; - goto fail; + goto root_item_alloc_fail; } ret = btrfs_find_free_objectid(tree_root, &objectid); if (ret) { pending->error = ret; - goto fail; + goto no_free_objectid; } btrfs_reloc_pre_snapshot(trans, pending, &to_reserve); @@ -988,22 +1036,22 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, to_reserve); if (ret) { pending->error = ret; - goto fail; + goto no_free_objectid; } } ret = btrfs_qgroup_inherit(trans, fs_info, root->root_key.objectid, objectid, pending->inherit); - kfree(pending->inherit); if (ret) { pending->error = ret; - goto fail; + goto no_free_objectid; } key.objectid = objectid; key.offset = (u64)-1; key.type = BTRFS_ROOT_ITEM_KEY; + rsv = trans->block_rsv; trans->block_rsv = &pending->block_rsv; dentry = pending->dentry; @@ -1017,24 +1065,21 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, */ ret = btrfs_set_inode_index(parent_inode, &index); BUG_ON(ret); /* -ENOMEM */ - ret = btrfs_insert_dir_item(trans, parent_root, - dentry->d_name.name, dentry->d_name.len, - parent_inode, &key, - BTRFS_FT_DIR, index); - if (ret == -EEXIST) { + + /* check if there is a file/dir which has the same name. */ + dir_item = btrfs_lookup_dir_item(NULL, parent_root, path, + btrfs_ino(parent_inode), + dentry->d_name.name, + dentry->d_name.len, 0); + if (dir_item != NULL && !IS_ERR(dir_item)) { pending->error = -EEXIST; - dput(parent); goto fail; - } else if (ret) { - goto abort_trans_dput; + } else if (IS_ERR(dir_item)) { + ret = PTR_ERR(dir_item); + btrfs_abort_transaction(trans, root, ret); + goto fail; } - - btrfs_i_size_write(parent_inode, parent_inode->i_size + - dentry->d_name.len * 2); - parent_inode->i_mtime = parent_inode->i_ctime = CURRENT_TIME; - ret = btrfs_update_inode(trans, parent_root, parent_inode); - if (ret) - goto abort_trans_dput; + btrfs_release_path(path); /* * pull in the delayed directory update @@ -1043,8 +1088,8 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, * snapshot */ ret = btrfs_run_delayed_items(trans, root); - if (ret) { /* Transaction aborted */ - dput(parent); + if (ret) { /* Transaction aborted */ + btrfs_abort_transaction(trans, root, ret); goto fail; } @@ -1079,7 +1124,8 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, if (ret) { btrfs_tree_unlock(old); free_extent_buffer(old); - goto abort_trans_dput; + btrfs_abort_transaction(trans, root, ret); + goto fail; } btrfs_set_lock_blocking(old); @@ -1088,8 +1134,10 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, /* clean up in any case */ btrfs_tree_unlock(old); free_extent_buffer(old); - if (ret) - goto abort_trans_dput; + if (ret) { + btrfs_abort_transaction(trans, root, ret); + goto fail; + } /* see comments in should_cow_block() */ root->force_cow = 1; @@ -1101,8 +1149,10 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, ret = btrfs_insert_root(trans, tree_root, &key, new_root_item); btrfs_tree_unlock(tmp); free_extent_buffer(tmp); - if (ret) - goto abort_trans_dput; + if (ret) { + btrfs_abort_transaction(trans, root, ret); + goto fail; + } /* * insert root back/forward references @@ -1111,32 +1161,58 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, parent_root->root_key.objectid, btrfs_ino(parent_inode), index, dentry->d_name.name, dentry->d_name.len); - dput(parent); - if (ret) + if (ret) { + btrfs_abort_transaction(trans, root, ret); goto fail; + } key.offset = (u64)-1; pending->snap = btrfs_read_fs_root_no_name(root->fs_info, &key); if (IS_ERR(pending->snap)) { ret = PTR_ERR(pending->snap); - goto abort_trans; + btrfs_abort_transaction(trans, root, ret); + goto fail; } ret = btrfs_reloc_post_snapshot(trans, pending); + if (ret) { + btrfs_abort_transaction(trans, root, ret); + goto fail; + } + + ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1); + if (ret) { + btrfs_abort_transaction(trans, root, ret); + goto fail; + } + + ret = btrfs_insert_dir_item(trans, parent_root, + dentry->d_name.name, dentry->d_name.len, + parent_inode, &key, + BTRFS_FT_DIR, index); + /* We have check then name at the beginning, so it is impossible. */ + BUG_ON(ret == -EEXIST); + if (ret) { + btrfs_abort_transaction(trans, root, ret); + goto fail; + } + + btrfs_i_size_write(parent_inode, parent_inode->i_size + + dentry->d_name.len * 2); + parent_inode->i_mtime = parent_inode->i_ctime = CURRENT_TIME; + ret = btrfs_update_inode_fallback(trans, parent_root, parent_inode); if (ret) - goto abort_trans; - ret = 0; + btrfs_abort_transaction(trans, root, ret); fail: - kfree(new_root_item); + dput(parent); trans->block_rsv = rsv; +no_free_objectid: + kfree(new_root_item); +root_item_alloc_fail: + btrfs_free_path(path); +path_alloc_fail: btrfs_block_rsv_release(root, &pending->block_rsv, (u64)-1); return ret; - -abort_trans_dput: - dput(parent); -abort_trans: - btrfs_abort_transaction(trans, root, ret); - goto fail; } /* @@ -1229,6 +1305,16 @@ static void do_async_commit(struct work_struct *work) struct btrfs_async_commit *ac = container_of(work, struct btrfs_async_commit, work.work); + /* + * We've got freeze protection passed with the transaction. + * Tell lockdep about it. + */ + rwsem_acquire_read( + &ac->root->fs_info->sb->s_writers.lock_map[SB_FREEZE_FS-1], + 0, 1, _THIS_IP_); + + current->journal_info = ac->newtrans; + btrfs_commit_transaction(ac->newtrans, ac->root); kfree(ac); } @@ -1258,6 +1344,14 @@ int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans, atomic_inc(&cur_trans->use_count); btrfs_end_transaction(trans, root); + + /* + * Tell lockdep we've released the freeze rwsem, since the + * async commit thread will be the one to unlock it. + */ + rwsem_release(&root->fs_info->sb->s_writers.lock_map[SB_FREEZE_FS-1], + 1, _THIS_IP_); + schedule_delayed_work(&ac->work, 0); /* wait for transaction to start and unblock */ @@ -1348,6 +1442,9 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, */ cur_trans->delayed_refs.flushing = 1; + if (!list_empty(&trans->new_bgs)) + btrfs_create_pending_block_groups(trans, root); + ret = btrfs_run_delayed_refs(trans, root, 0); if (ret) goto cleanup_transaction; @@ -1403,7 +1500,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, if (flush_on_commit || snap_pending) { btrfs_start_delalloc_inodes(root, 1); - btrfs_wait_ordered_extents(root, 0, 1); + btrfs_wait_ordered_extents(root, 1); } ret = btrfs_run_delayed_items(trans, root); @@ -1456,13 +1553,28 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, */ mutex_lock(&root->fs_info->reloc_mutex); - ret = btrfs_run_delayed_items(trans, root); + /* + * We needn't worry about the delayed items because we will + * deal with them in create_pending_snapshot(), which is the + * core function of the snapshot creation. + */ + ret = create_pending_snapshots(trans, root->fs_info); if (ret) { mutex_unlock(&root->fs_info->reloc_mutex); goto cleanup_transaction; } - ret = create_pending_snapshots(trans, root->fs_info); + /* + * We insert the dir indexes of the snapshots and update the inode + * of the snapshots' parents after the snapshot creation, so there + * are some delayed items which are not dealt with. Now deal with + * them. + * + * We needn't worry that this operation will corrupt the snapshots, + * because all the tree which are snapshoted will be forced to COW + * the nodes and leaves. + */ + ret = btrfs_run_delayed_items(trans, root); if (ret) { mutex_unlock(&root->fs_info->reloc_mutex); goto cleanup_transaction; @@ -1584,7 +1696,8 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, put_transaction(cur_trans); put_transaction(cur_trans); - sb_end_intwrite(root->fs_info->sb); + if (trans->type < TRANS_JOIN_NOLOCK) + sb_end_intwrite(root->fs_info->sb); trace_btrfs_transaction_commit(root); diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h index e8b8416c688b..80961947a6b2 100644 --- a/fs/btrfs/transaction.h +++ b/fs/btrfs/transaction.h @@ -47,6 +47,14 @@ struct btrfs_transaction { int aborted; }; +enum btrfs_trans_type { + TRANS_START, + TRANS_JOIN, + TRANS_USERSPACE, + TRANS_JOIN_NOLOCK, + TRANS_ATTACH, +}; + struct btrfs_trans_handle { u64 transid; u64 bytes_reserved; @@ -58,8 +66,9 @@ struct btrfs_trans_handle { struct btrfs_transaction *transaction; struct btrfs_block_rsv *block_rsv; struct btrfs_block_rsv *orig_rsv; - int aborted; - int adding_csums; + short aborted; + short adding_csums; + enum btrfs_trans_type type; /* * this root is only needed to validate that the root passed to * start_transaction is the same as the one passed to end_transaction. @@ -68,6 +77,7 @@ struct btrfs_trans_handle { struct btrfs_root *root; struct seq_list delayed_ref_elem; struct list_head qgroup_ref_list; + struct list_head new_bgs; }; struct btrfs_pending_snapshot { @@ -88,16 +98,18 @@ static inline void btrfs_set_inode_last_trans(struct btrfs_trans_handle *trans, { BTRFS_I(inode)->last_trans = trans->transaction->transid; BTRFS_I(inode)->last_sub_trans = BTRFS_I(inode)->root->log_transid; + BTRFS_I(inode)->last_log_commit = BTRFS_I(inode)->root->last_log_commit; } int btrfs_end_transaction(struct btrfs_trans_handle *trans, struct btrfs_root *root); -int btrfs_end_transaction_nolock(struct btrfs_trans_handle *trans, - struct btrfs_root *root); struct btrfs_trans_handle *btrfs_start_transaction(struct btrfs_root *root, int num_items); +struct btrfs_trans_handle *btrfs_start_transaction_noflush( + struct btrfs_root *root, int num_items); struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root); struct btrfs_trans_handle *btrfs_join_transaction_nolock(struct btrfs_root *root); +struct btrfs_trans_handle *btrfs_attach_transaction(struct btrfs_root *root); struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *root); int btrfs_wait_for_commit(struct btrfs_root *root, u64 transid); int btrfs_write_and_wait_transaction(struct btrfs_trans_handle *trans, diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index c86670f4f285..81e407d9677a 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -18,13 +18,16 @@ #include <linux/sched.h> #include <linux/slab.h> +#include <linux/list_sort.h> #include "ctree.h" #include "transaction.h" #include "disk-io.h" #include "locking.h" #include "print-tree.h" +#include "backref.h" #include "compat.h" #include "tree-log.h" +#include "hash.h" /* magic values for the inode_only field in btrfs_log_inode: * @@ -146,7 +149,7 @@ static int start_log_trans(struct btrfs_trans_handle *trans, root->log_multiple_pids = true; } - root->log_batch++; + atomic_inc(&root->log_batch); atomic_inc(&root->log_writers); mutex_unlock(&root->log_mutex); return 0; @@ -165,7 +168,7 @@ static int start_log_trans(struct btrfs_trans_handle *trans, err = ret; } mutex_unlock(&root->fs_info->tree_log_mutex); - root->log_batch++; + atomic_inc(&root->log_batch); atomic_inc(&root->log_writers); mutex_unlock(&root->log_mutex); return err; @@ -484,7 +487,6 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, int found_type; u64 mask = root->sectorsize - 1; u64 extent_end; - u64 alloc_hint; u64 start = key->offset; u64 saved_nbytes; struct btrfs_file_extent_item *item; @@ -550,8 +552,7 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, saved_nbytes = inode_get_bytes(inode); /* drop any overlapping extents */ - ret = btrfs_drop_extents(trans, inode, start, extent_end, - &alloc_hint, 1); + ret = btrfs_drop_extents(trans, root, inode, start, extent_end, 1); BUG_ON(ret); if (found_type == BTRFS_FILE_EXTENT_REG || @@ -744,6 +745,7 @@ out: */ static noinline int backref_in_log(struct btrfs_root *log, struct btrfs_key *key, + u64 ref_objectid, char *name, int namelen) { struct btrfs_path *path; @@ -764,8 +766,17 @@ static noinline int backref_in_log(struct btrfs_root *log, if (ret != 0) goto out; - item_size = btrfs_item_size_nr(path->nodes[0], path->slots[0]); ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]); + + if (key->type == BTRFS_INODE_EXTREF_KEY) { + if (btrfs_find_name_in_ext_backref(path, ref_objectid, + name, namelen, NULL)) + match = 1; + + goto out; + } + + item_size = btrfs_item_size_nr(path->nodes[0], path->slots[0]); ptr_end = ptr + item_size; while (ptr < ptr_end) { ref = (struct btrfs_inode_ref *)ptr; @@ -786,91 +797,42 @@ out: return match; } - -/* - * replay one inode back reference item found in the log tree. - * eb, slot and key refer to the buffer and key found in the log tree. - * root is the destination we are replaying into, and path is for temp - * use by this function. (it should be released on return). - */ -static noinline int add_inode_ref(struct btrfs_trans_handle *trans, +static inline int __add_inode_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root, - struct btrfs_root *log, struct btrfs_path *path, - struct extent_buffer *eb, int slot, - struct btrfs_key *key) + struct btrfs_root *log_root, + struct inode *dir, struct inode *inode, + struct extent_buffer *eb, + u64 inode_objectid, u64 parent_objectid, + u64 ref_index, char *name, int namelen, + int *search_done) { - struct btrfs_inode_ref *ref; - struct btrfs_dir_item *di; - struct inode *dir; - struct inode *inode; - unsigned long ref_ptr; - unsigned long ref_end; - char *name; - int namelen; int ret; - int search_done = 0; - - /* - * it is possible that we didn't log all the parent directories - * for a given inode. If we don't find the dir, just don't - * copy the back ref in. The link count fixup code will take - * care of the rest - */ - dir = read_one_inode(root, key->offset); - if (!dir) - return -ENOENT; - - inode = read_one_inode(root, key->objectid); - if (!inode) { - iput(dir); - return -EIO; - } - - ref_ptr = btrfs_item_ptr_offset(eb, slot); - ref_end = ref_ptr + btrfs_item_size_nr(eb, slot); + char *victim_name; + int victim_name_len; + struct extent_buffer *leaf; + struct btrfs_dir_item *di; + struct btrfs_key search_key; + struct btrfs_inode_extref *extref; again: - ref = (struct btrfs_inode_ref *)ref_ptr; - - namelen = btrfs_inode_ref_name_len(eb, ref); - name = kmalloc(namelen, GFP_NOFS); - BUG_ON(!name); - - read_extent_buffer(eb, name, (unsigned long)(ref + 1), namelen); - - /* if we already have a perfect match, we're done */ - if (inode_in_dir(root, path, btrfs_ino(dir), btrfs_ino(inode), - btrfs_inode_ref_index(eb, ref), - name, namelen)) { - goto out; - } - - /* - * look for a conflicting back reference in the metadata. - * if we find one we have to unlink that name of the file - * before we add our new link. Later on, we overwrite any - * existing back reference, and we don't want to create - * dangling pointers in the directory. - */ - - if (search_done) - goto insert; - - ret = btrfs_search_slot(NULL, root, key, path, 0, 0); + /* Search old style refs */ + search_key.objectid = inode_objectid; + search_key.type = BTRFS_INODE_REF_KEY; + search_key.offset = parent_objectid; + ret = btrfs_search_slot(NULL, root, &search_key, path, 0, 0); if (ret == 0) { - char *victim_name; - int victim_name_len; struct btrfs_inode_ref *victim_ref; unsigned long ptr; unsigned long ptr_end; - struct extent_buffer *leaf = path->nodes[0]; + + leaf = path->nodes[0]; /* are we trying to overwrite a back ref for the root directory * if so, just jump out, we're done */ - if (key->objectid == key->offset) - goto out_nowrite; + if (search_key.objectid == search_key.offset) + return 1; /* check all the names in this back reference to see * if they are in the log. if so, we allow them to stay @@ -889,7 +851,9 @@ again: (unsigned long)(victim_ref + 1), victim_name_len); - if (!backref_in_log(log, key, victim_name, + if (!backref_in_log(log_root, &search_key, + parent_objectid, + victim_name, victim_name_len)) { btrfs_inc_nlink(inode); btrfs_release_path(path); @@ -897,9 +861,14 @@ again: ret = btrfs_unlink_inode(trans, root, dir, inode, victim_name, victim_name_len); + BUG_ON(ret); btrfs_run_delayed_items(trans, root); + kfree(victim_name); + *search_done = 1; + goto again; } kfree(victim_name); + ptr = (unsigned long)(victim_ref + 1) + victim_name_len; } BUG_ON(ret); @@ -908,14 +877,78 @@ again: * NOTE: we have searched root tree and checked the * coresponding ref, it does not need to check again. */ - search_done = 1; + *search_done = 1; + } + btrfs_release_path(path); + + /* Same search but for extended refs */ + extref = btrfs_lookup_inode_extref(NULL, root, path, name, namelen, + inode_objectid, parent_objectid, 0, + 0); + if (!IS_ERR_OR_NULL(extref)) { + u32 item_size; + u32 cur_offset = 0; + unsigned long base; + struct inode *victim_parent; + + leaf = path->nodes[0]; + + item_size = btrfs_item_size_nr(leaf, path->slots[0]); + base = btrfs_item_ptr_offset(leaf, path->slots[0]); + + while (cur_offset < item_size) { + extref = (struct btrfs_inode_extref *)base + cur_offset; + + victim_name_len = btrfs_inode_extref_name_len(leaf, extref); + + if (btrfs_inode_extref_parent(leaf, extref) != parent_objectid) + goto next; + + victim_name = kmalloc(victim_name_len, GFP_NOFS); + read_extent_buffer(leaf, victim_name, (unsigned long)&extref->name, + victim_name_len); + + search_key.objectid = inode_objectid; + search_key.type = BTRFS_INODE_EXTREF_KEY; + search_key.offset = btrfs_extref_hash(parent_objectid, + victim_name, + victim_name_len); + ret = 0; + if (!backref_in_log(log_root, &search_key, + parent_objectid, victim_name, + victim_name_len)) { + ret = -ENOENT; + victim_parent = read_one_inode(root, + parent_objectid); + if (victim_parent) { + btrfs_inc_nlink(inode); + btrfs_release_path(path); + + ret = btrfs_unlink_inode(trans, root, + victim_parent, + inode, + victim_name, + victim_name_len); + btrfs_run_delayed_items(trans, root); + } + BUG_ON(ret); + iput(victim_parent); + kfree(victim_name); + *search_done = 1; + goto again; + } + kfree(victim_name); + BUG_ON(ret); +next: + cur_offset += victim_name_len + sizeof(*extref); + } + *search_done = 1; } btrfs_release_path(path); /* look for a conflicting sequence number */ di = btrfs_lookup_dir_index_item(trans, root, path, btrfs_ino(dir), - btrfs_inode_ref_index(eb, ref), - name, namelen, 0); + ref_index, name, namelen, 0); if (di && !IS_ERR(di)) { ret = drop_one_dir_item(trans, root, path, dir, di); BUG_ON(ret); @@ -931,25 +964,173 @@ again: } btrfs_release_path(path); -insert: - /* insert our name */ - ret = btrfs_add_link(trans, dir, inode, name, namelen, 0, - btrfs_inode_ref_index(eb, ref)); - BUG_ON(ret); + return 0; +} - btrfs_update_inode(trans, root, inode); +static int extref_get_fields(struct extent_buffer *eb, unsigned long ref_ptr, + u32 *namelen, char **name, u64 *index, + u64 *parent_objectid) +{ + struct btrfs_inode_extref *extref; -out: - ref_ptr = (unsigned long)(ref + 1) + namelen; - kfree(name); - if (ref_ptr < ref_end) - goto again; + extref = (struct btrfs_inode_extref *)ref_ptr; + + *namelen = btrfs_inode_extref_name_len(eb, extref); + *name = kmalloc(*namelen, GFP_NOFS); + if (*name == NULL) + return -ENOMEM; + + read_extent_buffer(eb, *name, (unsigned long)&extref->name, + *namelen); + + *index = btrfs_inode_extref_index(eb, extref); + if (parent_objectid) + *parent_objectid = btrfs_inode_extref_parent(eb, extref); + + return 0; +} + +static int ref_get_fields(struct extent_buffer *eb, unsigned long ref_ptr, + u32 *namelen, char **name, u64 *index) +{ + struct btrfs_inode_ref *ref; + + ref = (struct btrfs_inode_ref *)ref_ptr; + + *namelen = btrfs_inode_ref_name_len(eb, ref); + *name = kmalloc(*namelen, GFP_NOFS); + if (*name == NULL) + return -ENOMEM; + + read_extent_buffer(eb, *name, (unsigned long)(ref + 1), *namelen); + + *index = btrfs_inode_ref_index(eb, ref); + + return 0; +} + +/* + * replay one inode back reference item found in the log tree. + * eb, slot and key refer to the buffer and key found in the log tree. + * root is the destination we are replaying into, and path is for temp + * use by this function. (it should be released on return). + */ +static noinline int add_inode_ref(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + struct btrfs_root *log, + struct btrfs_path *path, + struct extent_buffer *eb, int slot, + struct btrfs_key *key) +{ + struct inode *dir; + struct inode *inode; + unsigned long ref_ptr; + unsigned long ref_end; + char *name; + int namelen; + int ret; + int search_done = 0; + int log_ref_ver = 0; + u64 parent_objectid; + u64 inode_objectid; + u64 ref_index = 0; + int ref_struct_size; + + ref_ptr = btrfs_item_ptr_offset(eb, slot); + ref_end = ref_ptr + btrfs_item_size_nr(eb, slot); + + if (key->type == BTRFS_INODE_EXTREF_KEY) { + struct btrfs_inode_extref *r; + + ref_struct_size = sizeof(struct btrfs_inode_extref); + log_ref_ver = 1; + r = (struct btrfs_inode_extref *)ref_ptr; + parent_objectid = btrfs_inode_extref_parent(eb, r); + } else { + ref_struct_size = sizeof(struct btrfs_inode_ref); + parent_objectid = key->offset; + } + inode_objectid = key->objectid; + + /* + * it is possible that we didn't log all the parent directories + * for a given inode. If we don't find the dir, just don't + * copy the back ref in. The link count fixup code will take + * care of the rest + */ + dir = read_one_inode(root, parent_objectid); + if (!dir) + return -ENOENT; + + inode = read_one_inode(root, inode_objectid); + if (!inode) { + iput(dir); + return -EIO; + } + + while (ref_ptr < ref_end) { + if (log_ref_ver) { + ret = extref_get_fields(eb, ref_ptr, &namelen, &name, + &ref_index, &parent_objectid); + /* + * parent object can change from one array + * item to another. + */ + if (!dir) + dir = read_one_inode(root, parent_objectid); + if (!dir) + return -ENOENT; + } else { + ret = ref_get_fields(eb, ref_ptr, &namelen, &name, + &ref_index); + } + if (ret) + return ret; + + /* if we already have a perfect match, we're done */ + if (!inode_in_dir(root, path, btrfs_ino(dir), btrfs_ino(inode), + ref_index, name, namelen)) { + /* + * look for a conflicting back reference in the + * metadata. if we find one we have to unlink that name + * of the file before we add our new link. Later on, we + * overwrite any existing back reference, and we don't + * want to create dangling pointers in the directory. + */ + + if (!search_done) { + ret = __add_inode_ref(trans, root, path, log, + dir, inode, eb, + inode_objectid, + parent_objectid, + ref_index, name, namelen, + &search_done); + if (ret == 1) + goto out; + BUG_ON(ret); + } + + /* insert our name */ + ret = btrfs_add_link(trans, dir, inode, name, namelen, + 0, ref_index); + BUG_ON(ret); + + btrfs_update_inode(trans, root, inode); + } + + ref_ptr = (unsigned long)(ref_ptr + ref_struct_size) + namelen; + kfree(name); + if (log_ref_ver) { + iput(dir); + dir = NULL; + } + } /* finally write the back reference in the inode */ ret = overwrite_item(trans, root, path, eb, slot, key); BUG_ON(ret); -out_nowrite: +out: btrfs_release_path(path); iput(dir); iput(inode); @@ -966,25 +1147,55 @@ static int insert_orphan_item(struct btrfs_trans_handle *trans, return ret; } +static int count_inode_extrefs(struct btrfs_root *root, + struct inode *inode, struct btrfs_path *path) +{ + int ret = 0; + int name_len; + unsigned int nlink = 0; + u32 item_size; + u32 cur_offset = 0; + u64 inode_objectid = btrfs_ino(inode); + u64 offset = 0; + unsigned long ptr; + struct btrfs_inode_extref *extref; + struct extent_buffer *leaf; -/* - * There are a few corners where the link count of the file can't - * be properly maintained during replay. So, instead of adding - * lots of complexity to the log code, we just scan the backrefs - * for any file that has been through replay. - * - * The scan will update the link count on the inode to reflect the - * number of back refs found. If it goes down to zero, the iput - * will free the inode. - */ -static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - struct inode *inode) + while (1) { + ret = btrfs_find_one_extref(root, inode_objectid, offset, path, + &extref, &offset); + if (ret) + break; + + leaf = path->nodes[0]; + item_size = btrfs_item_size_nr(leaf, path->slots[0]); + ptr = btrfs_item_ptr_offset(leaf, path->slots[0]); + + while (cur_offset < item_size) { + extref = (struct btrfs_inode_extref *) (ptr + cur_offset); + name_len = btrfs_inode_extref_name_len(leaf, extref); + + nlink++; + + cur_offset += name_len + sizeof(*extref); + } + + offset++; + btrfs_release_path(path); + } + btrfs_release_path(path); + + if (ret < 0) + return ret; + return nlink; +} + +static int count_inode_refs(struct btrfs_root *root, + struct inode *inode, struct btrfs_path *path) { - struct btrfs_path *path; int ret; struct btrfs_key key; - u64 nlink = 0; + unsigned int nlink = 0; unsigned long ptr; unsigned long ptr_end; int name_len; @@ -994,10 +1205,6 @@ static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans, key.type = BTRFS_INODE_REF_KEY; key.offset = (u64)-1; - path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; - while (1) { ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); if (ret < 0) @@ -1031,6 +1238,50 @@ static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans, btrfs_release_path(path); } btrfs_release_path(path); + + return nlink; +} + +/* + * There are a few corners where the link count of the file can't + * be properly maintained during replay. So, instead of adding + * lots of complexity to the log code, we just scan the backrefs + * for any file that has been through replay. + * + * The scan will update the link count on the inode to reflect the + * number of back refs found. If it goes down to zero, the iput + * will free the inode. + */ +static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + struct inode *inode) +{ + struct btrfs_path *path; + int ret; + u64 nlink = 0; + u64 ino = btrfs_ino(inode); + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + ret = count_inode_refs(root, inode, path); + if (ret < 0) + goto out; + + nlink = ret; + + ret = count_inode_extrefs(root, inode, path); + if (ret == -ENOENT) + ret = 0; + + if (ret < 0) + goto out; + + nlink += ret; + + ret = 0; + if (nlink != inode->i_nlink) { set_nlink(inode, nlink); btrfs_update_inode(trans, root, inode); @@ -1046,9 +1297,10 @@ static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans, ret = insert_orphan_item(trans, root, ino); BUG_ON(ret); } - btrfs_free_path(path); - return 0; +out: + btrfs_free_path(path); + return ret; } static noinline int fixup_inode_link_counts(struct btrfs_trans_handle *trans, @@ -1695,6 +1947,10 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb, ret = add_inode_ref(wc->trans, root, log, path, eb, i, &key); BUG_ON(ret && ret != -ENOENT); + } else if (key.type == BTRFS_INODE_EXTREF_KEY) { + ret = add_inode_ref(wc->trans, root, log, path, + eb, i, &key); + BUG_ON(ret && ret != -ENOENT); } else if (key.type == BTRFS_EXTENT_DATA_KEY) { ret = replay_one_extent(wc->trans, root, path, eb, i, &key); @@ -2037,7 +2293,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, if (atomic_read(&root->log_commit[(index1 + 1) % 2])) wait_log_commit(trans, root, root->log_transid - 1); while (1) { - unsigned long batch = root->log_batch; + int batch = atomic_read(&root->log_batch); /* when we're on an ssd, just kick the log commit out */ if (!btrfs_test_opt(root, SSD) && root->log_multiple_pids) { mutex_unlock(&root->log_mutex); @@ -2045,7 +2301,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, mutex_lock(&root->log_mutex); } wait_for_writer(trans, root); - if (batch == root->log_batch) + if (batch == atomic_read(&root->log_batch)) break; } @@ -2074,7 +2330,6 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, btrfs_set_root_node(&log->root_item, log->node); - root->log_batch = 0; root->log_transid++; log->log_transid = root->log_transid; root->log_start_pid = 0; @@ -2087,7 +2342,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, mutex_unlock(&root->log_mutex); mutex_lock(&log_root_tree->log_mutex); - log_root_tree->log_batch++; + atomic_inc(&log_root_tree->log_batch); atomic_inc(&log_root_tree->log_writers); mutex_unlock(&log_root_tree->log_mutex); @@ -2157,7 +2412,6 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, btrfs_set_super_log_root_level(root->fs_info->super_for_commit, btrfs_header_level(log_root_tree->node)); - log_root_tree->log_batch = 0; log_root_tree->log_transid++; smp_mb(); @@ -2171,9 +2425,12 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, * in and cause problems either. */ btrfs_scrub_pause_super(root); - write_ctree_super(trans, root->fs_info->tree_root, 1); + ret = write_ctree_super(trans, root->fs_info->tree_root, 1); btrfs_scrub_continue_super(root); - ret = 0; + if (ret) { + btrfs_abort_transaction(trans, root, ret); + goto out_wake_log_root; + } mutex_lock(&root->log_mutex); if (root->last_log_commit < log_transid) @@ -2209,7 +2466,8 @@ static void free_log_tree(struct btrfs_trans_handle *trans, while (1) { ret = find_first_extent_bit(&log->dirty_log_pages, - 0, &start, &end, EXTENT_DIRTY | EXTENT_NEW); + 0, &start, &end, EXTENT_DIRTY | EXTENT_NEW, + NULL); if (ret) break; @@ -2646,6 +2904,7 @@ static int drop_objectid_items(struct btrfs_trans_handle *trans, int ret; struct btrfs_key key; struct btrfs_key found_key; + int start_slot; key.objectid = objectid; key.type = max_key_type; @@ -2667,8 +2926,18 @@ static int drop_objectid_items(struct btrfs_trans_handle *trans, if (found_key.objectid != objectid) break; - ret = btrfs_del_item(trans, log, path); - if (ret) + found_key.offset = 0; + found_key.type = 0; + ret = btrfs_bin_search(path->nodes[0], &found_key, 0, + &start_slot); + + ret = btrfs_del_items(trans, log, path, start_slot, + path->slots[0] - start_slot + 1); + /* + * If start slot isn't 0 then we don't need to re-search, we've + * found the last guy with the objectid in this tree. + */ + if (ret || start_slot != 0) break; btrfs_release_path(path); } @@ -2678,14 +2947,64 @@ static int drop_objectid_items(struct btrfs_trans_handle *trans, return ret; } +static void fill_inode_item(struct btrfs_trans_handle *trans, + struct extent_buffer *leaf, + struct btrfs_inode_item *item, + struct inode *inode, int log_inode_only) +{ + btrfs_set_inode_uid(leaf, item, i_uid_read(inode)); + btrfs_set_inode_gid(leaf, item, i_gid_read(inode)); + btrfs_set_inode_mode(leaf, item, inode->i_mode); + btrfs_set_inode_nlink(leaf, item, inode->i_nlink); + + btrfs_set_timespec_sec(leaf, btrfs_inode_atime(item), + inode->i_atime.tv_sec); + btrfs_set_timespec_nsec(leaf, btrfs_inode_atime(item), + inode->i_atime.tv_nsec); + + btrfs_set_timespec_sec(leaf, btrfs_inode_mtime(item), + inode->i_mtime.tv_sec); + btrfs_set_timespec_nsec(leaf, btrfs_inode_mtime(item), + inode->i_mtime.tv_nsec); + + btrfs_set_timespec_sec(leaf, btrfs_inode_ctime(item), + inode->i_ctime.tv_sec); + btrfs_set_timespec_nsec(leaf, btrfs_inode_ctime(item), + inode->i_ctime.tv_nsec); + + btrfs_set_inode_nbytes(leaf, item, inode_get_bytes(inode)); + + btrfs_set_inode_sequence(leaf, item, inode->i_version); + btrfs_set_inode_transid(leaf, item, trans->transid); + btrfs_set_inode_rdev(leaf, item, inode->i_rdev); + btrfs_set_inode_flags(leaf, item, BTRFS_I(inode)->flags); + btrfs_set_inode_block_group(leaf, item, 0); + + if (log_inode_only) { + /* set the generation to zero so the recover code + * can tell the difference between an logging + * just to say 'this inode exists' and a logging + * to say 'update this inode with these values' + */ + btrfs_set_inode_generation(leaf, item, 0); + btrfs_set_inode_size(leaf, item, 0); + } else { + btrfs_set_inode_generation(leaf, item, + BTRFS_I(inode)->generation); + btrfs_set_inode_size(leaf, item, inode->i_size); + } + +} + static noinline int copy_items(struct btrfs_trans_handle *trans, - struct btrfs_root *log, + struct inode *inode, struct btrfs_path *dst_path, struct extent_buffer *src, int start_slot, int nr, int inode_only) { unsigned long src_offset; unsigned long dst_offset; + struct btrfs_root *log = BTRFS_I(inode)->root->log_root; struct btrfs_file_extent_item *extent; struct btrfs_inode_item *inode_item; int ret; @@ -2694,6 +3013,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans, char *ins_data; int i; struct list_head ordered_sums; + int skip_csum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; INIT_LIST_HEAD(&ordered_sums); @@ -2722,29 +3042,23 @@ static noinline int copy_items(struct btrfs_trans_handle *trans, src_offset = btrfs_item_ptr_offset(src, start_slot + i); - copy_extent_buffer(dst_path->nodes[0], src, dst_offset, - src_offset, ins_sizes[i]); - - if (inode_only == LOG_INODE_EXISTS && - ins_keys[i].type == BTRFS_INODE_ITEM_KEY) { + if (ins_keys[i].type == BTRFS_INODE_ITEM_KEY) { inode_item = btrfs_item_ptr(dst_path->nodes[0], dst_path->slots[0], struct btrfs_inode_item); - btrfs_set_inode_size(dst_path->nodes[0], inode_item, 0); - - /* set the generation to zero so the recover code - * can tell the difference between an logging - * just to say 'this inode exists' and a logging - * to say 'update this inode with these values' - */ - btrfs_set_inode_generation(dst_path->nodes[0], - inode_item, 0); + fill_inode_item(trans, dst_path->nodes[0], inode_item, + inode, inode_only == LOG_INODE_EXISTS); + } else { + copy_extent_buffer(dst_path->nodes[0], src, dst_offset, + src_offset, ins_sizes[i]); } + /* take a reference on file data extents so that truncates * or deletes of this inode don't have to relog the inode * again */ - if (btrfs_key_type(ins_keys + i) == BTRFS_EXTENT_DATA_KEY) { + if (btrfs_key_type(ins_keys + i) == BTRFS_EXTENT_DATA_KEY && + !skip_csum) { int found_type; extent = btrfs_item_ptr(src, start_slot + i, struct btrfs_file_extent_item); @@ -2753,8 +3067,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans, continue; found_type = btrfs_file_extent_type(src, extent); - if (found_type == BTRFS_FILE_EXTENT_REG || - found_type == BTRFS_FILE_EXTENT_PREALLOC) { + if (found_type == BTRFS_FILE_EXTENT_REG) { u64 ds, dl, cs, cl; ds = btrfs_file_extent_disk_bytenr(src, extent); @@ -2803,6 +3116,239 @@ static noinline int copy_items(struct btrfs_trans_handle *trans, return ret; } +static int extent_cmp(void *priv, struct list_head *a, struct list_head *b) +{ + struct extent_map *em1, *em2; + + em1 = list_entry(a, struct extent_map, list); + em2 = list_entry(b, struct extent_map, list); + + if (em1->start < em2->start) + return -1; + else if (em1->start > em2->start) + return 1; + return 0; +} + +struct log_args { + struct extent_buffer *src; + u64 next_offset; + int start_slot; + int nr; +}; + +static int log_one_extent(struct btrfs_trans_handle *trans, + struct inode *inode, struct btrfs_root *root, + struct extent_map *em, struct btrfs_path *path, + struct btrfs_path *dst_path, struct log_args *args) +{ + struct btrfs_root *log = root->log_root; + struct btrfs_file_extent_item *fi; + struct btrfs_key key; + u64 start = em->mod_start; + u64 search_start = start; + u64 len = em->mod_len; + u64 num_bytes; + int nritems; + int ret; + + if (BTRFS_I(inode)->logged_trans == trans->transid) { + ret = __btrfs_drop_extents(trans, log, inode, dst_path, start, + start + len, NULL, 0); + if (ret) + return ret; + } + + while (len) { + if (args->nr) + goto next_slot; +again: + key.objectid = btrfs_ino(inode); + key.type = BTRFS_EXTENT_DATA_KEY; + key.offset = search_start; + + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) + return ret; + + if (ret) { + /* + * A rare case were we can have an em for a section of a + * larger extent so we need to make sure that this em + * falls within the extent we've found. If not we just + * bail and go back to ye-olde way of doing things but + * it happens often enough in testing that we need to do + * this dance to make sure. + */ + do { + if (path->slots[0] == 0) { + btrfs_release_path(path); + if (search_start == 0) + return -ENOENT; + search_start--; + goto again; + } + + path->slots[0]--; + btrfs_item_key_to_cpu(path->nodes[0], &key, + path->slots[0]); + if (key.objectid != btrfs_ino(inode) || + key.type != BTRFS_EXTENT_DATA_KEY) { + btrfs_release_path(path); + return -ENOENT; + } + } while (key.offset > start); + + fi = btrfs_item_ptr(path->nodes[0], path->slots[0], + struct btrfs_file_extent_item); + num_bytes = btrfs_file_extent_num_bytes(path->nodes[0], + fi); + if (key.offset + num_bytes <= start) { + btrfs_release_path(path); + return -ENOENT; + } + } + args->src = path->nodes[0]; +next_slot: + btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); + fi = btrfs_item_ptr(args->src, path->slots[0], + struct btrfs_file_extent_item); + if (args->nr && + args->start_slot + args->nr == path->slots[0]) { + args->nr++; + } else if (args->nr) { + ret = copy_items(trans, inode, dst_path, args->src, + args->start_slot, args->nr, + LOG_INODE_ALL); + if (ret) + return ret; + args->nr = 1; + args->start_slot = path->slots[0]; + } else if (!args->nr) { + args->nr = 1; + args->start_slot = path->slots[0]; + } + nritems = btrfs_header_nritems(path->nodes[0]); + path->slots[0]++; + num_bytes = btrfs_file_extent_num_bytes(args->src, fi); + if (len < num_bytes) { + /* I _think_ this is ok, envision we write to a + * preallocated space that is adjacent to a previously + * written preallocated space that gets merged when we + * mark this preallocated space written. If we do not + * have the adjacent extent in cache then when we copy + * this extent it could end up being larger than our EM + * thinks it is, which is a-ok, so just set len to 0. + */ + len = 0; + } else { + len -= num_bytes; + } + start = key.offset + num_bytes; + args->next_offset = start; + search_start = start; + + if (path->slots[0] < nritems) { + if (len) + goto next_slot; + break; + } + + if (args->nr) { + ret = copy_items(trans, inode, dst_path, args->src, + args->start_slot, args->nr, + LOG_INODE_ALL); + if (ret) + return ret; + args->nr = 0; + btrfs_release_path(path); + } + } + + return 0; +} + +static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + struct inode *inode, + struct btrfs_path *path, + struct btrfs_path *dst_path) +{ + struct log_args args; + struct extent_map *em, *n; + struct list_head extents; + struct extent_map_tree *tree = &BTRFS_I(inode)->extent_tree; + u64 test_gen; + int ret = 0; + + INIT_LIST_HEAD(&extents); + + memset(&args, 0, sizeof(args)); + + write_lock(&tree->lock); + test_gen = root->fs_info->last_trans_committed; + + list_for_each_entry_safe(em, n, &tree->modified_extents, list) { + list_del_init(&em->list); + if (em->generation <= test_gen) + continue; + /* Need a ref to keep it from getting evicted from cache */ + atomic_inc(&em->refs); + set_bit(EXTENT_FLAG_LOGGING, &em->flags); + list_add_tail(&em->list, &extents); + } + + list_sort(NULL, &extents, extent_cmp); + + while (!list_empty(&extents)) { + em = list_entry(extents.next, struct extent_map, list); + + list_del_init(&em->list); + clear_bit(EXTENT_FLAG_LOGGING, &em->flags); + + /* + * If we had an error we just need to delete everybody from our + * private list. + */ + if (ret) { + free_extent_map(em); + continue; + } + + write_unlock(&tree->lock); + + /* + * If the previous EM and the last extent we left off on aren't + * sequential then we need to copy the items we have and redo + * our search + */ + if (args.nr && em->mod_start != args.next_offset) { + ret = copy_items(trans, inode, dst_path, args.src, + args.start_slot, args.nr, + LOG_INODE_ALL); + if (ret) { + free_extent_map(em); + write_lock(&tree->lock); + continue; + } + btrfs_release_path(path); + args.nr = 0; + } + + ret = log_one_extent(trans, inode, root, em, path, dst_path, &args); + free_extent_map(em); + write_lock(&tree->lock); + } + WARN_ON(!list_empty(&extents)); + write_unlock(&tree->lock); + + if (!ret && args.nr) + ret = copy_items(trans, inode, dst_path, args.src, + args.start_slot, args.nr, LOG_INODE_ALL); + btrfs_release_path(path); + return ret; +} + /* log a single inode in the tree log. * At least one parent directory for this inode must exist in the tree * or be logged already. @@ -2832,6 +3378,7 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, int nritems; int ins_start_slot = 0; int ins_nr; + bool fast_search = false; u64 ino = btrfs_ino(inode); log = root->log_root; @@ -2851,21 +3398,23 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, max_key.objectid = ino; - /* today the code can only do partial logging of directories */ - if (!S_ISDIR(inode->i_mode)) - inode_only = LOG_INODE_ALL; + /* today the code can only do partial logging of directories */ if (inode_only == LOG_INODE_EXISTS || S_ISDIR(inode->i_mode)) max_key.type = BTRFS_XATTR_ITEM_KEY; else max_key.type = (u8)-1; max_key.offset = (u64)-1; - ret = btrfs_commit_inode_delayed_items(trans, inode); - if (ret) { - btrfs_free_path(path); - btrfs_free_path(dst_path); - return ret; + /* Only run delayed items if we are a dir or a new file */ + if (S_ISDIR(inode->i_mode) || + BTRFS_I(inode)->generation > root->fs_info->last_trans_committed) { + ret = btrfs_commit_inode_delayed_items(trans, inode); + if (ret) { + btrfs_free_path(path); + btrfs_free_path(dst_path); + return ret; + } } mutex_lock(&BTRFS_I(inode)->log_mutex); @@ -2881,7 +3430,16 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, max_key_type = BTRFS_XATTR_ITEM_KEY; ret = drop_objectid_items(trans, log, path, ino, max_key_type); } else { - ret = btrfs_truncate_inode_items(trans, log, inode, 0, 0); + if (test_and_clear_bit(BTRFS_INODE_NEEDS_FULL_SYNC, + &BTRFS_I(inode)->runtime_flags)) { + ret = btrfs_truncate_inode_items(trans, log, + inode, 0, 0); + } else { + fast_search = true; + max_key.type = BTRFS_XATTR_ITEM_KEY; + ret = drop_objectid_items(trans, log, path, ino, + BTRFS_XATTR_ITEM_KEY); + } } if (ret) { err = ret; @@ -2912,7 +3470,7 @@ again: goto next_slot; } - ret = copy_items(trans, log, dst_path, src, ins_start_slot, + ret = copy_items(trans, inode, dst_path, src, ins_start_slot, ins_nr, inode_only); if (ret) { err = ret; @@ -2930,7 +3488,7 @@ next_slot: goto again; } if (ins_nr) { - ret = copy_items(trans, log, dst_path, src, + ret = copy_items(trans, inode, dst_path, src, ins_start_slot, ins_nr, inode_only); if (ret) { @@ -2951,8 +3509,7 @@ next_slot: break; } if (ins_nr) { - ret = copy_items(trans, log, dst_path, src, - ins_start_slot, + ret = copy_items(trans, inode, dst_path, src, ins_start_slot, ins_nr, inode_only); if (ret) { err = ret; @@ -2960,7 +3517,24 @@ next_slot: } ins_nr = 0; } - WARN_ON(ins_nr); + + if (fast_search) { + btrfs_release_path(path); + btrfs_release_path(dst_path); + ret = btrfs_log_changed_extents(trans, root, inode, path, + dst_path); + if (ret) { + err = ret; + goto out_unlock; + } + } else { + struct extent_map_tree *tree = &BTRFS_I(inode)->extent_tree; + struct extent_map *em, *n; + + list_for_each_entry_safe(em, n, &tree->modified_extents, list) + list_del_init(&em->list); + } + if (inode_only == LOG_INODE_ALL && S_ISDIR(inode->i_mode)) { btrfs_release_path(path); btrfs_release_path(dst_path); @@ -2971,6 +3545,7 @@ next_slot: } } BTRFS_I(inode)->logged_trans = trans->transid; + BTRFS_I(inode)->last_log_commit = BTRFS_I(inode)->last_sub_trans; out_unlock: mutex_unlock(&BTRFS_I(inode)->log_mutex); @@ -3138,7 +3713,7 @@ int btrfs_log_inode_parent(struct btrfs_trans_handle *trans, end_trans: dput(old_parent); if (ret < 0) { - BUG_ON(ret != -ENOSPC); + WARN_ON(ret != -ENOSPC); root->fs_info->last_trans_log_full_commit = trans->transid; ret = 1; } diff --git a/fs/btrfs/ulist.c b/fs/btrfs/ulist.c index ab942f46b3dd..99be4c138db6 100644 --- a/fs/btrfs/ulist.c +++ b/fs/btrfs/ulist.c @@ -143,14 +143,13 @@ EXPORT_SYMBOL(ulist_free); * In case of allocation failure -ENOMEM is returned and the ulist stays * unaltered. */ -int ulist_add(struct ulist *ulist, u64 val, unsigned long aux, - gfp_t gfp_mask) +int ulist_add(struct ulist *ulist, u64 val, u64 aux, gfp_t gfp_mask) { return ulist_add_merge(ulist, val, aux, NULL, gfp_mask); } -int ulist_add_merge(struct ulist *ulist, u64 val, unsigned long aux, - unsigned long *old_aux, gfp_t gfp_mask) +int ulist_add_merge(struct ulist *ulist, u64 val, u64 aux, + u64 *old_aux, gfp_t gfp_mask) { int i; diff --git a/fs/btrfs/ulist.h b/fs/btrfs/ulist.h index 21bdc8ec8130..21a1963439c3 100644 --- a/fs/btrfs/ulist.h +++ b/fs/btrfs/ulist.h @@ -33,7 +33,7 @@ struct ulist_iterator { */ struct ulist_node { u64 val; /* value to store */ - unsigned long aux; /* auxiliary value saved along with the val */ + u64 aux; /* auxiliary value saved along with the val */ }; struct ulist { @@ -65,10 +65,9 @@ void ulist_fini(struct ulist *ulist); void ulist_reinit(struct ulist *ulist); struct ulist *ulist_alloc(gfp_t gfp_mask); void ulist_free(struct ulist *ulist); -int ulist_add(struct ulist *ulist, u64 val, unsigned long aux, - gfp_t gfp_mask); -int ulist_add_merge(struct ulist *ulist, u64 val, unsigned long aux, - unsigned long *old_aux, gfp_t gfp_mask); +int ulist_add(struct ulist *ulist, u64 val, u64 aux, gfp_t gfp_mask); +int ulist_add_merge(struct ulist *ulist, u64 val, u64 aux, + u64 *old_aux, gfp_t gfp_mask); struct ulist_node *ulist_next(struct ulist *ulist, struct ulist_iterator *uiter); diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 88b969aeeb71..0f5ebb72a5ea 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -639,7 +639,7 @@ static int __btrfs_open_devices(struct btrfs_fs_devices *fs_devices, bdev = blkdev_get_by_path(device->name->str, flags, holder); if (IS_ERR(bdev)) { - printk(KERN_INFO "open %s failed\n", device->name->str); + printk(KERN_INFO "btrfs: open %s failed\n", device->name->str); goto error; } filemap_write_and_wait(bdev->bd_inode->i_mapping); @@ -1475,6 +1475,9 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path) free_fs_devices(cur_devices); } + root->fs_info->num_tolerated_disk_barrier_failures = + btrfs_calc_num_tolerated_disk_barrier_failures(root->fs_info); + /* * at this point, the device is zero sized. We want to * remove it from the devices list and zero out the old super @@ -1775,15 +1778,21 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path) if (seeding_dev) { ret = init_first_rw_device(trans, root, device); - if (ret) + if (ret) { + btrfs_abort_transaction(trans, root, ret); goto error_trans; + } ret = btrfs_finish_sprout(trans, root); - if (ret) + if (ret) { + btrfs_abort_transaction(trans, root, ret); goto error_trans; + } } else { ret = btrfs_add_device(trans, root, device); - if (ret) + if (ret) { + btrfs_abort_transaction(trans, root, ret); goto error_trans; + } } /* @@ -1793,6 +1802,8 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path) btrfs_clear_space_info_full(root->fs_info); unlock_chunks(root); + root->fs_info->num_tolerated_disk_barrier_failures = + btrfs_calc_num_tolerated_disk_barrier_failures(root->fs_info); ret = btrfs_commit_transaction(trans, root); if (seeding_dev) { @@ -1808,13 +1819,19 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path) "Failed to relocate sys chunks after " "device initialization. This can be fixed " "using the \"btrfs balance\" command."); + trans = btrfs_attach_transaction(root); + if (IS_ERR(trans)) { + if (PTR_ERR(trans) == -ENOENT) + return 0; + return PTR_ERR(trans); + } + ret = btrfs_commit_transaction(trans, root); } return ret; error_trans: unlock_chunks(root); - btrfs_abort_transaction(trans, root, ret); btrfs_end_transaction(trans, root); rcu_string_free(device->name); kfree(device); @@ -2804,6 +2821,26 @@ int btrfs_balance(struct btrfs_balance_control *bctl, } } + if (bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) { + int num_tolerated_disk_barrier_failures; + u64 target = bctl->sys.target; + + num_tolerated_disk_barrier_failures = + btrfs_calc_num_tolerated_disk_barrier_failures(fs_info); + if (num_tolerated_disk_barrier_failures > 0 && + (target & + (BTRFS_BLOCK_GROUP_DUP | BTRFS_BLOCK_GROUP_RAID0 | + BTRFS_AVAIL_ALLOC_BIT_SINGLE))) + num_tolerated_disk_barrier_failures = 0; + else if (num_tolerated_disk_barrier_failures > 1 && + (target & + (BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID10))) + num_tolerated_disk_barrier_failures = 1; + + fs_info->num_tolerated_disk_barrier_failures = + num_tolerated_disk_barrier_failures; + } + ret = insert_balance_item(fs_info->tree_root, bctl); if (ret && ret != -EEXIST) goto out; @@ -2836,6 +2873,11 @@ int btrfs_balance(struct btrfs_balance_control *bctl, __cancel_balance(fs_info); } + if (bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) { + fs_info->num_tolerated_disk_barrier_failures = + btrfs_calc_num_tolerated_disk_barrier_failures(fs_info); + } + wake_up(&fs_info->balance_wait_q); return ret; @@ -3608,12 +3650,16 @@ static noinline int init_first_rw_device(struct btrfs_trans_handle *trans, ret = __btrfs_alloc_chunk(trans, extent_root, &sys_map, &sys_chunk_size, &sys_stripe_size, sys_chunk_offset, alloc_profile); - if (ret) - goto abort; + if (ret) { + btrfs_abort_transaction(trans, root, ret); + goto out; + } ret = btrfs_add_device(trans, fs_info->chunk_root, device); - if (ret) - goto abort; + if (ret) { + btrfs_abort_transaction(trans, root, ret); + goto out; + } /* * Modifying chunk tree needs allocating new blocks from both @@ -3623,19 +3669,19 @@ static noinline int init_first_rw_device(struct btrfs_trans_handle *trans, */ ret = __finish_chunk_alloc(trans, extent_root, map, chunk_offset, chunk_size, stripe_size); - if (ret) - goto abort; + if (ret) { + btrfs_abort_transaction(trans, root, ret); + goto out; + } ret = __finish_chunk_alloc(trans, extent_root, sys_map, sys_chunk_offset, sys_chunk_size, sys_stripe_size); if (ret) - goto abort; + btrfs_abort_transaction(trans, root, ret); - return 0; +out: -abort: - btrfs_abort_transaction(trans, root, ret); return ret; } @@ -3760,7 +3806,7 @@ static int __btrfs_map_block(struct btrfs_mapping_tree *map_tree, int rw, read_unlock(&em_tree->lock); if (!em) { - printk(KERN_CRIT "unable to find logical %llu len %llu\n", + printk(KERN_CRIT "btrfs: unable to find logical %llu len %llu\n", (unsigned long long)logical, (unsigned long long)*length); BUG(); @@ -4217,7 +4263,7 @@ int btrfs_map_bio(struct btrfs_root *root, int rw, struct bio *bio, total_devs = bbio->num_stripes; if (map_length < length) { - printk(KERN_CRIT "mapping failed logical %llu bio len %llu " + printk(KERN_CRIT "btrfs: mapping failed logical %llu bio len %llu " "len %llu\n", (unsigned long long)logical, (unsigned long long)length, (unsigned long long)map_length); diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c index 92c20654cc55..9acb846c3e7f 100644 --- a/fs/btrfs/zlib.c +++ b/fs/btrfs/zlib.c @@ -97,7 +97,7 @@ static int zlib_compress_pages(struct list_head *ws, *total_in = 0; if (Z_OK != zlib_deflateInit(&workspace->def_strm, 3)) { - printk(KERN_WARNING "deflateInit failed\n"); + printk(KERN_WARNING "btrfs: deflateInit failed\n"); ret = -1; goto out; } @@ -125,7 +125,7 @@ static int zlib_compress_pages(struct list_head *ws, while (workspace->def_strm.total_in < len) { ret = zlib_deflate(&workspace->def_strm, Z_SYNC_FLUSH); if (ret != Z_OK) { - printk(KERN_DEBUG "btrfs deflate in loop returned %d\n", + printk(KERN_DEBUG "btrfs: deflate in loop returned %d\n", ret); zlib_deflateEnd(&workspace->def_strm); ret = -1; @@ -252,7 +252,7 @@ static int zlib_decompress_biovec(struct list_head *ws, struct page **pages_in, } if (Z_OK != zlib_inflateInit2(&workspace->inf_strm, wbits)) { - printk(KERN_WARNING "inflateInit failed\n"); + printk(KERN_WARNING "btrfs: inflateInit failed\n"); return -1; } while (workspace->inf_strm.total_in < srclen) { @@ -336,7 +336,7 @@ static int zlib_decompress(struct list_head *ws, unsigned char *data_in, } if (Z_OK != zlib_inflateInit2(&workspace->inf_strm, wbits)) { - printk(KERN_WARNING "inflateInit failed\n"); + printk(KERN_WARNING "btrfs: inflateInit failed\n"); return -1; } diff --git a/fs/buffer.c b/fs/buffer.c index 58e2e7b77372..b5f044283edb 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2312,12 +2312,6 @@ int __block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf, loff_t size; int ret; - /* - * Update file times before taking page lock. We may end up failing the - * fault so this update may be superfluous but who really cares... - */ - file_update_time(vma->vm_file); - lock_page(page); size = i_size_read(inode); if ((page->mapping != inode->i_mapping) || @@ -2355,6 +2349,13 @@ int block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf, struct super_block *sb = vma->vm_file->f_path.dentry->d_inode->i_sb; sb_start_pagefault(sb); + + /* + * Update file times before taking page lock. We may end up failing the + * fault so this update may be superfluous but who really cares... + */ + file_update_time(vma->vm_file); + ret = __block_page_mkwrite(vma, vmf, get_block); sb_end_pagefault(sb); return block_page_mkwrite_return(ret); diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 452e71a1b753..6690269f5dde 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -205,7 +205,7 @@ static int readpage_nounlock(struct file *filp, struct page *page) dout("readpage inode %p file %p page %p index %lu\n", inode, filp, page, page->index); err = ceph_osdc_readpages(osdc, ceph_vino(inode), &ci->i_layout, - page->index << PAGE_CACHE_SHIFT, &len, + (u64) page_offset(page), &len, ci->i_truncate_seq, ci->i_truncate_size, &page, 1, 0); if (err == -ENOENT) @@ -286,7 +286,7 @@ static int start_read(struct inode *inode, struct list_head *page_list, int max) int nr_pages = 0; int ret; - off = page->index << PAGE_CACHE_SHIFT; + off = (u64) page_offset(page); /* count pages */ next_index = page->index; @@ -308,8 +308,8 @@ static int start_read(struct inode *inode, struct list_head *page_list, int max) NULL, 0, ci->i_truncate_seq, ci->i_truncate_size, NULL, false, 1, 0); - if (!req) - return -ENOMEM; + if (IS_ERR(req)) + return PTR_ERR(req); /* build page vector */ nr_pages = len >> PAGE_CACHE_SHIFT; @@ -426,7 +426,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc) struct ceph_inode_info *ci; struct ceph_fs_client *fsc; struct ceph_osd_client *osdc; - loff_t page_off = page->index << PAGE_CACHE_SHIFT; + loff_t page_off = page_offset(page); int len = PAGE_CACHE_SIZE; loff_t i_size; int err = 0; @@ -817,8 +817,7 @@ get_more_pages: /* ok */ if (locked_pages == 0) { /* prepare async write request */ - offset = (unsigned long long)page->index - << PAGE_CACHE_SHIFT; + offset = (u64) page_offset(page); len = wsize; req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, @@ -832,8 +831,8 @@ get_more_pages: ci->i_truncate_size, &inode->i_mtime, true, 1, 0); - if (!req) { - rc = -ENOMEM; + if (IS_ERR(req)) { + rc = PTR_ERR(req); unlock_page(page); break; } @@ -1180,7 +1179,7 @@ static int ceph_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) struct inode *inode = vma->vm_file->f_dentry->d_inode; struct page *page = vmf->page; struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc; - loff_t off = page->index << PAGE_CACHE_SHIFT; + loff_t off = page_offset(page); loff_t size, len; int ret; @@ -1225,6 +1224,7 @@ out: static struct vm_operations_struct ceph_vmops = { .fault = filemap_fault, .page_mkwrite = ceph_page_mkwrite, + .remap_pages = generic_file_remap_pages, }; int ceph_mmap(struct file *file, struct vm_area_struct *vma) @@ -1235,6 +1235,5 @@ int ceph_mmap(struct file *file, struct vm_area_struct *vma) return -ENOEXEC; file_accessed(file); vma->vm_ops = &ceph_vmops; - vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 620daad201db..3251e9cc6401 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -1005,7 +1005,7 @@ static void __queue_cap_release(struct ceph_mds_session *session, BUG_ON(msg->front.iov_len + sizeof(*item) > PAGE_CACHE_SIZE); head = msg->front.iov_base; - head->num = cpu_to_le32(le32_to_cpu(head->num) + 1); + le32_add_cpu(&head->num, 1); item = msg->front.iov_base + msg->front.iov_len; item->ino = cpu_to_le64(ino); item->cap_id = cpu_to_le64(cap_id); diff --git a/fs/ceph/export.c b/fs/ceph/export.c index 8e1b60e557b6..9349bb37a2fe 100644 --- a/fs/ceph/export.c +++ b/fs/ceph/export.c @@ -90,6 +90,8 @@ static int ceph_encode_fh(struct inode *inode, u32 *rawfh, int *max_len, *max_len = handle_length; type = 255; } + if (dentry) + dput(dentry); return type; } @@ -99,7 +101,7 @@ static int ceph_encode_fh(struct inode *inode, u32 *rawfh, int *max_len, * FIXME: we should try harder by querying the mds for the ino. */ static struct dentry *__fh_to_dentry(struct super_block *sb, - struct ceph_nfs_fh *fh) + struct ceph_nfs_fh *fh, int fh_len) { struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc; struct inode *inode; @@ -107,6 +109,9 @@ static struct dentry *__fh_to_dentry(struct super_block *sb, struct ceph_vino vino; int err; + if (fh_len < sizeof(*fh) / 4) + return ERR_PTR(-ESTALE); + dout("__fh_to_dentry %llx\n", fh->ino); vino.ino = fh->ino; vino.snap = CEPH_NOSNAP; @@ -150,7 +155,7 @@ static struct dentry *__fh_to_dentry(struct super_block *sb, * convert connectable fh to dentry */ static struct dentry *__cfh_to_dentry(struct super_block *sb, - struct ceph_nfs_confh *cfh) + struct ceph_nfs_confh *cfh, int fh_len) { struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc; struct inode *inode; @@ -158,6 +163,9 @@ static struct dentry *__cfh_to_dentry(struct super_block *sb, struct ceph_vino vino; int err; + if (fh_len < sizeof(*cfh) / 4) + return ERR_PTR(-ESTALE); + dout("__cfh_to_dentry %llx (%llx/%x)\n", cfh->ino, cfh->parent_ino, cfh->parent_name_hash); @@ -207,9 +215,11 @@ static struct dentry *ceph_fh_to_dentry(struct super_block *sb, struct fid *fid, int fh_len, int fh_type) { if (fh_type == 1) - return __fh_to_dentry(sb, (struct ceph_nfs_fh *)fid->raw); + return __fh_to_dentry(sb, (struct ceph_nfs_fh *)fid->raw, + fh_len); else - return __cfh_to_dentry(sb, (struct ceph_nfs_confh *)fid->raw); + return __cfh_to_dentry(sb, (struct ceph_nfs_confh *)fid->raw, + fh_len); } /* @@ -230,6 +240,8 @@ static struct dentry *ceph_fh_to_parent(struct super_block *sb, if (fh_type == 1) return ERR_PTR(-ESTALE); + if (fh_len < sizeof(*cfh) / 4) + return ERR_PTR(-ESTALE); pr_debug("fh_to_parent %llx/%d\n", cfh->parent_ino, cfh->parent_name_hash); diff --git a/fs/ceph/file.c b/fs/ceph/file.c index ecebbc09bfc7..5840d2aaed15 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -536,8 +536,8 @@ more: do_sync, ci->i_truncate_seq, ci->i_truncate_size, &mtime, false, 2, page_align); - if (!req) - return -ENOMEM; + if (IS_ERR(req)) + return PTR_ERR(req); if (file->f_flags & O_DIRECT) { pages = ceph_get_direct_page_vector(data, num_pages, false); diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c index 1396ceb46797..36549a46e311 100644 --- a/fs/ceph/ioctl.c +++ b/fs/ceph/ioctl.c @@ -187,14 +187,18 @@ static long ceph_ioctl_get_dataloc(struct file *file, void __user *arg) u64 tmp; struct ceph_object_layout ol; struct ceph_pg pgid; + int r; /* copy and validate */ if (copy_from_user(&dl, arg, sizeof(dl))) return -EFAULT; down_read(&osdc->map_sem); - ceph_calc_file_object_mapping(&ci->i_layout, dl.file_offset, &len, - &dl.object_no, &dl.object_offset, &olen); + r = ceph_calc_file_object_mapping(&ci->i_layout, dl.file_offset, &len, + &dl.object_no, &dl.object_offset, + &olen); + if (r < 0) + return -EIO; dl.file_offset -= dl.object_offset; dl.object_size = ceph_file_layout_object_size(ci->i_layout); dl.block_size = ceph_file_layout_su(ci->i_layout); diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index a5a735422aa7..1bcf712655d9 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -2625,7 +2625,8 @@ static void check_new_map(struct ceph_mds_client *mdsc, ceph_mdsmap_is_laggy(newmap, i) ? " (laggy)" : "", session_state_name(s->s_state)); - if (memcmp(ceph_mdsmap_get_addr(oldmap, i), + if (i >= newmap->m_max_mds || + memcmp(ceph_mdsmap_get_addr(oldmap, i), ceph_mdsmap_get_addr(newmap, i), sizeof(struct ceph_entity_addr))) { if (s->s_state == CEPH_MDS_SESSION_OPENING) { diff --git a/fs/ceph/super.c b/fs/ceph/super.c index 3a42d9326378..2eb43f211325 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -307,7 +307,10 @@ static int parse_mount_options(struct ceph_mount_options **pfsopt, { struct ceph_mount_options *fsopt; const char *dev_name_end; - int err = -ENOMEM; + int err; + + if (!dev_name || !*dev_name) + return -EINVAL; fsopt = kzalloc(sizeof(*fsopt), GFP_KERNEL); if (!fsopt) @@ -328,21 +331,33 @@ static int parse_mount_options(struct ceph_mount_options **pfsopt, fsopt->max_readdir_bytes = CEPH_MAX_READDIR_BYTES_DEFAULT; fsopt->congestion_kb = default_congestion_kb(); - /* ip1[:port1][,ip2[:port2]...]:/subdir/in/fs */ + /* + * Distinguish the server list from the path in "dev_name". + * Internally we do not include the leading '/' in the path. + * + * "dev_name" will look like: + * <server_spec>[,<server_spec>...]:[<path>] + * where + * <server_spec> is <ip>[:<port>] + * <path> is optional, but if present must begin with '/' + */ + dev_name_end = strchr(dev_name, '/'); + if (dev_name_end) { + /* skip over leading '/' for path */ + *path = dev_name_end + 1; + } else { + /* path is empty */ + dev_name_end = dev_name + strlen(dev_name); + *path = dev_name_end; + } err = -EINVAL; - if (!dev_name) - goto out; - *path = strstr(dev_name, ":/"); - if (*path == NULL) { - pr_err("device name is missing path (no :/ in %s)\n", + dev_name_end--; /* back up to ':' separator */ + if (*dev_name_end != ':') { + pr_err("device name is missing path (no : separator in %s)\n", dev_name); goto out; } - dev_name_end = *path; dout("device name '%.*s'\n", (int)(dev_name_end - dev_name), dev_name); - - /* path on server */ - *path += 2; dout("server path '%s'\n", *path); *popt = ceph_parse_options(options, dev_name, dev_name_end, diff --git a/fs/char_dev.c b/fs/char_dev.c index 3f152b92a94a..afc2bb691780 100644 --- a/fs/char_dev.c +++ b/fs/char_dev.c @@ -471,9 +471,19 @@ static int exact_lock(dev_t dev, void *data) */ int cdev_add(struct cdev *p, dev_t dev, unsigned count) { + int error; + p->dev = dev; p->count = count; - return kobj_map(cdev_map, dev, count, NULL, exact_match, exact_lock, p); + + error = kobj_map(cdev_map, dev, count, NULL, + exact_match, exact_lock, p); + if (error) + return error; + + kobject_get(p->kobj.parent); + + return 0; } static void cdev_unmap(dev_t dev, unsigned count) @@ -498,14 +508,20 @@ void cdev_del(struct cdev *p) static void cdev_default_release(struct kobject *kobj) { struct cdev *p = container_of(kobj, struct cdev, kobj); + struct kobject *parent = kobj->parent; + cdev_purge(p); + kobject_put(parent); } static void cdev_dynamic_release(struct kobject *kobj) { struct cdev *p = container_of(kobj, struct cdev, kobj); + struct kobject *parent = kobj->parent; + cdev_purge(p); kfree(p); + kobject_put(parent); } static struct kobj_type ktype_cdev_default = { diff --git a/fs/cifs/cifs_spnego.c b/fs/cifs/cifs_spnego.c index e622863b292f..086f381d6489 100644 --- a/fs/cifs/cifs_spnego.c +++ b/fs/cifs/cifs_spnego.c @@ -31,18 +31,18 @@ /* create a new cifs key */ static int -cifs_spnego_key_instantiate(struct key *key, const void *data, size_t datalen) +cifs_spnego_key_instantiate(struct key *key, struct key_preparsed_payload *prep) { char *payload; int ret; ret = -ENOMEM; - payload = kmalloc(datalen, GFP_KERNEL); + payload = kmalloc(prep->datalen, GFP_KERNEL); if (!payload) goto error; /* attach the data */ - memcpy(payload, data, datalen); + memcpy(payload, prep->data, prep->datalen); key->payload.data = payload; ret = 0; diff --git a/fs/cifs/cifs_unicode.c b/fs/cifs/cifs_unicode.c index 53cf2aabce87..71d5d0a5f6b2 100644 --- a/fs/cifs/cifs_unicode.c +++ b/fs/cifs/cifs_unicode.c @@ -203,6 +203,27 @@ cifs_strtoUTF16(__le16 *to, const char *from, int len, int i; wchar_t wchar_to; /* needed to quiet sparse */ + /* special case for utf8 to handle no plane0 chars */ + if (!strcmp(codepage->charset, "utf8")) { + /* + * convert utf8 -> utf16, we assume we have enough space + * as caller should have assumed conversion does not overflow + * in destination len is length in wchar_t units (16bits) + */ + i = utf8s_to_utf16s(from, len, UTF16_LITTLE_ENDIAN, + (wchar_t *) to, len); + + /* if success terminate and exit */ + if (i >= 0) + goto success; + /* + * if fails fall back to UCS encoding as this + * function should not return negative values + * currently can fail only if source contains + * invalid encoded characters + */ + } + for (i = 0; len && *from; i++, from += charlen, len -= charlen) { charlen = codepage->char2uni(from, len, &wchar_to); if (charlen < 1) { @@ -215,6 +236,7 @@ cifs_strtoUTF16(__le16 *to, const char *from, int len, put_unaligned_le16(wchar_to, &to[i]); } +success: put_unaligned_le16(0, &to[i]); return i; } diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index 2ee5c54797fa..fc783e264420 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -167,17 +167,17 @@ static struct shrinker cifs_shrinker = { }; static int -cifs_idmap_key_instantiate(struct key *key, const void *data, size_t datalen) +cifs_idmap_key_instantiate(struct key *key, struct key_preparsed_payload *prep) { char *payload; - payload = kmalloc(datalen, GFP_KERNEL); + payload = kmalloc(prep->datalen, GFP_KERNEL); if (!payload) return -ENOMEM; - memcpy(payload, data, datalen); + memcpy(payload, prep->data, prep->datalen); key->payload.data = payload; - key->datalen = datalen; + key->datalen = prep->datalen; return 0; } diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 2fdbe08a7a23..5c670b998ffb 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -67,6 +67,7 @@ enum { /* Mount options that take no arguments */ Opt_user_xattr, Opt_nouser_xattr, Opt_forceuid, Opt_noforceuid, + Opt_forcegid, Opt_noforcegid, Opt_noblocksend, Opt_noautotune, Opt_hard, Opt_soft, Opt_perm, Opt_noperm, Opt_mapchars, Opt_nomapchars, Opt_sfu, @@ -117,6 +118,8 @@ static const match_table_t cifs_mount_option_tokens = { { Opt_nouser_xattr, "nouser_xattr" }, { Opt_forceuid, "forceuid" }, { Opt_noforceuid, "noforceuid" }, + { Opt_forcegid, "forcegid" }, + { Opt_noforcegid, "noforcegid" }, { Opt_noblocksend, "noblocksend" }, { Opt_noautotune, "noautotune" }, { Opt_hard, "hard" }, @@ -1195,6 +1198,12 @@ cifs_parse_mount_options(const char *mountdata, const char *devname, case Opt_noforceuid: override_uid = 0; break; + case Opt_forcegid: + override_gid = 1; + break; + case Opt_noforcegid: + override_gid = 0; + break; case Opt_noblocksend: vol->noblocksnd = 1; break; diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 7d7bbdc4c8e7..edb25b4bbb95 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -3003,6 +3003,7 @@ cifs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) static struct vm_operations_struct cifs_file_vm_ops = { .fault = filemap_fault, .page_mkwrite = cifs_page_mkwrite, + .remap_pages = generic_file_remap_pages, }; int cifs_file_strict_mmap(struct file *file, struct vm_area_struct *vma) diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 2126ab185045..76d974c952fe 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -183,6 +183,12 @@ smb_send_kvec(struct TCP_Server_Info *server, struct kvec *iov, size_t n_vec, rc = kernel_sendmsg(ssocket, &smb_msg, &iov[first_vec], n_vec - first_vec, remaining); if (rc == -ENOSPC || rc == -EAGAIN) { + /* + * Catch if a low level driver returns -ENOSPC. This + * WARN_ON will be removed by 3.10 if no one reports + * seeing this. + */ + WARN_ON_ONCE(rc == -ENOSPC); i++; if (i >= 14 || (!server->noblocksnd && (i > 2))) { cERROR(1, "sends on sock %p stuck for 15 " diff --git a/fs/compat.c b/fs/compat.c index b7a24d0ca30d..015e1e1f87c6 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -776,16 +776,16 @@ asmlinkage long compat_sys_mount(const char __user * dev_name, char *kernel_type; unsigned long data_page; char *kernel_dev; - char *dir_page; + struct filename *dir; int retval; retval = copy_mount_string(type, &kernel_type); if (retval < 0) goto out; - dir_page = getname(dir_name); - retval = PTR_ERR(dir_page); - if (IS_ERR(dir_page)) + dir = getname(dir_name); + retval = PTR_ERR(dir); + if (IS_ERR(dir)) goto out1; retval = copy_mount_string(dev_name, &kernel_dev); @@ -807,7 +807,7 @@ asmlinkage long compat_sys_mount(const char __user * dev_name, } } - retval = do_mount(kernel_dev, dir_page, kernel_type, + retval = do_mount(kernel_dev, dir->name, kernel_type, flags, (void*)data_page); out4: @@ -815,7 +815,7 @@ asmlinkage long compat_sys_mount(const char __user * dev_name, out3: kfree(kernel_dev); out2: - putname(dir_page); + putname(dir); out1: kfree(kernel_type); out: diff --git a/fs/compat_binfmt_elf.c b/fs/compat_binfmt_elf.c index 112e45a17e99..a81147e2e4ef 100644 --- a/fs/compat_binfmt_elf.c +++ b/fs/compat_binfmt_elf.c @@ -38,6 +38,13 @@ #define elf_addr_t Elf32_Addr /* + * Some data types as stored in coredump. + */ +#define user_long_t compat_long_t +#define user_siginfo_t compat_siginfo_t +#define copy_siginfo_to_user copy_siginfo_to_user32 + +/* * The machine-dependent core note format types are defined in elfcore-compat.h, * which requires asm/elf.h to define compat_elf_gregset_t et al. */ diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index f5054025f9da..4c6285fff598 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -210,6 +210,8 @@ static int do_video_set_spu_palette(unsigned int fd, unsigned int cmd, err = get_user(palp, &up->palette); err |= get_user(length, &up->length); + if (err) + return -EFAULT; up_native = compat_alloc_user_space(sizeof(struct video_spu_palette)); err = put_user(compat_ptr(palp), &up_native->palette); diff --git a/fs/coredump.c b/fs/coredump.c index f045bbad6822..ce47379bfa61 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -14,6 +14,7 @@ #include <linux/key.h> #include <linux/personality.h> #include <linux/binfmts.h> +#include <linux/coredump.h> #include <linux/utsname.h> #include <linux/pid_namespace.h> #include <linux/module.h> @@ -39,6 +40,7 @@ #include <trace/events/task.h> #include "internal.h" +#include "coredump.h" #include <trace/events/sched.h> @@ -147,7 +149,7 @@ put_exe_file: * name into corename, which must have space for at least * CORENAME_MAX_SIZE bytes plus one byte for the zero terminator. */ -static int format_corename(struct core_name *cn, long signr) +static int format_corename(struct core_name *cn, struct coredump_params *cprm) { const struct cred *cred = current_cred(); const char *pat_ptr = core_pattern; @@ -192,9 +194,13 @@ static int format_corename(struct core_name *cn, long signr) case 'g': err = cn_printf(cn, "%d", cred->gid); break; + case 'd': + err = cn_printf(cn, "%d", + __get_dumpable(cprm->mm_flags)); + break; /* signal that caused the coredump */ case 's': - err = cn_printf(cn, "%ld", signr); + err = cn_printf(cn, "%ld", cprm->siginfo->si_signo); break; /* UNIX time of coredump */ case 't': { @@ -444,14 +450,15 @@ static int umh_pipe_setup(struct subprocess_info *info, struct cred *new) cp->file = files[1]; - replace_fd(0, files[0], 0); + err = replace_fd(0, files[0], 0); + fput(files[0]); /* and disallow core files too */ current->signal->rlim[RLIMIT_CORE] = (struct rlimit){1, 1}; - return 0; + return err; } -void do_coredump(long signr, int exit_code, struct pt_regs *regs) +void do_coredump(siginfo_t *siginfo, struct pt_regs *regs) { struct core_state core_state; struct core_name cn; @@ -466,7 +473,7 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) bool need_nonrelative = false; static atomic_t core_dump_count = ATOMIC_INIT(0); struct coredump_params cprm = { - .signr = signr, + .siginfo = siginfo, .regs = regs, .limit = rlimit(RLIMIT_CORE), /* @@ -477,7 +484,7 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) .mm_flags = mm->flags, }; - audit_core_dumps(signr); + audit_core_dumps(siginfo->si_signo); binfmt = mm->binfmt; if (!binfmt || !binfmt->core_dump) @@ -501,7 +508,7 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) need_nonrelative = true; } - retval = coredump_wait(exit_code, &core_state); + retval = coredump_wait(siginfo->si_signo, &core_state); if (retval < 0) goto fail_creds; @@ -513,7 +520,7 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) */ clear_thread_flag(TIF_SIGPENDING); - ispipe = format_corename(&cn, signr); + ispipe = format_corename(&cn, &cprm); if (ispipe) { int dump_count; diff --git a/fs/coredump.h b/fs/coredump.h new file mode 100644 index 000000000000..e39ff072110d --- /dev/null +++ b/fs/coredump.h @@ -0,0 +1,6 @@ +#ifndef _FS_COREDUMP_H +#define _FS_COREDUMP_H + +extern int __get_dumpable(unsigned long mm_flags); + +#endif diff --git a/fs/eventpoll.c b/fs/eventpoll.c index cd96649bfe62..da72250ddc1c 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -346,7 +346,7 @@ static inline struct epitem *ep_item_from_epqueue(poll_table *p) /* Tells if the epoll_ctl(2) operation needs an event copy from userspace */ static inline int ep_op_has_event(int op) { - return op != EPOLL_CTL_DEL; + return op == EPOLL_CTL_ADD || op == EPOLL_CTL_MOD; } /* Initialize the poll safe wake up structure */ @@ -676,6 +676,34 @@ static int ep_remove(struct eventpoll *ep, struct epitem *epi) return 0; } +/* + * Disables a "struct epitem" in the eventpoll set. Returns -EBUSY if the item + * had no event flags set, indicating that another thread may be currently + * handling that item's events (in the case that EPOLLONESHOT was being + * used). Otherwise a zero result indicates that the item has been disabled + * from receiving events. A disabled item may be re-enabled via + * EPOLL_CTL_MOD. Must be called with "mtx" held. + */ +static int ep_disable(struct eventpoll *ep, struct epitem *epi) +{ + int result = 0; + unsigned long flags; + + spin_lock_irqsave(&ep->lock, flags); + if (epi->event.events & ~EP_PRIVATE_BITS) { + if (ep_is_linked(&epi->rdllink)) + list_del_init(&epi->rdllink); + /* Ensure ep_poll_callback will not add epi back onto ready + list: */ + epi->event.events &= EP_PRIVATE_BITS; + } + else + result = -EBUSY; + spin_unlock_irqrestore(&ep->lock, flags); + + return result; +} + static void ep_free(struct eventpoll *ep) { struct rb_node *rbp; @@ -1020,8 +1048,6 @@ static void ep_rbtree_insert(struct eventpoll *ep, struct epitem *epi) rb_insert_color(&epi->rbn, &ep->rbr); } - - #define PATH_ARR_SIZE 5 /* * These are the number paths of length 1 to 5, that we are allowing to emanate @@ -1787,6 +1813,12 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd, } else error = -ENOENT; break; + case EPOLL_CTL_DISABLE: + if (epi) + error = ep_disable(ep, epi); + else + error = -ENOENT; + break; } mutex_unlock(&ep->mtx); diff --git a/fs/exec.c b/fs/exec.c index 48fb26ef8a1b..0039055b1fc6 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -59,10 +59,10 @@ #include <asm/uaccess.h> #include <asm/mmu_context.h> #include <asm/tlb.h> -#include <asm/exec.h> #include <trace/events/task.h> #include "internal.h" +#include "coredump.h" #include <trace/events/sched.h> @@ -105,7 +105,7 @@ static inline void put_binfmt(struct linux_binfmt * fmt) SYSCALL_DEFINE1(uselib, const char __user *, library) { struct file *file; - char *tmp = getname(library); + struct filename *tmp = getname(library); int error = PTR_ERR(tmp); static const struct open_flags uselib_flags = { .open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, @@ -391,7 +391,7 @@ struct user_arg_ptr { union { const char __user *const __user *native; #ifdef CONFIG_COMPAT - compat_uptr_t __user *compat; + const compat_uptr_t __user *compat; #endif } ptr; }; @@ -602,7 +602,7 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) * process cleanup to remove whatever mess we made. */ if (length != move_page_tables(vma, old_start, - vma, new_start, length)) + vma, new_start, length, false)) return -ENOMEM; lru_add_drain(); @@ -751,13 +751,14 @@ struct file *open_exec(const char *name) { struct file *file; int err; + struct filename tmp = { .name = name }; static const struct open_flags open_exec_flags = { .open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, .acc_mode = MAY_EXEC | MAY_OPEN, .intent = LOOKUP_OPEN }; - file = do_filp_open(AT_FDCWD, name, &open_exec_flags, LOOKUP_FOLLOW); + file = do_filp_open(AT_FDCWD, &tmp, &open_exec_flags, LOOKUP_FOLLOW); if (IS_ERR(file)) goto out; @@ -877,9 +878,11 @@ static int de_thread(struct task_struct *tsk) sig->notify_count--; while (sig->notify_count) { - __set_current_state(TASK_UNINTERRUPTIBLE); + __set_current_state(TASK_KILLABLE); spin_unlock_irq(lock); schedule(); + if (unlikely(__fatal_signal_pending(tsk))) + goto killed; spin_lock_irq(lock); } spin_unlock_irq(lock); @@ -897,9 +900,11 @@ static int de_thread(struct task_struct *tsk) write_lock_irq(&tasklist_lock); if (likely(leader->exit_state)) break; - __set_current_state(TASK_UNINTERRUPTIBLE); + __set_current_state(TASK_KILLABLE); write_unlock_irq(&tasklist_lock); schedule(); + if (unlikely(__fatal_signal_pending(tsk))) + goto killed; } /* @@ -993,6 +998,14 @@ no_thread_group: BUG_ON(!thread_group_leader(tsk)); return 0; + +killed: + /* protects against exit_notify() and __exit_signal() */ + read_lock(&tasklist_lock); + sig->group_exit_task = NULL; + sig->notify_count = 0; + read_unlock(&tasklist_lock); + return -EAGAIN; } char *get_task_comm(char *buf, struct task_struct *tsk) @@ -1070,7 +1083,8 @@ int flush_old_exec(struct linux_binprm * bprm) bprm->mm = NULL; /* We're using it now */ set_fs(USER_DS); - current->flags &= ~(PF_RANDOMIZE | PF_FORKNOEXEC | PF_KTHREAD); + current->flags &= + ~(PF_RANDOMIZE | PF_FORKNOEXEC | PF_KTHREAD | PF_NOFREEZE); flush_thread(); current->personality &= ~bprm->per_clear; @@ -1096,7 +1110,7 @@ void setup_new_exec(struct linux_binprm * bprm) current->sas_ss_sp = current->sas_ss_size = 0; if (uid_eq(current_euid(), current_uid()) && gid_eq(current_egid(), current_gid())) - set_dumpable(current->mm, 1); + set_dumpable(current->mm, SUID_DUMPABLE_ENABLED); else set_dumpable(current->mm, suid_dumpable); @@ -1561,9 +1575,9 @@ int do_execve(const char *filename, } #ifdef CONFIG_COMPAT -int compat_do_execve(char *filename, - compat_uptr_t __user *__argv, - compat_uptr_t __user *__envp, +int compat_do_execve(const char *filename, + const compat_uptr_t __user *__argv, + const compat_uptr_t __user *__envp, struct pt_regs *regs) { struct user_arg_ptr argv = { @@ -1645,3 +1659,56 @@ int get_dumpable(struct mm_struct *mm) { return __get_dumpable(mm->flags); } + +#ifdef __ARCH_WANT_SYS_EXECVE +SYSCALL_DEFINE3(execve, + const char __user *, filename, + const char __user *const __user *, argv, + const char __user *const __user *, envp) +{ + struct filename *path = getname(filename); + int error = PTR_ERR(path); + if (!IS_ERR(path)) { + error = do_execve(path->name, argv, envp, current_pt_regs()); + putname(path); + } + return error; +} +#ifdef CONFIG_COMPAT +asmlinkage long compat_sys_execve(const char __user * filename, + const compat_uptr_t __user * argv, + const compat_uptr_t __user * envp) +{ + struct filename *path = getname(filename); + int error = PTR_ERR(path); + if (!IS_ERR(path)) { + error = compat_do_execve(path->name, argv, envp, + current_pt_regs()); + putname(path); + } + return error; +} +#endif +#endif + +#ifdef __ARCH_WANT_KERNEL_EXECVE +int kernel_execve(const char *filename, + const char *const argv[], + const char *const envp[]) +{ + struct pt_regs *p = current_pt_regs(); + int ret; + + ret = do_execve(filename, + (const char __user *const __user *)argv, + (const char __user *const __user *)envp, p); + if (ret < 0) + return ret; + + /* + * We were successful. We won't be returning to our caller, but + * instead to user space by manipulating the kernel stack. + */ + ret_from_kernel_execve(p); +} +#endif diff --git a/fs/exofs/ore.c b/fs/exofs/ore.c index 1585db1aa365..f936cb50dc0d 100644 --- a/fs/exofs/ore.c +++ b/fs/exofs/ore.c @@ -814,8 +814,8 @@ static int _write_mirror(struct ore_io_state *ios, int cur_comp) struct bio *bio; if (per_dev != master_dev) { - bio = bio_kmalloc(GFP_KERNEL, - master_dev->bio->bi_max_vecs); + bio = bio_clone_kmalloc(master_dev->bio, + GFP_KERNEL); if (unlikely(!bio)) { ORE_DBGMSG( "Failed to allocate BIO size=%u\n", @@ -824,7 +824,6 @@ static int _write_mirror(struct ore_io_state *ios, int cur_comp) goto out; } - __bio_clone(bio, master_dev->bio); bio->bi_bdev = NULL; bio->bi_next = NULL; per_dev->offset = master_dev->offset; diff --git a/fs/exofs/ore_raid.c b/fs/exofs/ore_raid.c index 5f376d14fdcc..b963f38ac298 100644 --- a/fs/exofs/ore_raid.c +++ b/fs/exofs/ore_raid.c @@ -203,7 +203,7 @@ static unsigned _sp2d_min_pg(struct __stripe_pages_2d *sp2d) static unsigned _sp2d_max_pg(struct __stripe_pages_2d *sp2d) { - unsigned p; + int p; for (p = sp2d->pages_in_unit - 1; p >= 0; --p) { struct __1_page_stripe *_1ps = &sp2d->_1p_stripes[p]; diff --git a/fs/exofs/super.c b/fs/exofs/super.c index 59e3bbfac0b1..5e59280d42d7 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -389,8 +389,6 @@ static int exofs_sync_fs(struct super_block *sb, int wait) if (unlikely(ret)) goto out; - lock_super(sb); - ios->length = offsetof(struct exofs_fscb, s_dev_table_oid); memset(fscb, 0, ios->length); fscb->s_nextid = cpu_to_le64(sbi->s_nextid); @@ -406,8 +404,6 @@ static int exofs_sync_fs(struct super_block *sb, int wait) if (unlikely(ret)) EXOFS_ERR("%s: ore_write failed.\n", __func__); - - unlock_super(sb); out: EXOFS_DBGMSG("s_nextid=0x%llx ret=%d\n", _LLU(sbi->s_nextid), ret); ore_put_io_state(ios); diff --git a/fs/exofs/sys.c b/fs/exofs/sys.c index 5a7b691e748b..1b4f2f95fc37 100644 --- a/fs/exofs/sys.c +++ b/fs/exofs/sys.c @@ -80,8 +80,13 @@ static ssize_t uri_show(struct exofs_dev *edp, char *buf) static ssize_t uri_store(struct exofs_dev *edp, const char *buf, size_t len) { + uint8_t *new_uri; + edp->urilen = strlen(buf) + 1; - edp->uri = krealloc(edp->uri, edp->urilen, GFP_KERNEL); + new_uri = krealloc(edp->uri, edp->urilen, GFP_KERNEL); + if (new_uri == NULL) + return -ENOMEM; + edp->uri = new_uri; strncpy(edp->uri, buf, edp->urilen); return edp->urilen; } diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 6c205d0c565b..fa04d023177e 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -469,7 +469,7 @@ static int parse_options(char *options, struct super_block *sb) uid = make_kuid(current_user_ns(), option); if (!uid_valid(uid)) { ext2_msg(sb, KERN_ERR, "Invalid uid value %d", option); - return -1; + return 0; } sbi->s_resuid = uid; @@ -480,7 +480,7 @@ static int parse_options(char *options, struct super_block *sb) gid = make_kgid(current_user_ns(), option); if (!gid_valid(gid)) { ext2_msg(sb, KERN_ERR, "Invalid gid value %d", option); - return -1; + return 0; } sbi->s_resgid = gid; break; diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index 8f4fddac01a6..890b8947c546 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -46,8 +46,7 @@ static struct buffer_head *ext3_append(handle_t *handle, *block = inode->i_size >> inode->i_sb->s_blocksize_bits; - bh = ext3_bread(handle, inode, *block, 1, err); - if (bh) { + if ((bh = ext3_dir_bread(handle, inode, *block, 1, err))) { inode->i_size += inode->i_sb->s_blocksize; EXT3_I(inode)->i_disksize = inode->i_size; *err = ext3_journal_get_write_access(handle, bh); @@ -339,8 +338,10 @@ dx_probe(struct qstr *entry, struct inode *dir, u32 hash; frame->bh = NULL; - if (!(bh = ext3_bread (NULL,dir, 0, 0, err))) + if (!(bh = ext3_dir_bread(NULL, dir, 0, 0, err))) { + *err = ERR_BAD_DX_DIR; goto fail; + } root = (struct dx_root *) bh->b_data; if (root->info.hash_version != DX_HASH_TEA && root->info.hash_version != DX_HASH_HALF_MD4 && @@ -436,8 +437,10 @@ dx_probe(struct qstr *entry, struct inode *dir, frame->entries = entries; frame->at = at; if (!indirect--) return frame; - if (!(bh = ext3_bread (NULL,dir, dx_get_block(at), 0, err))) + if (!(bh = ext3_dir_bread(NULL, dir, dx_get_block(at), 0, err))) { + *err = ERR_BAD_DX_DIR; goto fail2; + } at = entries = ((struct dx_node *) bh->b_data)->entries; if (dx_get_limit(entries) != dx_node_limit (dir)) { ext3_warning(dir->i_sb, __func__, @@ -535,8 +538,8 @@ static int ext3_htree_next_block(struct inode *dir, __u32 hash, * block so no check is necessary */ while (num_frames--) { - if (!(bh = ext3_bread(NULL, dir, dx_get_block(p->at), - 0, &err))) + if (!(bh = ext3_dir_bread(NULL, dir, dx_get_block(p->at), + 0, &err))) return err; /* Failure */ p++; brelse (p->bh); @@ -559,10 +562,11 @@ static int htree_dirblock_to_tree(struct file *dir_file, { struct buffer_head *bh; struct ext3_dir_entry_2 *de, *top; - int err, count = 0; + int err = 0, count = 0; dxtrace(printk("In htree dirblock_to_tree: block %d\n", block)); - if (!(bh = ext3_bread (NULL, dir, block, 0, &err))) + + if (!(bh = ext3_dir_bread(NULL, dir, block, 0, &err))) return err; de = (struct ext3_dir_entry_2 *) bh->b_data; @@ -976,7 +980,7 @@ static struct buffer_head * ext3_dx_find_entry(struct inode *dir, return NULL; do { block = dx_get_block(frame->at); - if (!(bh = ext3_bread (NULL,dir, block, 0, err))) + if (!(bh = ext3_dir_bread (NULL, dir, block, 0, err))) goto errout; retval = search_dirblock(bh, dir, entry, @@ -1458,9 +1462,9 @@ static int ext3_add_entry (handle_t *handle, struct dentry *dentry, } blocks = dir->i_size >> sb->s_blocksize_bits; for (block = 0; block < blocks; block++) { - bh = ext3_bread(handle, dir, block, 0, &retval); - if(!bh) + if (!(bh = ext3_dir_bread(handle, dir, block, 0, &retval))) return retval; + retval = add_dirent_to_buf(handle, dentry, inode, NULL, bh); if (retval != -ENOSPC) return retval; @@ -1500,7 +1504,7 @@ static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, entries = frame->entries; at = frame->at; - if (!(bh = ext3_bread(handle,dir, dx_get_block(frame->at), 0, &err))) + if (!(bh = ext3_dir_bread(handle, dir, dx_get_block(frame->at), 0, &err))) goto cleanup; BUFFER_TRACE(bh, "get_write_access"); @@ -1790,8 +1794,7 @@ retry: inode->i_op = &ext3_dir_inode_operations; inode->i_fop = &ext3_dir_operations; inode->i_size = EXT3_I(inode)->i_disksize = inode->i_sb->s_blocksize; - dir_block = ext3_bread (handle, inode, 0, 1, &err); - if (!dir_block) + if (!(dir_block = ext3_dir_bread(handle, inode, 0, 1, &err))) goto out_clear_inode; BUFFER_TRACE(dir_block, "get_write_access"); @@ -1859,7 +1862,7 @@ static int empty_dir (struct inode * inode) sb = inode->i_sb; if (inode->i_size < EXT3_DIR_REC_LEN(1) + EXT3_DIR_REC_LEN(2) || - !(bh = ext3_bread (NULL, inode, 0, 0, &err))) { + !(bh = ext3_dir_bread(NULL, inode, 0, 0, &err))) { if (err) ext3_error(inode->i_sb, __func__, "error %d reading directory #%lu offset 0", @@ -1890,9 +1893,8 @@ static int empty_dir (struct inode * inode) (void *) de >= (void *) (bh->b_data+sb->s_blocksize)) { err = 0; brelse (bh); - bh = ext3_bread (NULL, inode, - offset >> EXT3_BLOCK_SIZE_BITS(sb), 0, &err); - if (!bh) { + if (!(bh = ext3_dir_bread (NULL, inode, + offset >> EXT3_BLOCK_SIZE_BITS(sb), 0, &err))) { if (err) ext3_error(sb, __func__, "error %d reading directory" @@ -2388,7 +2390,7 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry, goto end_rename; } retval = -EIO; - dir_bh = ext3_bread (handle, old_inode, 0, 0, &retval); + dir_bh = ext3_dir_bread(handle, old_inode, 0, 0, &retval); if (!dir_bh) goto end_rename; if (le32_to_cpu(PARENT_INO(dir_bh->b_data)) != old_dir->i_ino) diff --git a/fs/ext3/namei.h b/fs/ext3/namei.h index f2ce2b0065c9..46304d8c9f0a 100644 --- a/fs/ext3/namei.h +++ b/fs/ext3/namei.h @@ -6,3 +6,22 @@ */ extern struct dentry *ext3_get_parent(struct dentry *child); + +static inline struct buffer_head *ext3_dir_bread(handle_t *handle, + struct inode *inode, + int block, int create, + int *err) +{ + struct buffer_head *bh; + + bh = ext3_bread(handle, inode, block, create, err); + + if (!bh && !(*err)) { + *err = -EIO; + ext3_error(inode->i_sb, __func__, + "Directory hole detected on inode %lu\n", + inode->i_ino); + return NULL; + } + return bh; +} diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 17ae5c83d234..5366393528df 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -1001,7 +1001,7 @@ static int parse_options (char *options, struct super_block *sb, uid = make_kuid(current_user_ns(), option); if (!uid_valid(uid)) { ext3_msg(sb, KERN_ERR, "Invalid uid value %d", option); - return -1; + return 0; } sbi->s_resuid = uid; @@ -1012,7 +1012,7 @@ static int parse_options (char *options, struct super_block *sb, gid = make_kgid(current_user_ns(), option); if (!gid_valid(gid)) { ext3_msg(sb, KERN_ERR, "Invalid gid value %d", option); - return -1; + return 0; } sbi->s_resgid = gid; break; @@ -2578,11 +2578,9 @@ out: static int ext3_unfreeze(struct super_block *sb) { if (!(sb->s_flags & MS_RDONLY)) { - lock_super(sb); /* Reser the needs_recovery flag before the fs is unlocked. */ EXT3_SET_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); ext3_commit_super(sb, EXT3_SB(sb)->s_es, 1); - unlock_super(sb); journal_unlock_updates(EXT3_SB(sb)->s_journal); } return 0; @@ -2602,7 +2600,6 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data) #endif /* Store the original options */ - lock_super(sb); old_sb_flags = sb->s_flags; old_opts.s_mount_opt = sbi->s_mount_opt; old_opts.s_resuid = sbi->s_resuid; @@ -2708,8 +2705,6 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data) old_opts.s_qf_names[i] != sbi->s_qf_names[i]) kfree(old_opts.s_qf_names[i]); #endif - unlock_super(sb); - if (enable_quota) dquot_resume(sb, -1); return 0; @@ -2728,7 +2723,6 @@ restore_opts: sbi->s_qf_names[i] = old_opts.s_qf_names[i]; } #endif - unlock_super(sb); return err; } diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 1b5089067d01..cf1821784a16 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -174,8 +174,7 @@ void ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, ext4_free_inodes_set(sb, gdp, 0); ext4_itable_unused_set(sb, gdp, 0); memset(bh->b_data, 0xff, sb->s_blocksize); - ext4_block_bitmap_csum_set(sb, block_group, gdp, bh, - EXT4_BLOCKS_PER_GROUP(sb) / 8); + ext4_block_bitmap_csum_set(sb, block_group, gdp, bh); return; } memset(bh->b_data, 0, sb->s_blocksize); @@ -212,8 +211,7 @@ void ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, */ ext4_mark_bitmap_end(num_clusters_in_group(sb, block_group), sb->s_blocksize * 8, bh->b_data); - ext4_block_bitmap_csum_set(sb, block_group, gdp, bh, - EXT4_BLOCKS_PER_GROUP(sb) / 8); + ext4_block_bitmap_csum_set(sb, block_group, gdp, bh); ext4_group_desc_csum_set(sb, block_group, gdp); } @@ -350,7 +348,7 @@ void ext4_validate_block_bitmap(struct super_block *sb, return; } if (unlikely(!ext4_block_bitmap_csum_verify(sb, block_group, - desc, bh, EXT4_BLOCKS_PER_GROUP(sb) / 8))) { + desc, bh))) { ext4_unlock_group(sb, block_group); ext4_error(sb, "bg %u: bad block bitmap checksum", block_group); return; diff --git a/fs/ext4/bitmap.c b/fs/ext4/bitmap.c index 5c2d1813ebe9..3285aa5a706a 100644 --- a/fs/ext4/bitmap.c +++ b/fs/ext4/bitmap.c @@ -58,11 +58,12 @@ void ext4_inode_bitmap_csum_set(struct super_block *sb, ext4_group_t group, int ext4_block_bitmap_csum_verify(struct super_block *sb, ext4_group_t group, struct ext4_group_desc *gdp, - struct buffer_head *bh, int sz) + struct buffer_head *bh) { __u32 hi; __u32 provided, calculated; struct ext4_sb_info *sbi = EXT4_SB(sb); + int sz = EXT4_CLUSTERS_PER_GROUP(sb) / 8; if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) @@ -84,8 +85,9 @@ int ext4_block_bitmap_csum_verify(struct super_block *sb, ext4_group_t group, void ext4_block_bitmap_csum_set(struct super_block *sb, ext4_group_t group, struct ext4_group_desc *gdp, - struct buffer_head *bh, int sz) + struct buffer_head *bh) { + int sz = EXT4_CLUSTERS_PER_GROUP(sb) / 8; __u32 csum; struct ext4_sb_info *sbi = EXT4_SB(sb); diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index c3411d4ce2da..3c20de1d59d0 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -186,7 +186,6 @@ struct mpage_da_data { #define EXT4_IO_END_ERROR 0x0002 #define EXT4_IO_END_QUEUED 0x0004 #define EXT4_IO_END_DIRECT 0x0008 -#define EXT4_IO_END_IN_FSYNC 0x0010 struct ext4_io_page { struct page *p_page; @@ -912,9 +911,7 @@ struct ext4_inode_info { struct list_head i_completed_io_list; spinlock_t i_completed_io_lock; atomic_t i_ioend_count; /* Number of outstanding io_end structs */ - /* current io_end structure for async DIO write*/ - ext4_io_end_t *cur_aio_dio; - atomic_t i_aiodio_unwritten; /* Nr. of inflight conversions pending */ + atomic_t i_unwritten; /* Nr. of inflight conversions pending */ spinlock_t i_block_reservation_lock; @@ -1233,6 +1230,7 @@ struct ext4_sb_info { spinlock_t s_md_lock; unsigned short *s_mb_offsets; unsigned int *s_mb_maxs; + unsigned int s_group_info_size; /* tunables */ unsigned long s_stripe; @@ -1243,6 +1241,7 @@ struct ext4_sb_info { unsigned int s_mb_order2_reqs; unsigned int s_mb_group_prealloc; unsigned int s_max_writeback_mb_bump; + unsigned int s_max_dir_size_kb; /* where last allocation was done - for stream allocation */ unsigned long s_mb_last_group; unsigned long s_mb_last_start; @@ -1270,8 +1269,12 @@ struct ext4_sb_info { unsigned long s_sectors_written_start; u64 s_kbytes_written; + /* the size of zero-out chunk */ + unsigned int s_extent_max_zeroout_kb; + unsigned int s_log_groups_per_flex; struct flex_groups *s_flex_groups; + ext4_group_t s_flex_groups_allocated; /* workqueue for dio unwritten */ struct workqueue_struct *dio_unwritten_wq; @@ -1328,10 +1331,20 @@ static inline void ext4_set_io_unwritten_flag(struct inode *inode, { if (!(io_end->flag & EXT4_IO_END_UNWRITTEN)) { io_end->flag |= EXT4_IO_END_UNWRITTEN; - atomic_inc(&EXT4_I(inode)->i_aiodio_unwritten); + atomic_inc(&EXT4_I(inode)->i_unwritten); } } +static inline ext4_io_end_t *ext4_inode_aio(struct inode *inode) +{ + return inode->i_private; +} + +static inline void ext4_inode_aio_set(struct inode *inode, ext4_io_end_t *io) +{ + inode->i_private = io; +} + /* * Inode dynamic state flags */ @@ -1345,6 +1358,8 @@ enum { EXT4_STATE_DIO_UNWRITTEN, /* need convert on dio done*/ EXT4_STATE_NEWENTRY, /* File just added to dir */ EXT4_STATE_DELALLOC_RESERVED, /* blks already reserved for delalloc */ + EXT4_STATE_DIOREAD_LOCK, /* Disable support for dio read + nolocking */ }; #define EXT4_INODE_BIT_FNS(name, field, offset) \ @@ -1867,10 +1882,10 @@ int ext4_inode_bitmap_csum_verify(struct super_block *sb, ext4_group_t group, struct buffer_head *bh, int sz); void ext4_block_bitmap_csum_set(struct super_block *sb, ext4_group_t group, struct ext4_group_desc *gdp, - struct buffer_head *bh, int sz); + struct buffer_head *bh); int ext4_block_bitmap_csum_verify(struct super_block *sb, ext4_group_t group, struct ext4_group_desc *gdp, - struct buffer_head *bh, int sz); + struct buffer_head *bh); /* balloc.c */ extern void ext4_validate_block_bitmap(struct super_block *sb, @@ -1932,7 +1947,7 @@ extern void ext4_htree_free_dir_info(struct dir_private_info *p); /* fsync.c */ extern int ext4_sync_file(struct file *, loff_t, loff_t, int); -extern int ext4_flush_completed_IO(struct inode *); +extern int ext4_flush_unwritten_io(struct inode *); /* hash.c */ extern int ext4fs_dirhash(const char *name, int len, struct @@ -1966,6 +1981,8 @@ extern void ext4_exit_mballoc(void); extern void ext4_free_blocks(handle_t *handle, struct inode *inode, struct buffer_head *bh, ext4_fsblk_t block, unsigned long count, int flags); +extern int ext4_mb_alloc_groupinfo(struct super_block *sb, + ext4_group_t ngroups); extern int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t i, struct ext4_group_desc *desc); extern int ext4_group_add_blocks(handle_t *handle, struct super_block *sb, @@ -2046,11 +2063,12 @@ extern int ext4_resize_fs(struct super_block *sb, ext4_fsblk_t n_blocks_count); extern int ext4_calculate_overhead(struct super_block *sb); extern int ext4_superblock_csum_verify(struct super_block *sb, struct ext4_super_block *es); -extern void ext4_superblock_csum_set(struct super_block *sb, - struct ext4_super_block *es); +extern void ext4_superblock_csum_set(struct super_block *sb); extern void *ext4_kvmalloc(size_t size, gfp_t flags); extern void *ext4_kvzalloc(size_t size, gfp_t flags); extern void ext4_kvfree(void *ptr); +extern int ext4_alloc_flex_bg_array(struct super_block *sb, + ext4_group_t ngroup); extern __printf(4, 5) void __ext4_error(struct super_block *, const char *, unsigned int, const char *, ...); @@ -2352,6 +2370,7 @@ extern const struct file_operations ext4_dir_operations; extern const struct inode_operations ext4_file_inode_operations; extern const struct file_operations ext4_file_operations; extern loff_t ext4_llseek(struct file *file, loff_t offset, int origin); +extern void ext4_unwritten_wait(struct inode *inode); /* namei.c */ extern const struct inode_operations ext4_dir_inode_operations; @@ -2400,11 +2419,11 @@ extern int ext4_move_extents(struct file *o_filp, struct file *d_filp, /* page-io.c */ extern int __init ext4_init_pageio(void); +extern void ext4_add_complete_io(ext4_io_end_t *io_end); extern void ext4_exit_pageio(void); extern void ext4_ioend_wait(struct inode *); extern void ext4_free_io_end(ext4_io_end_t *io); extern ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags); -extern int ext4_end_io_nolock(ext4_io_end_t *io); extern void ext4_io_submit(struct ext4_io_submit *io); extern int ext4_bio_write_page(struct ext4_io_submit *io, struct page *page, @@ -2452,6 +2471,21 @@ static inline void set_bitmap_uptodate(struct buffer_head *bh) set_bit(BH_BITMAP_UPTODATE, &(bh)->b_state); } +/* + * Disable DIO read nolock optimization, so new dioreaders will be forced + * to grab i_mutex + */ +static inline void ext4_inode_block_unlocked_dio(struct inode *inode) +{ + ext4_set_inode_state(inode, EXT4_STATE_DIOREAD_LOCK); + smp_mb(); +} +static inline void ext4_inode_resume_unlocked_dio(struct inode *inode) +{ + smp_mb(); + ext4_clear_inode_state(inode, EXT4_STATE_DIOREAD_LOCK); +} + #define in_range(b, first, len) ((b) >= (first) && (b) <= (first) + (len) - 1) /* For ioend & aio unwritten conversion wait queues */ diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c index bfa65b49d424..b4323ba846b5 100644 --- a/fs/ext4/ext4_jbd2.c +++ b/fs/ext4/ext4_jbd2.c @@ -143,17 +143,13 @@ int __ext4_handle_dirty_super(const char *where, unsigned int line, struct buffer_head *bh = EXT4_SB(sb)->s_sbh; int err = 0; + ext4_superblock_csum_set(sb); if (ext4_handle_valid(handle)) { - ext4_superblock_csum_set(sb, - (struct ext4_super_block *)bh->b_data); err = jbd2_journal_dirty_metadata(handle, bh); if (err) ext4_journal_abort_handle(where, line, __func__, bh, handle, err); - } else { - ext4_superblock_csum_set(sb, - (struct ext4_super_block *)bh->b_data); + } else mark_buffer_dirty(bh); - } return err; } diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index aabbb3f53683..7011ac967208 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -52,6 +52,9 @@ #define EXT4_EXT_MARK_UNINIT1 0x2 /* mark first half uninitialized */ #define EXT4_EXT_MARK_UNINIT2 0x4 /* mark second half uninitialized */ +#define EXT4_EXT_DATA_VALID1 0x8 /* first half contains valid data */ +#define EXT4_EXT_DATA_VALID2 0x10 /* second half contains valid data */ + static __le32 ext4_extent_block_csum(struct inode *inode, struct ext4_extent_header *eh) { @@ -1177,7 +1180,7 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, le32_to_cpu(EXT_FIRST_INDEX(neh)->ei_block), ext4_idx_pblock(EXT_FIRST_INDEX(neh))); - neh->eh_depth = cpu_to_le16(le16_to_cpu(neh->eh_depth) + 1); + le16_add_cpu(&neh->eh_depth, 1); ext4_mark_inode_dirty(handle, inode); out: brelse(bh); @@ -1656,16 +1659,60 @@ static int ext4_ext_try_to_merge_right(struct inode *inode, } /* + * This function does a very simple check to see if we can collapse + * an extent tree with a single extent tree leaf block into the inode. + */ +static void ext4_ext_try_to_merge_up(handle_t *handle, + struct inode *inode, + struct ext4_ext_path *path) +{ + size_t s; + unsigned max_root = ext4_ext_space_root(inode, 0); + ext4_fsblk_t blk; + + if ((path[0].p_depth != 1) || + (le16_to_cpu(path[0].p_hdr->eh_entries) != 1) || + (le16_to_cpu(path[1].p_hdr->eh_entries) > max_root)) + return; + + /* + * We need to modify the block allocation bitmap and the block + * group descriptor to release the extent tree block. If we + * can't get the journal credits, give up. + */ + if (ext4_journal_extend(handle, 2)) + return; + + /* + * Copy the extent data up to the inode + */ + blk = ext4_idx_pblock(path[0].p_idx); + s = le16_to_cpu(path[1].p_hdr->eh_entries) * + sizeof(struct ext4_extent_idx); + s += sizeof(struct ext4_extent_header); + + memcpy(path[0].p_hdr, path[1].p_hdr, s); + path[0].p_depth = 0; + path[0].p_ext = EXT_FIRST_EXTENT(path[0].p_hdr) + + (path[1].p_ext - EXT_FIRST_EXTENT(path[1].p_hdr)); + path[0].p_hdr->eh_max = cpu_to_le16(max_root); + + brelse(path[1].p_bh); + ext4_free_blocks(handle, inode, NULL, blk, 1, + EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET); +} + +/* * This function tries to merge the @ex extent to neighbours in the tree. * return 1 if merge left else 0. */ -static int ext4_ext_try_to_merge(struct inode *inode, +static void ext4_ext_try_to_merge(handle_t *handle, + struct inode *inode, struct ext4_ext_path *path, struct ext4_extent *ex) { struct ext4_extent_header *eh; unsigned int depth; int merge_done = 0; - int ret = 0; depth = ext_depth(inode); BUG_ON(path[depth].p_hdr == NULL); @@ -1675,9 +1722,9 @@ static int ext4_ext_try_to_merge(struct inode *inode, merge_done = ext4_ext_try_to_merge_right(inode, path, ex - 1); if (!merge_done) - ret = ext4_ext_try_to_merge_right(inode, path, ex); + (void) ext4_ext_try_to_merge_right(inode, path, ex); - return ret; + ext4_ext_try_to_merge_up(handle, inode, path); } /* @@ -1893,7 +1940,7 @@ has_space: merge: /* try to merge extents */ if (!(flag & EXT4_GET_BLOCKS_PRE_IO)) - ext4_ext_try_to_merge(inode, path, nearex); + ext4_ext_try_to_merge(handle, inode, path, nearex); /* time to correct all indexes above */ @@ -1901,7 +1948,7 @@ merge: if (err) goto cleanup; - err = ext4_ext_dirty(handle, inode, path + depth); + err = ext4_ext_dirty(handle, inode, path + path->p_depth); cleanup: if (npath) { @@ -2092,13 +2139,10 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, } /* - * ext4_ext_check_cache() + * ext4_ext_in_cache() * Checks to see if the given block is in the cache. * If it is, the cached extent is stored in the given - * cache extent pointer. If the cached extent is a hole, - * this routine should be used instead of - * ext4_ext_in_cache if the calling function needs to - * know the size of the hole. + * cache extent pointer. * * @inode: The files inode * @block: The block to look for in the cache @@ -2107,8 +2151,10 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, * * Return 0 if cache is invalid; 1 if the cache is valid */ -static int ext4_ext_check_cache(struct inode *inode, ext4_lblk_t block, - struct ext4_ext_cache *ex){ +static int +ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block, + struct ext4_extent *ex) +{ struct ext4_ext_cache *cex; struct ext4_sb_info *sbi; int ret = 0; @@ -2125,7 +2171,9 @@ static int ext4_ext_check_cache(struct inode *inode, ext4_lblk_t block, goto errout; if (in_range(block, cex->ec_block, cex->ec_len)) { - memcpy(ex, cex, sizeof(struct ext4_ext_cache)); + ex->ee_block = cpu_to_le32(cex->ec_block); + ext4_ext_store_pblock(ex, cex->ec_start); + ex->ee_len = cpu_to_le16(cex->ec_len); ext_debug("%u cached by %u:%u:%llu\n", block, cex->ec_block, cex->ec_len, cex->ec_start); @@ -2138,37 +2186,6 @@ errout: } /* - * ext4_ext_in_cache() - * Checks to see if the given block is in the cache. - * If it is, the cached extent is stored in the given - * extent pointer. - * - * @inode: The files inode - * @block: The block to look for in the cache - * @ex: Pointer where the cached extent will be stored - * if it contains block - * - * Return 0 if cache is invalid; 1 if the cache is valid - */ -static int -ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block, - struct ext4_extent *ex) -{ - struct ext4_ext_cache cex; - int ret = 0; - - if (ext4_ext_check_cache(inode, block, &cex)) { - ex->ee_block = cpu_to_le32(cex.ec_block); - ext4_ext_store_pblock(ex, cex.ec_start); - ex->ee_len = cpu_to_le16(cex.ec_len); - ret = 1; - } - - return ret; -} - - -/* * ext4_ext_rm_idx: * removes index from the index block. */ @@ -2274,10 +2291,13 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode, struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); unsigned short ee_len = ext4_ext_get_actual_len(ex); ext4_fsblk_t pblk; - int flags = EXT4_FREE_BLOCKS_FORGET; + int flags = 0; if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) - flags |= EXT4_FREE_BLOCKS_METADATA; + flags |= EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET; + else if (ext4_should_journal_data(inode)) + flags |= EXT4_FREE_BLOCKS_FORGET; + /* * For bigalloc file systems, we never free a partial cluster * at the beginning of the extent. Instead, we make a note @@ -2572,7 +2592,7 @@ static int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start, struct ext4_ext_path *path = NULL; ext4_fsblk_t partial_cluster = 0; handle_t *handle; - int i = 0, err; + int i = 0, err = 0; ext_debug("truncate since %u to %u\n", start, end); @@ -2604,12 +2624,16 @@ again: return PTR_ERR(path); } depth = ext_depth(inode); + /* Leaf not may not exist only if inode has no blocks at all */ ex = path[depth].p_ext; if (!ex) { - ext4_ext_drop_refs(path); - kfree(path); - path = NULL; - goto cont; + if (depth) { + EXT4_ERROR_INODE(inode, + "path[%d].p_hdr == NULL", + depth); + err = -EIO; + } + goto out; } ee_block = le32_to_cpu(ex->ee_block); @@ -2641,8 +2665,6 @@ again: goto out; } } -cont: - /* * We start scanning from right side, freeing all the blocks * after i_size and walking into the tree depth-wise. @@ -2895,6 +2917,9 @@ static int ext4_split_extent_at(handle_t *handle, unsigned int ee_len, depth; int err = 0; + BUG_ON((split_flag & (EXT4_EXT_DATA_VALID1 | EXT4_EXT_DATA_VALID2)) == + (EXT4_EXT_DATA_VALID1 | EXT4_EXT_DATA_VALID2)); + ext_debug("ext4_split_extents_at: inode %lu, logical" "block %llu\n", inode->i_ino, (unsigned long long)split); @@ -2924,9 +2949,9 @@ static int ext4_split_extent_at(handle_t *handle, ext4_ext_mark_initialized(ex); if (!(flags & EXT4_GET_BLOCKS_PRE_IO)) - ext4_ext_try_to_merge(inode, path, ex); + ext4_ext_try_to_merge(handle, inode, path, ex); - err = ext4_ext_dirty(handle, inode, path + depth); + err = ext4_ext_dirty(handle, inode, path + path->p_depth); goto out; } @@ -2953,13 +2978,20 @@ static int ext4_split_extent_at(handle_t *handle, err = ext4_ext_insert_extent(handle, inode, path, &newex, flags); if (err == -ENOSPC && (EXT4_EXT_MAY_ZEROOUT & split_flag)) { - err = ext4_ext_zeroout(inode, &orig_ex); + if (split_flag & (EXT4_EXT_DATA_VALID1|EXT4_EXT_DATA_VALID2)) { + if (split_flag & EXT4_EXT_DATA_VALID1) + err = ext4_ext_zeroout(inode, ex2); + else + err = ext4_ext_zeroout(inode, ex); + } else + err = ext4_ext_zeroout(inode, &orig_ex); + if (err) goto fix_extent_len; /* update the extent length and mark as initialized */ ex->ee_len = cpu_to_le16(ee_len); - ext4_ext_try_to_merge(inode, path, ex); - err = ext4_ext_dirty(handle, inode, path + depth); + ext4_ext_try_to_merge(handle, inode, path, ex); + err = ext4_ext_dirty(handle, inode, path + path->p_depth); goto out; } else if (err) goto fix_extent_len; @@ -3006,12 +3038,13 @@ static int ext4_split_extent(handle_t *handle, uninitialized = ext4_ext_is_uninitialized(ex); if (map->m_lblk + map->m_len < ee_block + ee_len) { - split_flag1 = split_flag & EXT4_EXT_MAY_ZEROOUT ? - EXT4_EXT_MAY_ZEROOUT : 0; + split_flag1 = split_flag & EXT4_EXT_MAY_ZEROOUT; flags1 = flags | EXT4_GET_BLOCKS_PRE_IO; if (uninitialized) split_flag1 |= EXT4_EXT_MARK_UNINIT1 | EXT4_EXT_MARK_UNINIT2; + if (split_flag & EXT4_EXT_DATA_VALID2) + split_flag1 |= EXT4_EXT_DATA_VALID1; err = ext4_split_extent_at(handle, inode, path, map->m_lblk + map->m_len, split_flag1, flags1); if (err) @@ -3024,8 +3057,8 @@ static int ext4_split_extent(handle_t *handle, return PTR_ERR(path); if (map->m_lblk >= ee_block) { - split_flag1 = split_flag & EXT4_EXT_MAY_ZEROOUT ? - EXT4_EXT_MAY_ZEROOUT : 0; + split_flag1 = split_flag & (EXT4_EXT_MAY_ZEROOUT | + EXT4_EXT_DATA_VALID2); if (uninitialized) split_flag1 |= EXT4_EXT_MARK_UNINIT1; if (split_flag & EXT4_EXT_MARK_UNINIT2) @@ -3041,7 +3074,6 @@ out: return err ? err : map->m_len; } -#define EXT4_EXT_ZERO_LEN 7 /* * This function is called by ext4_ext_map_blocks() if someone tries to write * to an uninitialized extent. It may result in splitting the uninitialized @@ -3067,13 +3099,14 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, struct ext4_map_blocks *map, struct ext4_ext_path *path) { + struct ext4_sb_info *sbi; struct ext4_extent_header *eh; struct ext4_map_blocks split_map; struct ext4_extent zero_ex; struct ext4_extent *ex; ext4_lblk_t ee_block, eof_block; unsigned int ee_len, depth; - int allocated; + int allocated, max_zeroout = 0; int err = 0; int split_flag = 0; @@ -3081,6 +3114,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, "block %llu, max_blocks %u\n", inode->i_ino, (unsigned long long)map->m_lblk, map->m_len); + sbi = EXT4_SB(inode->i_sb); eof_block = (inode->i_size + inode->i_sb->s_blocksize - 1) >> inode->i_sb->s_blocksize_bits; if (eof_block < map->m_lblk + map->m_len) @@ -3180,9 +3214,12 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, */ split_flag |= ee_block + ee_len <= eof_block ? EXT4_EXT_MAY_ZEROOUT : 0; - /* If extent has less than 2*EXT4_EXT_ZERO_LEN zerout directly */ - if (ee_len <= 2*EXT4_EXT_ZERO_LEN && - (EXT4_EXT_MAY_ZEROOUT & split_flag)) { + if (EXT4_EXT_MAY_ZEROOUT & split_flag) + max_zeroout = sbi->s_extent_max_zeroout_kb >> + inode->i_sb->s_blocksize_bits; + + /* If extent is less than s_max_zeroout_kb, zeroout directly */ + if (max_zeroout && (ee_len <= max_zeroout)) { err = ext4_ext_zeroout(inode, ex); if (err) goto out; @@ -3191,8 +3228,8 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, if (err) goto out; ext4_ext_mark_initialized(ex); - ext4_ext_try_to_merge(inode, path, ex); - err = ext4_ext_dirty(handle, inode, path + depth); + ext4_ext_try_to_merge(handle, inode, path, ex); + err = ext4_ext_dirty(handle, inode, path + path->p_depth); goto out; } @@ -3206,9 +3243,8 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, split_map.m_lblk = map->m_lblk; split_map.m_len = map->m_len; - if (allocated > map->m_len) { - if (allocated <= EXT4_EXT_ZERO_LEN && - (EXT4_EXT_MAY_ZEROOUT & split_flag)) { + if (max_zeroout && (allocated > map->m_len)) { + if (allocated <= max_zeroout) { /* case 3 */ zero_ex.ee_block = cpu_to_le32(map->m_lblk); @@ -3220,9 +3256,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, goto out; split_map.m_lblk = map->m_lblk; split_map.m_len = allocated; - } else if ((map->m_lblk - ee_block + map->m_len < - EXT4_EXT_ZERO_LEN) && - (EXT4_EXT_MAY_ZEROOUT & split_flag)) { + } else if (map->m_lblk - ee_block + map->m_len < max_zeroout) { /* case 2 */ if (map->m_lblk != ee_block) { zero_ex.ee_block = ex->ee_block; @@ -3242,7 +3276,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, } allocated = ext4_split_extent(handle, inode, path, - &split_map, split_flag, 0); + &split_map, split_flag, 0); if (allocated < 0) err = allocated; @@ -3256,7 +3290,7 @@ out: * to an uninitialized extent. * * Writing to an uninitialized extent may result in splitting the uninitialized - * extent into multiple /initialized uninitialized extents (up to three) + * extent into multiple initialized/uninitialized extents (up to three) * There are three possibilities: * a> There is no split required: Entire extent should be uninitialized * b> Splits in two extents: Write is happening at either end of the extent @@ -3303,26 +3337,47 @@ static int ext4_split_unwritten_extents(handle_t *handle, split_flag |= ee_block + ee_len <= eof_block ? EXT4_EXT_MAY_ZEROOUT : 0; split_flag |= EXT4_EXT_MARK_UNINIT2; - + if (flags & EXT4_GET_BLOCKS_CONVERT) + split_flag |= EXT4_EXT_DATA_VALID2; flags |= EXT4_GET_BLOCKS_PRE_IO; return ext4_split_extent(handle, inode, path, map, split_flag, flags); } static int ext4_convert_unwritten_extents_endio(handle_t *handle, - struct inode *inode, - struct ext4_ext_path *path) + struct inode *inode, + struct ext4_map_blocks *map, + struct ext4_ext_path *path) { struct ext4_extent *ex; + ext4_lblk_t ee_block; + unsigned int ee_len; int depth; int err = 0; depth = ext_depth(inode); ex = path[depth].p_ext; + ee_block = le32_to_cpu(ex->ee_block); + ee_len = ext4_ext_get_actual_len(ex); ext_debug("ext4_convert_unwritten_extents_endio: inode %lu, logical" "block %llu, max_blocks %u\n", inode->i_ino, - (unsigned long long)le32_to_cpu(ex->ee_block), - ext4_ext_get_actual_len(ex)); + (unsigned long long)ee_block, ee_len); + + /* If extent is larger than requested then split is required */ + if (ee_block != map->m_lblk || ee_len > map->m_len) { + err = ext4_split_unwritten_extents(handle, inode, map, path, + EXT4_GET_BLOCKS_CONVERT); + if (err < 0) + goto out; + ext4_ext_drop_refs(path); + path = ext4_ext_find_extent(inode, map->m_lblk, path); + if (IS_ERR(path)) { + err = PTR_ERR(path); + goto out; + } + depth = ext_depth(inode); + ex = path[depth].p_ext; + } err = ext4_ext_get_access(handle, inode, path + depth); if (err) @@ -3333,10 +3388,10 @@ static int ext4_convert_unwritten_extents_endio(handle_t *handle, /* note: ext4_ext_correct_indexes() isn't needed here because * borders are not changed */ - ext4_ext_try_to_merge(inode, path, ex); + ext4_ext_try_to_merge(handle, inode, path, ex); /* Mark modified extent as dirty */ - err = ext4_ext_dirty(handle, inode, path + depth); + err = ext4_ext_dirty(handle, inode, path + path->p_depth); out: ext4_ext_show_leaf(inode, path); return err; @@ -3600,7 +3655,7 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode, { int ret = 0; int err = 0; - ext4_io_end_t *io = EXT4_I(inode)->cur_aio_dio; + ext4_io_end_t *io = ext4_inode_aio(inode); ext_debug("ext4_ext_handle_uninitialized_extents: inode %lu, logical " "block %llu, max_blocks %u, flags %x, allocated %u\n", @@ -3615,6 +3670,8 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode, if ((flags & EXT4_GET_BLOCKS_PRE_IO)) { ret = ext4_split_unwritten_extents(handle, inode, map, path, flags); + if (ret <= 0) + goto out; /* * Flag the inode(non aio case) or end_io struct (aio case) * that this IO needs to conversion to written when IO is @@ -3630,7 +3687,7 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode, } /* IO end_io complete, convert the filled extent to written */ if ((flags & EXT4_GET_BLOCKS_CONVERT)) { - ret = ext4_convert_unwritten_extents_endio(handle, inode, + ret = ext4_convert_unwritten_extents_endio(handle, inode, map, path); if (ret >= 0) { ext4_update_inode_fsync_trans(handle, inode, 1); @@ -3858,8 +3915,9 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, unsigned int allocated = 0, offset = 0; unsigned int allocated_clusters = 0; struct ext4_allocation_request ar; - ext4_io_end_t *io = EXT4_I(inode)->cur_aio_dio; + ext4_io_end_t *io = ext4_inode_aio(inode); ext4_lblk_t cluster_offset; + int set_unwritten = 0; ext_debug("blocks %u/%u requested for inode %lu\n", map->m_lblk, map->m_len, inode->i_ino); @@ -4082,13 +4140,8 @@ got_allocated_blocks: * For non asycn direct IO case, flag the inode state * that we need to perform conversion when IO is done. */ - if ((flags & EXT4_GET_BLOCKS_PRE_IO)) { - if (io) - ext4_set_io_unwritten_flag(inode, io); - else - ext4_set_inode_state(inode, - EXT4_STATE_DIO_UNWRITTEN); - } + if ((flags & EXT4_GET_BLOCKS_PRE_IO)) + set_unwritten = 1; if (ext4_should_dioread_nolock(inode)) map->m_flags |= EXT4_MAP_UNINIT; } @@ -4100,6 +4153,15 @@ got_allocated_blocks: if (!err) err = ext4_ext_insert_extent(handle, inode, path, &newex, flags); + + if (!err && set_unwritten) { + if (io) + ext4_set_io_unwritten_flag(inode, io); + else + ext4_set_inode_state(inode, + EXT4_STATE_DIO_UNWRITTEN); + } + if (err && free_on_err) { int fb_flags = flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE ? EXT4_FREE_BLOCKS_NO_QUOT_UPDATE : 0; @@ -4241,7 +4303,7 @@ void ext4_ext_truncate(struct inode *inode) * finish any pending end_io work so we won't run the risk of * converting any truncated blocks to initialized later */ - ext4_flush_completed_IO(inode); + ext4_flush_unwritten_io(inode); /* * probably first extent we're gonna free will be last in block @@ -4401,6 +4463,9 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) */ if (len <= EXT_UNINIT_MAX_LEN << blkbits) flags |= EXT4_GET_BLOCKS_NO_NORMALIZE; + + /* Prevent race condition between unwritten */ + ext4_flush_unwritten_io(inode); retry: while (ret >= 0 && ret < max_blocks) { map.m_lblk = map.m_lblk + ret; @@ -4769,9 +4834,32 @@ int ext4_ext_punch_hole(struct file *file, loff_t offset, loff_t length) loff_t first_page_offset, last_page_offset; int credits, err = 0; + /* + * Write out all dirty pages to avoid race conditions + * Then release them. + */ + if (mapping->nrpages && mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) { + err = filemap_write_and_wait_range(mapping, + offset, offset + length - 1); + + if (err) + return err; + } + + mutex_lock(&inode->i_mutex); + /* It's not possible punch hole on append only file */ + if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) { + err = -EPERM; + goto out_mutex; + } + if (IS_SWAPFILE(inode)) { + err = -ETXTBSY; + goto out_mutex; + } + /* No need to punch hole beyond i_size */ if (offset >= inode->i_size) - return 0; + goto out_mutex; /* * If the hole extends beyond i_size, set the hole @@ -4789,35 +4877,26 @@ int ext4_ext_punch_hole(struct file *file, loff_t offset, loff_t length) first_page_offset = first_page << PAGE_CACHE_SHIFT; last_page_offset = last_page << PAGE_CACHE_SHIFT; - /* - * Write out all dirty pages to avoid race conditions - * Then release them. - */ - if (mapping->nrpages && mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) { - err = filemap_write_and_wait_range(mapping, - offset, offset + length - 1); - - if (err) - return err; - } - /* Now release the pages */ if (last_page_offset > first_page_offset) { truncate_pagecache_range(inode, first_page_offset, last_page_offset - 1); } - /* finish any pending end_io work */ - ext4_flush_completed_IO(inode); + /* Wait all existing dio workers, newcomers will block on i_mutex */ + ext4_inode_block_unlocked_dio(inode); + err = ext4_flush_unwritten_io(inode); + if (err) + goto out_dio; + inode_dio_wait(inode); credits = ext4_writepage_trans_blocks(inode); handle = ext4_journal_start(inode, credits); - if (IS_ERR(handle)) - return PTR_ERR(handle); + if (IS_ERR(handle)) { + err = PTR_ERR(handle); + goto out_dio; + } - err = ext4_orphan_add(handle, inode); - if (err) - goto out; /* * Now we need to zero out the non-page-aligned data in the @@ -4903,10 +4982,13 @@ int ext4_ext_punch_hole(struct file *file, loff_t offset, loff_t length) up_write(&EXT4_I(inode)->i_data_sem); out: - ext4_orphan_del(handle, inode); inode->i_mtime = inode->i_ctime = ext4_current_time(inode); ext4_mark_inode_dirty(handle, inode); ext4_journal_stop(handle); +out_dio: + ext4_inode_resume_unlocked_dio(inode); +out_mutex: + mutex_unlock(&inode->i_mutex); return err; } int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 3b0e3bdaabfc..bf3966bccd34 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -55,11 +55,11 @@ static int ext4_release_file(struct inode *inode, struct file *filp) return 0; } -static void ext4_aiodio_wait(struct inode *inode) +void ext4_unwritten_wait(struct inode *inode) { wait_queue_head_t *wq = ext4_ioend_wq(inode); - wait_event(*wq, (atomic_read(&EXT4_I(inode)->i_aiodio_unwritten) == 0)); + wait_event(*wq, (atomic_read(&EXT4_I(inode)->i_unwritten) == 0)); } /* @@ -116,7 +116,7 @@ ext4_file_dio_write(struct kiocb *iocb, const struct iovec *iov, "performance will be poor.", inode->i_ino, current->comm); mutex_lock(ext4_aio_mutex(inode)); - ext4_aiodio_wait(inode); + ext4_unwritten_wait(inode); } BUG_ON(iocb->ki_pos != pos); @@ -207,6 +207,7 @@ ext4_file_write(struct kiocb *iocb, const struct iovec *iov, static const struct vm_operations_struct ext4_file_vm_ops = { .fault = filemap_fault, .page_mkwrite = ext4_page_mkwrite, + .remap_pages = generic_file_remap_pages, }; static int ext4_file_mmap(struct file *file, struct vm_area_struct *vma) @@ -217,7 +218,6 @@ static int ext4_file_mmap(struct file *file, struct vm_area_struct *vma) return -ENOEXEC; file_accessed(file); vma->vm_ops = &ext4_file_vm_ops; - vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index 2a1dcea4f12e..be1d89f385b4 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -34,87 +34,6 @@ #include <trace/events/ext4.h> -static void dump_completed_IO(struct inode * inode) -{ -#ifdef EXT4FS_DEBUG - struct list_head *cur, *before, *after; - ext4_io_end_t *io, *io0, *io1; - unsigned long flags; - - if (list_empty(&EXT4_I(inode)->i_completed_io_list)){ - ext4_debug("inode %lu completed_io list is empty\n", inode->i_ino); - return; - } - - ext4_debug("Dump inode %lu completed_io list \n", inode->i_ino); - spin_lock_irqsave(&EXT4_I(inode)->i_completed_io_lock, flags); - list_for_each_entry(io, &EXT4_I(inode)->i_completed_io_list, list){ - cur = &io->list; - before = cur->prev; - io0 = container_of(before, ext4_io_end_t, list); - after = cur->next; - io1 = container_of(after, ext4_io_end_t, list); - - ext4_debug("io 0x%p from inode %lu,prev 0x%p,next 0x%p\n", - io, inode->i_ino, io0, io1); - } - spin_unlock_irqrestore(&EXT4_I(inode)->i_completed_io_lock, flags); -#endif -} - -/* - * This function is called from ext4_sync_file(). - * - * When IO is completed, the work to convert unwritten extents to - * written is queued on workqueue but may not get immediately - * scheduled. When fsync is called, we need to ensure the - * conversion is complete before fsync returns. - * The inode keeps track of a list of pending/completed IO that - * might needs to do the conversion. This function walks through - * the list and convert the related unwritten extents for completed IO - * to written. - * The function return the number of pending IOs on success. - */ -int ext4_flush_completed_IO(struct inode *inode) -{ - ext4_io_end_t *io; - struct ext4_inode_info *ei = EXT4_I(inode); - unsigned long flags; - int ret = 0; - int ret2 = 0; - - dump_completed_IO(inode); - spin_lock_irqsave(&ei->i_completed_io_lock, flags); - while (!list_empty(&ei->i_completed_io_list)){ - io = list_entry(ei->i_completed_io_list.next, - ext4_io_end_t, list); - list_del_init(&io->list); - io->flag |= EXT4_IO_END_IN_FSYNC; - /* - * Calling ext4_end_io_nolock() to convert completed - * IO to written. - * - * When ext4_sync_file() is called, run_queue() may already - * about to flush the work corresponding to this io structure. - * It will be upset if it founds the io structure related - * to the work-to-be schedule is freed. - * - * Thus we need to keep the io structure still valid here after - * conversion finished. The io structure has a flag to - * avoid double converting from both fsync and background work - * queue work. - */ - spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); - ret = ext4_end_io_nolock(io); - if (ret < 0) - ret2 = ret; - spin_lock_irqsave(&ei->i_completed_io_lock, flags); - io->flag &= ~EXT4_IO_END_IN_FSYNC; - } - spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); - return (ret2 < 0) ? ret2 : 0; -} - /* * If we're not journaling and this is a just-created file, we have to * sync our parent directory (if it was freshly created) since @@ -203,7 +122,7 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync) struct inode *inode = file->f_mapping->host; struct ext4_inode_info *ei = EXT4_I(inode); journal_t *journal = EXT4_SB(inode->i_sb)->s_journal; - int ret; + int ret, err; tid_t commit_tid; bool needs_barrier = false; @@ -219,7 +138,7 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync) if (inode->i_sb->s_flags & MS_RDONLY) goto out; - ret = ext4_flush_completed_IO(inode); + ret = ext4_flush_unwritten_io(inode); if (ret < 0) goto out; @@ -255,8 +174,11 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync) needs_barrier = true; jbd2_log_start_commit(journal, commit_tid); ret = jbd2_log_wait_commit(journal, commit_tid); - if (needs_barrier) - blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL); + if (needs_barrier) { + err = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL); + if (!ret) + ret = err; + } out: mutex_unlock(&inode->i_mutex); trace_ext4_sync_file_exit(inode, ret); diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index 26154b81b836..4facdd29a350 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -697,6 +697,15 @@ got_group: if (!gdp) goto fail; + /* + * Check free inodes count before loading bitmap. + */ + if (ext4_free_inodes_count(sb, gdp) == 0) { + if (++group == ngroups) + group = 0; + continue; + } + brelse(inode_bitmap_bh); inode_bitmap_bh = ext4_read_inode_bitmap(sb, group); if (!inode_bitmap_bh) @@ -753,9 +762,7 @@ got: ext4_free_group_clusters_set(sb, gdp, ext4_free_clusters_after_init(sb, group, gdp)); ext4_block_bitmap_csum_set(sb, group, gdp, - block_bitmap_bh, - EXT4_BLOCKS_PER_GROUP(sb) / - 8); + block_bitmap_bh); ext4_group_desc_csum_set(sb, group, gdp); } ext4_unlock_group(sb, group); diff --git a/fs/ext4/indirect.c b/fs/ext4/indirect.c index 830e1b2bf145..792e388e7b44 100644 --- a/fs/ext4/indirect.c +++ b/fs/ext4/indirect.c @@ -807,16 +807,30 @@ ssize_t ext4_ind_direct_IO(int rw, struct kiocb *iocb, retry: if (rw == READ && ext4_should_dioread_nolock(inode)) { - if (unlikely(!list_empty(&ei->i_completed_io_list))) { + if (unlikely(atomic_read(&EXT4_I(inode)->i_unwritten))) { mutex_lock(&inode->i_mutex); - ext4_flush_completed_IO(inode); + ext4_flush_unwritten_io(inode); mutex_unlock(&inode->i_mutex); } + /* + * Nolock dioread optimization may be dynamically disabled + * via ext4_inode_block_unlocked_dio(). Check inode's state + * while holding extra i_dio_count ref. + */ + atomic_inc(&inode->i_dio_count); + smp_mb(); + if (unlikely(ext4_test_inode_state(inode, + EXT4_STATE_DIOREAD_LOCK))) { + inode_dio_done(inode); + goto locked; + } ret = __blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, offset, nr_segs, ext4_get_block, NULL, NULL, 0); + inode_dio_done(inode); } else { +locked: ret = blockdev_direct_IO(rw, iocb, inode, iov, offset, nr_segs, ext4_get_block); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index c862ee5fe79d..b3c243b9afa5 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -732,11 +732,13 @@ struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode, err = ext4_map_blocks(handle, inode, &map, create ? EXT4_GET_BLOCKS_CREATE : 0); + /* ensure we send some value back into *errp */ + *errp = 0; + if (err < 0) *errp = err; if (err <= 0) return NULL; - *errp = 0; bh = sb_getblk(inode->i_sb, map.m_pblk); if (!bh) { @@ -1954,9 +1956,6 @@ out: return ret; } -static int ext4_set_bh_endio(struct buffer_head *bh, struct inode *inode); -static void ext4_end_io_buffer_write(struct buffer_head *bh, int uptodate); - /* * Note that we don't need to start a transaction unless we're journaling data * because we should have holes filled from ext4_page_mkwrite(). We even don't @@ -2463,6 +2462,16 @@ static int ext4_nonda_switch(struct super_block *sb) free_blocks = EXT4_C2B(sbi, percpu_counter_read_positive(&sbi->s_freeclusters_counter)); dirty_blocks = percpu_counter_read_positive(&sbi->s_dirtyclusters_counter); + /* + * Start pushing delalloc when 1/2 of free blocks are dirty. + */ + if (dirty_blocks && (free_blocks < 2 * dirty_blocks) && + !writeback_in_progress(sb->s_bdi) && + down_read_trylock(&sb->s_umount)) { + writeback_inodes_sb(sb, WB_REASON_FS_FREE_SPACE); + up_read(&sb->s_umount); + } + if (2 * free_blocks < 3 * dirty_blocks || free_blocks < (dirty_blocks + EXT4_FREECLUSTERS_WATERMARK)) { /* @@ -2471,13 +2480,6 @@ static int ext4_nonda_switch(struct super_block *sb) */ return 1; } - /* - * Even if we don't switch but are nearing capacity, - * start pushing delalloc when 1/2 of free blocks are dirty. - */ - if (free_blocks < 2 * dirty_blocks) - writeback_inodes_sb_if_idle(sb, WB_REASON_FS_FREE_SPACE); - return 0; } @@ -2879,9 +2881,6 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset, { struct inode *inode = iocb->ki_filp->f_path.dentry->d_inode; ext4_io_end_t *io_end = iocb->private; - struct workqueue_struct *wq; - unsigned long flags; - struct ext4_inode_info *ei; /* if not async direct IO or dio with 0 bytes write, just return */ if (!io_end || !size) @@ -2910,24 +2909,14 @@ out: io_end->iocb = iocb; io_end->result = ret; } - wq = EXT4_SB(io_end->inode->i_sb)->dio_unwritten_wq; - /* Add the io_end to per-inode completed aio dio list*/ - ei = EXT4_I(io_end->inode); - spin_lock_irqsave(&ei->i_completed_io_lock, flags); - list_add_tail(&io_end->list, &ei->i_completed_io_list); - spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); - - /* queue the work to convert unwritten extents to written */ - queue_work(wq, &io_end->work); + ext4_add_complete_io(io_end); } static void ext4_end_io_buffer_write(struct buffer_head *bh, int uptodate) { ext4_io_end_t *io_end = bh->b_private; - struct workqueue_struct *wq; struct inode *inode; - unsigned long flags; if (!test_clear_buffer_uninit(bh) || !io_end) goto out; @@ -2946,15 +2935,7 @@ static void ext4_end_io_buffer_write(struct buffer_head *bh, int uptodate) */ inode = io_end->inode; ext4_set_io_unwritten_flag(inode, io_end); - - /* Add the io_end to per-inode completed io list*/ - spin_lock_irqsave(&EXT4_I(inode)->i_completed_io_lock, flags); - list_add_tail(&io_end->list, &EXT4_I(inode)->i_completed_io_list); - spin_unlock_irqrestore(&EXT4_I(inode)->i_completed_io_lock, flags); - - wq = EXT4_SB(inode->i_sb)->dio_unwritten_wq; - /* queue the work to convert unwritten extents to written */ - queue_work(wq, &io_end->work); + ext4_add_complete_io(io_end); out: bh->b_private = NULL; bh->b_end_io = NULL; @@ -3029,6 +3010,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, overwrite = *((int *)iocb->private); if (overwrite) { + atomic_inc(&inode->i_dio_count); down_read(&EXT4_I(inode)->i_data_sem); mutex_unlock(&inode->i_mutex); } @@ -3054,7 +3036,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, * hook to the iocb. */ iocb->private = NULL; - EXT4_I(inode)->cur_aio_dio = NULL; + ext4_inode_aio_set(inode, NULL); if (!is_sync_kiocb(iocb)) { ext4_io_end_t *io_end = ext4_init_io_end(inode, GFP_NOFS); @@ -3071,7 +3053,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, * is a unwritten extents needs to be converted * when IO is completed. */ - EXT4_I(inode)->cur_aio_dio = iocb->private; + ext4_inode_aio_set(inode, io_end); } if (overwrite) @@ -3091,7 +3073,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, NULL, DIO_LOCKING); if (iocb->private) - EXT4_I(inode)->cur_aio_dio = NULL; + ext4_inode_aio_set(inode, NULL); /* * The io_end structure takes a reference to the inode, * that structure needs to be destroyed and the @@ -3126,6 +3108,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, retake_lock: /* take i_mutex locking again if we do a ovewrite dio */ if (overwrite) { + inode_dio_done(inode); up_read(&EXT4_I(inode)->i_data_sem); mutex_lock(&inode->i_mutex); } @@ -4052,6 +4035,7 @@ static int ext4_do_update_inode(handle_t *handle, struct ext4_inode_info *ei = EXT4_I(inode); struct buffer_head *bh = iloc->bh; int err = 0, rc, block; + int need_datasync = 0; uid_t i_uid; gid_t i_gid; @@ -4102,7 +4086,10 @@ static int ext4_do_update_inode(handle_t *handle, raw_inode->i_file_acl_high = cpu_to_le16(ei->i_file_acl >> 32); raw_inode->i_file_acl_lo = cpu_to_le32(ei->i_file_acl); - ext4_isize_set(raw_inode, ei->i_disksize); + if (ei->i_disksize != ext4_isize(raw_inode)) { + ext4_isize_set(raw_inode, ei->i_disksize); + need_datasync = 1; + } if (ei->i_disksize > 0x7fffffffULL) { struct super_block *sb = inode->i_sb; if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, @@ -4155,7 +4142,7 @@ static int ext4_do_update_inode(handle_t *handle, err = rc; ext4_clear_inode_state(inode, EXT4_STATE_NEW); - ext4_update_inode_fsync_trans(handle, inode, 0); + ext4_update_inode_fsync_trans(handle, inode, need_datasync); out_brelse: brelse(bh); ext4_std_error(inode->i_sb, err); @@ -4298,7 +4285,6 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) } if (attr->ia_valid & ATTR_SIZE) { - inode_dio_wait(inode); if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); @@ -4347,8 +4333,17 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) } if (attr->ia_valid & ATTR_SIZE) { - if (attr->ia_size != i_size_read(inode)) + if (attr->ia_size != i_size_read(inode)) { truncate_setsize(inode, attr->ia_size); + /* Inode size will be reduced, wait for dio in flight. + * Temporarily disable dioread_nolock to prevent + * livelock. */ + if (orphan) { + ext4_inode_block_unlocked_dio(inode); + inode_dio_wait(inode); + ext4_inode_resume_unlocked_dio(inode); + } + } ext4_truncate(inode); } @@ -4727,6 +4722,10 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) return err; } + /* Wait for all existing dio workers */ + ext4_inode_block_unlocked_dio(inode); + inode_dio_wait(inode); + jbd2_journal_lock_updates(journal); /* @@ -4746,6 +4745,7 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) ext4_set_aops(inode); jbd2_journal_unlock_updates(journal); + ext4_inode_resume_unlocked_dio(inode); /* Finally we can mark the inode as dirty. */ @@ -4780,6 +4780,7 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) int retries = 0; sb_start_pagefault(inode->i_sb); + file_update_time(vma->vm_file); /* Delalloc case is easy... */ if (test_opt(inode->i_sb, DELALLOC) && !ext4_should_journal_data(inode) && diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index 5439d6a56e99..5747f52f7c72 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -366,26 +366,11 @@ group_add_out: return -EOPNOTSUPP; } - if (EXT4_HAS_INCOMPAT_FEATURE(sb, - EXT4_FEATURE_INCOMPAT_META_BG)) { - ext4_msg(sb, KERN_ERR, - "Online resizing not (yet) supported with meta_bg"); - return -EOPNOTSUPP; - } - if (copy_from_user(&n_blocks_count, (__u64 __user *)arg, sizeof(__u64))) { return -EFAULT; } - if (n_blocks_count > MAX_32_NUM && - !EXT4_HAS_INCOMPAT_FEATURE(sb, - EXT4_FEATURE_INCOMPAT_64BIT)) { - ext4_msg(sb, KERN_ERR, - "File system only supports 32-bit block numbers"); - return -EOPNOTSUPP; - } - err = ext4_resize_begin(sb); if (err) return err; @@ -420,13 +405,6 @@ resizefs_out: if (!blk_queue_discard(q)) return -EOPNOTSUPP; - if (EXT4_HAS_RO_COMPAT_FEATURE(sb, - EXT4_FEATURE_RO_COMPAT_BIGALLOC)) { - ext4_msg(sb, KERN_ERR, - "FITRIM not supported with bigalloc"); - return -EOPNOTSUPP; - } - if (copy_from_user(&range, (struct fstrim_range __user *)arg, sizeof(range))) return -EFAULT; diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 08778f6cdfe9..526e55358606 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -24,6 +24,7 @@ #include "ext4_jbd2.h" #include "mballoc.h" #include <linux/debugfs.h> +#include <linux/log2.h> #include <linux/slab.h> #include <trace/events/ext4.h> @@ -1338,17 +1339,17 @@ static void mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b, mb_check_buddy(e4b); } -static int mb_find_extent(struct ext4_buddy *e4b, int order, int block, +static int mb_find_extent(struct ext4_buddy *e4b, int block, int needed, struct ext4_free_extent *ex) { int next = block; - int max; + int max, order; void *buddy; assert_spin_locked(ext4_group_lock_ptr(e4b->bd_sb, e4b->bd_group)); BUG_ON(ex == NULL); - buddy = mb_find_buddy(e4b, order, &max); + buddy = mb_find_buddy(e4b, 0, &max); BUG_ON(buddy == NULL); BUG_ON(block >= max); if (mb_test_bit(block, buddy)) { @@ -1358,12 +1359,9 @@ static int mb_find_extent(struct ext4_buddy *e4b, int order, int block, return 0; } - /* FIXME dorp order completely ? */ - if (likely(order == 0)) { - /* find actual order */ - order = mb_find_order_for_block(e4b, block); - block = block >> order; - } + /* find actual order */ + order = mb_find_order_for_block(e4b, block); + block = block >> order; ex->fe_len = 1 << order; ex->fe_start = block << order; @@ -1549,7 +1547,7 @@ static void ext4_mb_check_limits(struct ext4_allocation_context *ac, /* recheck chunk's availability - we don't know * when it was found (within this lock-unlock * period or not) */ - max = mb_find_extent(e4b, 0, bex->fe_start, gex->fe_len, &ex); + max = mb_find_extent(e4b, bex->fe_start, gex->fe_len, &ex); if (max >= gex->fe_len) { ext4_mb_use_best_found(ac, e4b); return; @@ -1641,7 +1639,7 @@ int ext4_mb_try_best_found(struct ext4_allocation_context *ac, return err; ext4_lock_group(ac->ac_sb, group); - max = mb_find_extent(e4b, 0, ex.fe_start, ex.fe_len, &ex); + max = mb_find_extent(e4b, ex.fe_start, ex.fe_len, &ex); if (max > 0) { ac->ac_b_ex = ex; @@ -1662,17 +1660,20 @@ int ext4_mb_find_by_goal(struct ext4_allocation_context *ac, int max; int err; struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + struct ext4_group_info *grp = ext4_get_group_info(ac->ac_sb, group); struct ext4_free_extent ex; if (!(ac->ac_flags & EXT4_MB_HINT_TRY_GOAL)) return 0; + if (grp->bb_free == 0) + return 0; err = ext4_mb_load_buddy(ac->ac_sb, group, e4b); if (err) return err; ext4_lock_group(ac->ac_sb, group); - max = mb_find_extent(e4b, 0, ac->ac_g_ex.fe_start, + max = mb_find_extent(e4b, ac->ac_g_ex.fe_start, ac->ac_g_ex.fe_len, &ex); if (max >= ac->ac_g_ex.fe_len && ac->ac_g_ex.fe_len == sbi->s_stripe) { @@ -1788,7 +1789,7 @@ void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac, break; } - mb_find_extent(e4b, 0, i, ac->ac_g_ex.fe_len, &ex); + mb_find_extent(e4b, i, ac->ac_g_ex.fe_len, &ex); BUG_ON(ex.fe_len <= 0); if (free < ex.fe_len) { ext4_grp_locked_error(sb, e4b->bd_group, 0, 0, @@ -1840,7 +1841,7 @@ void ext4_mb_scan_aligned(struct ext4_allocation_context *ac, while (i < EXT4_CLUSTERS_PER_GROUP(sb)) { if (!mb_test_bit(i, bitmap)) { - max = mb_find_extent(e4b, 0, i, sbi->s_stripe, &ex); + max = mb_find_extent(e4b, i, sbi->s_stripe, &ex); if (max >= sbi->s_stripe) { ac->ac_found++; ac->ac_b_ex = ex; @@ -1862,6 +1863,12 @@ static int ext4_mb_good_group(struct ext4_allocation_context *ac, BUG_ON(cr < 0 || cr >= 4); + free = grp->bb_free; + if (free == 0) + return 0; + if (cr <= 2 && free < ac->ac_g_ex.fe_len) + return 0; + /* We only do this if the grp has never been initialized */ if (unlikely(EXT4_MB_GRP_NEED_INIT(grp))) { int ret = ext4_mb_init_group(ac->ac_sb, group); @@ -1869,10 +1876,7 @@ static int ext4_mb_good_group(struct ext4_allocation_context *ac, return 0; } - free = grp->bb_free; fragments = grp->bb_fragments; - if (free == 0) - return 0; if (fragments == 0) return 0; @@ -2163,6 +2167,39 @@ static struct kmem_cache *get_groupinfo_cache(int blocksize_bits) return cachep; } +/* + * Allocate the top-level s_group_info array for the specified number + * of groups + */ +int ext4_mb_alloc_groupinfo(struct super_block *sb, ext4_group_t ngroups) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + unsigned size; + struct ext4_group_info ***new_groupinfo; + + size = (ngroups + EXT4_DESC_PER_BLOCK(sb) - 1) >> + EXT4_DESC_PER_BLOCK_BITS(sb); + if (size <= sbi->s_group_info_size) + return 0; + + size = roundup_pow_of_two(sizeof(*sbi->s_group_info) * size); + new_groupinfo = ext4_kvzalloc(size, GFP_KERNEL); + if (!new_groupinfo) { + ext4_msg(sb, KERN_ERR, "can't allocate buddy meta group"); + return -ENOMEM; + } + if (sbi->s_group_info) { + memcpy(new_groupinfo, sbi->s_group_info, + sbi->s_group_info_size * sizeof(*sbi->s_group_info)); + ext4_kvfree(sbi->s_group_info); + } + sbi->s_group_info = new_groupinfo; + sbi->s_group_info_size = size / sizeof(*sbi->s_group_info); + ext4_debug("allocated s_groupinfo array for %d meta_bg's\n", + sbi->s_group_info_size); + return 0; +} + /* Create and initialize ext4_group_info data for the given group. */ int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group, struct ext4_group_desc *desc) @@ -2195,12 +2232,11 @@ int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group, sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)]; i = group & (EXT4_DESC_PER_BLOCK(sb) - 1); - meta_group_info[i] = kmem_cache_alloc(cachep, GFP_KERNEL); + meta_group_info[i] = kmem_cache_zalloc(cachep, GFP_KERNEL); if (meta_group_info[i] == NULL) { ext4_msg(sb, KERN_ERR, "can't allocate buddy mem"); goto exit_group_info; } - memset(meta_group_info[i], 0, kmem_cache_size(cachep)); set_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &(meta_group_info[i]->bb_state)); @@ -2252,49 +2288,14 @@ static int ext4_mb_init_backend(struct super_block *sb) ext4_group_t ngroups = ext4_get_groups_count(sb); ext4_group_t i; struct ext4_sb_info *sbi = EXT4_SB(sb); - struct ext4_super_block *es = sbi->s_es; - int num_meta_group_infos; - int num_meta_group_infos_max; - int array_size; + int err; struct ext4_group_desc *desc; struct kmem_cache *cachep; - /* This is the number of blocks used by GDT */ - num_meta_group_infos = (ngroups + EXT4_DESC_PER_BLOCK(sb) - - 1) >> EXT4_DESC_PER_BLOCK_BITS(sb); - - /* - * This is the total number of blocks used by GDT including - * the number of reserved blocks for GDT. - * The s_group_info array is allocated with this value - * to allow a clean online resize without a complex - * manipulation of pointer. - * The drawback is the unused memory when no resize - * occurs but it's very low in terms of pages - * (see comments below) - * Need to handle this properly when META_BG resizing is allowed - */ - num_meta_group_infos_max = num_meta_group_infos + - le16_to_cpu(es->s_reserved_gdt_blocks); + err = ext4_mb_alloc_groupinfo(sb, ngroups); + if (err) + return err; - /* - * array_size is the size of s_group_info array. We round it - * to the next power of two because this approximation is done - * internally by kmalloc so we can have some more memory - * for free here (e.g. may be used for META_BG resize). - */ - array_size = 1; - while (array_size < sizeof(*sbi->s_group_info) * - num_meta_group_infos_max) - array_size = array_size << 1; - /* An 8TB filesystem with 64-bit pointers requires a 4096 byte - * kmalloc. A 128kb malloc should suffice for a 256TB filesystem. - * So a two level scheme suffices for now. */ - sbi->s_group_info = ext4_kvzalloc(array_size, GFP_KERNEL); - if (sbi->s_group_info == NULL) { - ext4_msg(sb, KERN_ERR, "can't allocate buddy meta group"); - return -ENOMEM; - } sbi->s_buddy_cache = new_inode(sb); if (sbi->s_buddy_cache == NULL) { ext4_msg(sb, KERN_ERR, "can't get new inode"); @@ -2322,7 +2323,7 @@ err_freebuddy: cachep = get_groupinfo_cache(sb->s_blocksize_bits); while (i-- > 0) kmem_cache_free(cachep, ext4_get_group_info(sb, i)); - i = num_meta_group_infos; + i = sbi->s_group_info_size; while (i-- > 0) kfree(sbi->s_group_info[i]); iput(sbi->s_buddy_cache); @@ -2804,8 +2805,7 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, } len = ext4_free_group_clusters(sb, gdp) - ac->ac_b_ex.fe_len; ext4_free_group_clusters_set(sb, gdp, len); - ext4_block_bitmap_csum_set(sb, ac->ac_b_ex.fe_group, gdp, bitmap_bh, - EXT4_BLOCKS_PER_GROUP(sb) / 8); + ext4_block_bitmap_csum_set(sb, ac->ac_b_ex.fe_group, gdp, bitmap_bh); ext4_group_desc_csum_set(sb, ac->ac_b_ex.fe_group, gdp); ext4_unlock_group(sb, ac->ac_b_ex.fe_group); @@ -4008,7 +4008,6 @@ ext4_mb_initialize_context(struct ext4_allocation_context *ac, ext4_get_group_no_and_offset(sb, goal, &group, &block); /* set up allocation goals */ - memset(ac, 0, sizeof(struct ext4_allocation_context)); ac->ac_b_ex.fe_logical = ar->logical & ~(sbi->s_cluster_ratio - 1); ac->ac_status = AC_STATUS_CONTINUE; ac->ac_sb = sb; @@ -4291,7 +4290,7 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle, } } - ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS); + ac = kmem_cache_zalloc(ext4_ac_cachep, GFP_NOFS); if (!ac) { ar->len = 0; *errp = -ENOMEM; @@ -4657,6 +4656,8 @@ do_more: * with group lock held. generate_buddy look at * them with group lock_held */ + if (test_opt(sb, DISCARD)) + ext4_issue_discard(sb, block_group, bit, count); ext4_lock_group(sb, block_group); mb_clear_bits(bitmap_bh->b_data, bit, count_clusters); mb_free_blocks(inode, &e4b, bit, count_clusters); @@ -4664,8 +4665,7 @@ do_more: ret = ext4_free_group_clusters(sb, gdp) + count_clusters; ext4_free_group_clusters_set(sb, gdp, ret); - ext4_block_bitmap_csum_set(sb, block_group, gdp, bitmap_bh, - EXT4_BLOCKS_PER_GROUP(sb) / 8); + ext4_block_bitmap_csum_set(sb, block_group, gdp, bitmap_bh); ext4_group_desc_csum_set(sb, block_group, gdp); ext4_unlock_group(sb, block_group); percpu_counter_add(&sbi->s_freeclusters_counter, count_clusters); @@ -4809,8 +4809,7 @@ int ext4_group_add_blocks(handle_t *handle, struct super_block *sb, mb_free_blocks(NULL, &e4b, bit, count); blk_free_count = blocks_freed + ext4_free_group_clusters(sb, desc); ext4_free_group_clusters_set(sb, desc, blk_free_count); - ext4_block_bitmap_csum_set(sb, block_group, desc, bitmap_bh, - EXT4_BLOCKS_PER_GROUP(sb) / 8); + ext4_block_bitmap_csum_set(sb, block_group, desc, bitmap_bh); ext4_group_desc_csum_set(sb, block_group, desc); ext4_unlock_group(sb, block_group); percpu_counter_add(&sbi->s_freeclusters_counter, @@ -4988,10 +4987,12 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range) start = range->start >> sb->s_blocksize_bits; end = start + (range->len >> sb->s_blocksize_bits) - 1; - minlen = range->minlen >> sb->s_blocksize_bits; + minlen = EXT4_NUM_B2C(EXT4_SB(sb), + range->minlen >> sb->s_blocksize_bits); - if (unlikely(minlen > EXT4_CLUSTERS_PER_GROUP(sb)) || - unlikely(start >= max_blks)) + if (minlen > EXT4_CLUSTERS_PER_GROUP(sb) || + start >= max_blks || + range->len < sb->s_blocksize) return -EINVAL; if (end >= max_blks) end = max_blks - 1; @@ -5048,6 +5049,6 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range) atomic_set(&EXT4_SB(sb)->s_last_trim_minblks, minlen); out: - range->len = trimmed * sb->s_blocksize; + range->len = EXT4_C2B(EXT4_SB(sb), trimmed) << sb->s_blocksize_bits; return ret; } diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index c070618c21ce..3ccd889ba953 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -65,11 +65,6 @@ extern u8 mb_enable_debug; #define MB_DEFAULT_MIN_TO_SCAN 10 /* - * How many groups mballoc will scan looking for the best chunk - */ -#define MB_DEFAULT_MAX_GROUPS_TO_SCAN 5 - -/* * with 'ext4_mb_stats' allocator will collect stats that will be * shown at umount. The collecting costs though! */ diff --git a/fs/ext4/move_extent.c b/fs/ext4/move_extent.c index c5826c623e7a..292daeeed455 100644 --- a/fs/ext4/move_extent.c +++ b/fs/ext4/move_extent.c @@ -141,55 +141,21 @@ mext_next_extent(struct inode *inode, struct ext4_ext_path *path, } /** - * mext_check_null_inode - NULL check for two inodes - * - * If inode1 or inode2 is NULL, return -EIO. Otherwise, return 0. - */ -static int -mext_check_null_inode(struct inode *inode1, struct inode *inode2, - const char *function, unsigned int line) -{ - int ret = 0; - - if (inode1 == NULL) { - __ext4_error(inode2->i_sb, function, line, - "Both inodes should not be NULL: " - "inode1 NULL inode2 %lu", inode2->i_ino); - ret = -EIO; - } else if (inode2 == NULL) { - __ext4_error(inode1->i_sb, function, line, - "Both inodes should not be NULL: " - "inode1 %lu inode2 NULL", inode1->i_ino); - ret = -EIO; - } - return ret; -} - -/** * double_down_write_data_sem - Acquire two inodes' write lock of i_data_sem * - * @orig_inode: original inode structure - * @donor_inode: donor inode structure - * Acquire write lock of i_data_sem of the two inodes (orig and donor) by - * i_ino order. + * Acquire write lock of i_data_sem of the two inodes */ static void -double_down_write_data_sem(struct inode *orig_inode, struct inode *donor_inode) +double_down_write_data_sem(struct inode *first, struct inode *second) { - struct inode *first = orig_inode, *second = donor_inode; + if (first < second) { + down_write(&EXT4_I(first)->i_data_sem); + down_write_nested(&EXT4_I(second)->i_data_sem, SINGLE_DEPTH_NESTING); + } else { + down_write(&EXT4_I(second)->i_data_sem); + down_write_nested(&EXT4_I(first)->i_data_sem, SINGLE_DEPTH_NESTING); - /* - * Use the inode number to provide the stable locking order instead - * of its address, because the C language doesn't guarantee you can - * compare pointers that don't come from the same array. - */ - if (donor_inode->i_ino < orig_inode->i_ino) { - first = donor_inode; - second = orig_inode; } - - down_write(&EXT4_I(first)->i_data_sem); - down_write_nested(&EXT4_I(second)->i_data_sem, SINGLE_DEPTH_NESTING); } /** @@ -604,9 +570,8 @@ mext_calc_swap_extents(struct ext4_extent *tmp_dext, diff = donor_off - le32_to_cpu(tmp_dext->ee_block); ext4_ext_store_pblock(tmp_dext, ext4_ext_pblock(tmp_dext) + diff); - tmp_dext->ee_block = - cpu_to_le32(le32_to_cpu(tmp_dext->ee_block) + diff); - tmp_dext->ee_len = cpu_to_le16(le16_to_cpu(tmp_dext->ee_len) - diff); + le32_add_cpu(&tmp_dext->ee_block, diff); + le16_add_cpu(&tmp_dext->ee_len, -diff); if (max_count < ext4_ext_get_actual_len(tmp_dext)) tmp_dext->ee_len = cpu_to_le16(max_count); @@ -629,6 +594,43 @@ mext_calc_swap_extents(struct ext4_extent *tmp_dext, } /** + * mext_check_coverage - Check that all extents in range has the same type + * + * @inode: inode in question + * @from: block offset of inode + * @count: block count to be checked + * @uninit: extents expected to be uninitialized + * @err: pointer to save error value + * + * Return 1 if all extents in range has expected type, and zero otherwise. + */ +static int +mext_check_coverage(struct inode *inode, ext4_lblk_t from, ext4_lblk_t count, + int uninit, int *err) +{ + struct ext4_ext_path *path = NULL; + struct ext4_extent *ext; + ext4_lblk_t last = from + count; + while (from < last) { + *err = get_ext_path(inode, from, &path); + if (*err) + return 0; + ext = path[ext_depth(inode)].p_ext; + if (!ext) { + ext4_ext_drop_refs(path); + return 0; + } + if (uninit != ext4_ext_is_uninitialized(ext)) { + ext4_ext_drop_refs(path); + return 0; + } + from += ext4_ext_get_actual_len(ext); + ext4_ext_drop_refs(path); + } + return 1; +} + +/** * mext_replace_branches - Replace original extents with new extents * * @handle: journal handle @@ -663,9 +665,6 @@ mext_replace_branches(handle_t *handle, struct inode *orig_inode, int replaced_count = 0; int dext_alen; - /* Protect extent trees against block allocations via delalloc */ - double_down_write_data_sem(orig_inode, donor_inode); - /* Get the original extent for the block "orig_off" */ *err = get_ext_path(orig_inode, orig_off, &orig_path); if (*err) @@ -764,12 +763,122 @@ out: ext4_ext_invalidate_cache(orig_inode); ext4_ext_invalidate_cache(donor_inode); - double_up_write_data_sem(orig_inode, donor_inode); - return replaced_count; } /** + * mext_page_double_lock - Grab and lock pages on both @inode1 and @inode2 + * + * @inode1: the inode structure + * @inode2: the inode structure + * @index: page index + * @page: result page vector + * + * Grab two locked pages for inode's by inode order + */ +static int +mext_page_double_lock(struct inode *inode1, struct inode *inode2, + pgoff_t index, struct page *page[2]) +{ + struct address_space *mapping[2]; + unsigned fl = AOP_FLAG_NOFS; + + BUG_ON(!inode1 || !inode2); + if (inode1 < inode2) { + mapping[0] = inode1->i_mapping; + mapping[1] = inode2->i_mapping; + } else { + mapping[0] = inode2->i_mapping; + mapping[1] = inode1->i_mapping; + } + + page[0] = grab_cache_page_write_begin(mapping[0], index, fl); + if (!page[0]) + return -ENOMEM; + + page[1] = grab_cache_page_write_begin(mapping[1], index, fl); + if (!page[1]) { + unlock_page(page[0]); + page_cache_release(page[0]); + return -ENOMEM; + } + + if (inode1 > inode2) { + struct page *tmp; + tmp = page[0]; + page[0] = page[1]; + page[1] = tmp; + } + return 0; +} + +/* Force page buffers uptodate w/o dropping page's lock */ +static int +mext_page_mkuptodate(struct page *page, unsigned from, unsigned to) +{ + struct inode *inode = page->mapping->host; + sector_t block; + struct buffer_head *bh, *head, *arr[MAX_BUF_PER_PAGE]; + unsigned int blocksize, block_start, block_end; + int i, err, nr = 0, partial = 0; + BUG_ON(!PageLocked(page)); + BUG_ON(PageWriteback(page)); + + if (PageUptodate(page)) + return 0; + + blocksize = 1 << inode->i_blkbits; + if (!page_has_buffers(page)) + create_empty_buffers(page, blocksize, 0); + + head = page_buffers(page); + block = (sector_t)page->index << (PAGE_CACHE_SHIFT - inode->i_blkbits); + for (bh = head, block_start = 0; bh != head || !block_start; + block++, block_start = block_end, bh = bh->b_this_page) { + block_end = block_start + blocksize; + if (block_end <= from || block_start >= to) { + if (!buffer_uptodate(bh)) + partial = 1; + continue; + } + if (buffer_uptodate(bh)) + continue; + if (!buffer_mapped(bh)) { + int err = 0; + err = ext4_get_block(inode, block, bh, 0); + if (err) { + SetPageError(page); + return err; + } + if (!buffer_mapped(bh)) { + zero_user(page, block_start, blocksize); + if (!err) + set_buffer_uptodate(bh); + continue; + } + } + BUG_ON(nr >= MAX_BUF_PER_PAGE); + arr[nr++] = bh; + } + /* No io required */ + if (!nr) + goto out; + + for (i = 0; i < nr; i++) { + bh = arr[i]; + if (!bh_uptodate_or_lock(bh)) { + err = bh_submit_read(bh); + if (err) + return err; + } + } +out: + if (!partial) + SetPageUptodate(page); + return 0; +} + +/** * move_extent_per_page - Move extent data per page * * @o_filp: file structure of original file @@ -791,26 +900,24 @@ move_extent_per_page(struct file *o_filp, struct inode *donor_inode, int block_len_in_page, int uninit, int *err) { struct inode *orig_inode = o_filp->f_dentry->d_inode; - struct address_space *mapping = orig_inode->i_mapping; - struct buffer_head *bh; - struct page *page = NULL; - const struct address_space_operations *a_ops = mapping->a_ops; + struct page *pagep[2] = {NULL, NULL}; handle_t *handle; ext4_lblk_t orig_blk_offset; long long offs = orig_page_offset << PAGE_CACHE_SHIFT; unsigned long blocksize = orig_inode->i_sb->s_blocksize; unsigned int w_flags = 0; unsigned int tmp_data_size, data_size, replaced_size; - void *fsdata; - int i, jblocks; - int err2 = 0; + int err2, jblocks, retries = 0; int replaced_count = 0; + int from = data_offset_in_page << orig_inode->i_blkbits; int blocks_per_page = PAGE_CACHE_SIZE >> orig_inode->i_blkbits; /* * It needs twice the amount of ordinary journal buffers because * inode and donor_inode may change each different metadata blocks. */ +again: + *err = 0; jblocks = ext4_writepage_trans_blocks(orig_inode) * 2; handle = ext4_journal_start(orig_inode, jblocks); if (IS_ERR(handle)) { @@ -824,19 +931,6 @@ move_extent_per_page(struct file *o_filp, struct inode *donor_inode, orig_blk_offset = orig_page_offset * blocks_per_page + data_offset_in_page; - /* - * If orig extent is uninitialized one, - * it's not necessary force the page into memory - * and then force it to be written out again. - * Just swap data blocks between orig and donor. - */ - if (uninit) { - replaced_count = mext_replace_branches(handle, orig_inode, - donor_inode, orig_blk_offset, - block_len_in_page, err); - goto out2; - } - offs = (long long)orig_blk_offset << orig_inode->i_blkbits; /* Calculate data_size */ @@ -858,75 +952,120 @@ move_extent_per_page(struct file *o_filp, struct inode *donor_inode, replaced_size = data_size; - *err = a_ops->write_begin(o_filp, mapping, offs, data_size, w_flags, - &page, &fsdata); + *err = mext_page_double_lock(orig_inode, donor_inode, orig_page_offset, + pagep); if (unlikely(*err < 0)) - goto out; - - if (!PageUptodate(page)) { - mapping->a_ops->readpage(o_filp, page); - lock_page(page); - } - + goto stop_journal; /* - * try_to_release_page() doesn't call releasepage in writeback mode. - * We should care about the order of writing to the same file - * by multiple move extent processes. - * It needs to call wait_on_page_writeback() to wait for the - * writeback of the page. + * If orig extent was uninitialized it can become initialized + * at any time after i_data_sem was dropped, in order to + * serialize with delalloc we have recheck extent while we + * hold page's lock, if it is still the case data copy is not + * necessary, just swap data blocks between orig and donor. */ - wait_on_page_writeback(page); + if (uninit) { + double_down_write_data_sem(orig_inode, donor_inode); + /* If any of extents in range became initialized we have to + * fallback to data copying */ + uninit = mext_check_coverage(orig_inode, orig_blk_offset, + block_len_in_page, 1, err); + if (*err) + goto drop_data_sem; - /* Release old bh and drop refs */ - try_to_release_page(page, 0); + uninit &= mext_check_coverage(donor_inode, orig_blk_offset, + block_len_in_page, 1, err); + if (*err) + goto drop_data_sem; + + if (!uninit) { + double_up_write_data_sem(orig_inode, donor_inode); + goto data_copy; + } + if ((page_has_private(pagep[0]) && + !try_to_release_page(pagep[0], 0)) || + (page_has_private(pagep[1]) && + !try_to_release_page(pagep[1], 0))) { + *err = -EBUSY; + goto drop_data_sem; + } + replaced_count = mext_replace_branches(handle, orig_inode, + donor_inode, orig_blk_offset, + block_len_in_page, err); + drop_data_sem: + double_up_write_data_sem(orig_inode, donor_inode); + goto unlock_pages; + } +data_copy: + *err = mext_page_mkuptodate(pagep[0], from, from + replaced_size); + if (*err) + goto unlock_pages; + + /* At this point all buffers in range are uptodate, old mapping layout + * is no longer required, try to drop it now. */ + if ((page_has_private(pagep[0]) && !try_to_release_page(pagep[0], 0)) || + (page_has_private(pagep[1]) && !try_to_release_page(pagep[1], 0))) { + *err = -EBUSY; + goto unlock_pages; + } replaced_count = mext_replace_branches(handle, orig_inode, donor_inode, - orig_blk_offset, block_len_in_page, - &err2); - if (err2) { + orig_blk_offset, + block_len_in_page, err); + if (*err) { if (replaced_count) { block_len_in_page = replaced_count; replaced_size = block_len_in_page << orig_inode->i_blkbits; } else - goto out; + goto unlock_pages; } + /* Perform all necessary steps similar write_begin()/write_end() + * but keeping in mind that i_size will not change */ + *err = __block_write_begin(pagep[0], from, from + replaced_size, + ext4_get_block); + if (!*err) + *err = block_commit_write(pagep[0], from, from + replaced_size); - if (!page_has_buffers(page)) - create_empty_buffers(page, 1 << orig_inode->i_blkbits, 0); - - bh = page_buffers(page); - for (i = 0; i < data_offset_in_page; i++) - bh = bh->b_this_page; - - for (i = 0; i < block_len_in_page; i++) { - *err = ext4_get_block(orig_inode, - (sector_t)(orig_blk_offset + i), bh, 0); - if (*err < 0) - goto out; - - if (bh->b_this_page != NULL) - bh = bh->b_this_page; - } - - *err = a_ops->write_end(o_filp, mapping, offs, data_size, replaced_size, - page, fsdata); - page = NULL; - -out: - if (unlikely(page)) { - if (PageLocked(page)) - unlock_page(page); - page_cache_release(page); - ext4_journal_stop(handle); - } -out2: + if (unlikely(*err < 0)) + goto repair_branches; + + /* Even in case of data=writeback it is reasonable to pin + * inode to transaction, to prevent unexpected data loss */ + *err = ext4_jbd2_file_inode(handle, orig_inode); + +unlock_pages: + unlock_page(pagep[0]); + page_cache_release(pagep[0]); + unlock_page(pagep[1]); + page_cache_release(pagep[1]); +stop_journal: ext4_journal_stop(handle); - - if (err2) - *err = err2; - + /* Buffer was busy because probably is pinned to journal transaction, + * force transaction commit may help to free it. */ + if (*err == -EBUSY && ext4_should_retry_alloc(orig_inode->i_sb, + &retries)) + goto again; return replaced_count; + +repair_branches: + /* + * This should never ever happen! + * Extents are swapped already, but we are not able to copy data. + * Try to swap extents to it's original places + */ + double_down_write_data_sem(orig_inode, donor_inode); + replaced_count = mext_replace_branches(handle, donor_inode, orig_inode, + orig_blk_offset, + block_len_in_page, &err2); + double_up_write_data_sem(orig_inode, donor_inode); + if (replaced_count != block_len_in_page) { + EXT4_ERROR_INODE_BLOCK(orig_inode, (sector_t)(orig_blk_offset), + "Unable to copy data block," + " data will be lost."); + *err = -EIO; + } + replaced_count = 0; + goto unlock_pages; } /** @@ -969,14 +1108,6 @@ mext_check_arguments(struct inode *orig_inode, return -EINVAL; } - /* Files should be in the same ext4 FS */ - if (orig_inode->i_sb != donor_inode->i_sb) { - ext4_debug("ext4 move extent: The argument files " - "should be in same FS [ino:orig %lu, donor %lu]\n", - orig_inode->i_ino, donor_inode->i_ino); - return -EINVAL; - } - /* Ext4 move extent supports only extent based file */ if (!(ext4_test_inode_flag(orig_inode, EXT4_INODE_EXTENTS))) { ext4_debug("ext4 move extent: orig file is not extents " @@ -1002,7 +1133,6 @@ mext_check_arguments(struct inode *orig_inode, } if ((orig_start >= EXT_MAX_BLOCKS) || - (donor_start >= EXT_MAX_BLOCKS) || (*len > EXT_MAX_BLOCKS) || (orig_start + *len >= EXT_MAX_BLOCKS)) { ext4_debug("ext4 move extent: Can't handle over [%u] blocks " @@ -1072,35 +1202,19 @@ mext_check_arguments(struct inode *orig_inode, * @inode1: the inode structure * @inode2: the inode structure * - * Lock two inodes' i_mutex by i_ino order. - * If inode1 or inode2 is NULL, return -EIO. Otherwise, return 0. + * Lock two inodes' i_mutex */ -static int +static void mext_inode_double_lock(struct inode *inode1, struct inode *inode2) { - int ret = 0; - - BUG_ON(inode1 == NULL && inode2 == NULL); - - ret = mext_check_null_inode(inode1, inode2, __func__, __LINE__); - if (ret < 0) - goto out; - - if (inode1 == inode2) { - mutex_lock(&inode1->i_mutex); - goto out; - } - - if (inode1->i_ino < inode2->i_ino) { + BUG_ON(inode1 == inode2); + if (inode1 < inode2) { mutex_lock_nested(&inode1->i_mutex, I_MUTEX_PARENT); mutex_lock_nested(&inode2->i_mutex, I_MUTEX_CHILD); } else { mutex_lock_nested(&inode2->i_mutex, I_MUTEX_PARENT); mutex_lock_nested(&inode1->i_mutex, I_MUTEX_CHILD); } - -out: - return ret; } /** @@ -1109,28 +1223,13 @@ out: * @inode1: the inode that is released first * @inode2: the inode that is released second * - * If inode1 or inode2 is NULL, return -EIO. Otherwise, return 0. */ -static int +static void mext_inode_double_unlock(struct inode *inode1, struct inode *inode2) { - int ret = 0; - - BUG_ON(inode1 == NULL && inode2 == NULL); - - ret = mext_check_null_inode(inode1, inode2, __func__, __LINE__); - if (ret < 0) - goto out; - - if (inode1) - mutex_unlock(&inode1->i_mutex); - - if (inode2 && inode2 != inode1) - mutex_unlock(&inode2->i_mutex); - -out: - return ret; + mutex_unlock(&inode1->i_mutex); + mutex_unlock(&inode2->i_mutex); } /** @@ -1187,16 +1286,23 @@ ext4_move_extents(struct file *o_filp, struct file *d_filp, ext4_lblk_t block_end, seq_start, add_blocks, file_end, seq_blocks = 0; ext4_lblk_t rest_blocks; pgoff_t orig_page_offset = 0, seq_end_page; - int ret1, ret2, depth, last_extent = 0; + int ret, depth, last_extent = 0; int blocks_per_page = PAGE_CACHE_SIZE >> orig_inode->i_blkbits; int data_offset_in_page; int block_len_in_page; int uninit; - /* orig and donor should be different file */ - if (orig_inode->i_ino == donor_inode->i_ino) { + if (orig_inode->i_sb != donor_inode->i_sb) { + ext4_debug("ext4 move extent: The argument files " + "should be in same FS [ino:orig %lu, donor %lu]\n", + orig_inode->i_ino, donor_inode->i_ino); + return -EINVAL; + } + + /* orig and donor should be different inodes */ + if (orig_inode == donor_inode) { ext4_debug("ext4 move extent: The argument files should not " - "be same file [ino:orig %lu, donor %lu]\n", + "be same inode [ino:orig %lu, donor %lu]\n", orig_inode->i_ino, donor_inode->i_ino); return -EINVAL; } @@ -1208,18 +1314,27 @@ ext4_move_extents(struct file *o_filp, struct file *d_filp, orig_inode->i_ino, donor_inode->i_ino); return -EINVAL; } - + /* TODO: This is non obvious task to swap blocks for inodes with full + jornaling enabled */ + if (ext4_should_journal_data(orig_inode) || + ext4_should_journal_data(donor_inode)) { + return -EINVAL; + } /* Protect orig and donor inodes against a truncate */ - ret1 = mext_inode_double_lock(orig_inode, donor_inode); - if (ret1 < 0) - return ret1; + mext_inode_double_lock(orig_inode, donor_inode); + + /* Wait for all existing dio workers */ + ext4_inode_block_unlocked_dio(orig_inode); + ext4_inode_block_unlocked_dio(donor_inode); + inode_dio_wait(orig_inode); + inode_dio_wait(donor_inode); /* Protect extent tree against block allocations via delalloc */ double_down_write_data_sem(orig_inode, donor_inode); /* Check the filesystem environment whether move_extent can be done */ - ret1 = mext_check_arguments(orig_inode, donor_inode, orig_start, + ret = mext_check_arguments(orig_inode, donor_inode, orig_start, donor_start, &len); - if (ret1) + if (ret) goto out; file_end = (i_size_read(orig_inode) - 1) >> orig_inode->i_blkbits; @@ -1227,13 +1342,13 @@ ext4_move_extents(struct file *o_filp, struct file *d_filp, if (file_end < block_end) len -= block_end - file_end; - ret1 = get_ext_path(orig_inode, block_start, &orig_path); - if (ret1) + ret = get_ext_path(orig_inode, block_start, &orig_path); + if (ret) goto out; /* Get path structure to check the hole */ - ret1 = get_ext_path(orig_inode, block_start, &holecheck_path); - if (ret1) + ret = get_ext_path(orig_inode, block_start, &holecheck_path); + if (ret) goto out; depth = ext_depth(orig_inode); @@ -1252,13 +1367,13 @@ ext4_move_extents(struct file *o_filp, struct file *d_filp, last_extent = mext_next_extent(orig_inode, holecheck_path, &ext_cur); if (last_extent < 0) { - ret1 = last_extent; + ret = last_extent; goto out; } last_extent = mext_next_extent(orig_inode, orig_path, &ext_dummy); if (last_extent < 0) { - ret1 = last_extent; + ret = last_extent; goto out; } seq_start = le32_to_cpu(ext_cur->ee_block); @@ -1272,7 +1387,7 @@ ext4_move_extents(struct file *o_filp, struct file *d_filp, if (le32_to_cpu(ext_cur->ee_block) > block_end) { ext4_debug("ext4 move extent: The specified range of file " "may be the hole\n"); - ret1 = -EINVAL; + ret = -EINVAL; goto out; } @@ -1292,7 +1407,7 @@ ext4_move_extents(struct file *o_filp, struct file *d_filp, last_extent = mext_next_extent(orig_inode, holecheck_path, &ext_cur); if (last_extent < 0) { - ret1 = last_extent; + ret = last_extent; break; } add_blocks = ext4_ext_get_actual_len(ext_cur); @@ -1349,18 +1464,18 @@ ext4_move_extents(struct file *o_filp, struct file *d_filp, orig_page_offset, data_offset_in_page, block_len_in_page, uninit, - &ret1); + &ret); /* Count how many blocks we have exchanged */ *moved_len += block_len_in_page; - if (ret1 < 0) + if (ret < 0) break; if (*moved_len > len) { EXT4_ERROR_INODE(orig_inode, "We replaced blocks too much! " "sum of replaced: %llu requested: %llu", *moved_len, len); - ret1 = -EIO; + ret = -EIO; break; } @@ -1374,22 +1489,22 @@ ext4_move_extents(struct file *o_filp, struct file *d_filp, } double_down_write_data_sem(orig_inode, donor_inode); - if (ret1 < 0) + if (ret < 0) break; /* Decrease buffer counter */ if (holecheck_path) ext4_ext_drop_refs(holecheck_path); - ret1 = get_ext_path(orig_inode, seq_start, &holecheck_path); - if (ret1) + ret = get_ext_path(orig_inode, seq_start, &holecheck_path); + if (ret) break; depth = holecheck_path->p_depth; /* Decrease buffer counter */ if (orig_path) ext4_ext_drop_refs(orig_path); - ret1 = get_ext_path(orig_inode, seq_start, &orig_path); - if (ret1) + ret = get_ext_path(orig_inode, seq_start, &orig_path); + if (ret) break; ext_cur = holecheck_path[depth].p_ext; @@ -1412,12 +1527,9 @@ out: kfree(holecheck_path); } double_up_write_data_sem(orig_inode, donor_inode); - ret2 = mext_inode_double_unlock(orig_inode, donor_inode); - - if (ret1) - return ret1; - else if (ret2) - return ret2; + ext4_inode_resume_unlocked_dio(orig_inode); + ext4_inode_resume_unlocked_dio(donor_inode); + mext_inode_double_unlock(orig_inode, donor_inode); - return 0; + return ret; } diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 2a42cc04466f..6d600a69fc9d 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -55,6 +55,13 @@ static struct buffer_head *ext4_append(handle_t *handle, { struct buffer_head *bh; + if (unlikely(EXT4_SB(inode->i_sb)->s_max_dir_size_kb && + ((inode->i_size >> 10) >= + EXT4_SB(inode->i_sb)->s_max_dir_size_kb))) { + *err = -ENOSPC; + return NULL; + } + *block = inode->i_size >> inode->i_sb->s_blocksize_bits; bh = ext4_bread(handle, inode, *block, 1, err); @@ -67,6 +74,12 @@ static struct buffer_head *ext4_append(handle_t *handle, bh = NULL; } } + if (!bh && !(*err)) { + *err = -EIO; + ext4_error(inode->i_sb, + "Directory hole detected on inode %lu\n", + inode->i_ino); + } return bh; } @@ -594,8 +607,11 @@ dx_probe(const struct qstr *d_name, struct inode *dir, u32 hash; frame->bh = NULL; - if (!(bh = ext4_bread (NULL,dir, 0, 0, err))) + if (!(bh = ext4_bread(NULL, dir, 0, 0, err))) { + if (*err == 0) + *err = ERR_BAD_DX_DIR; goto fail; + } root = (struct dx_root *) bh->b_data; if (root->info.hash_version != DX_HASH_TEA && root->info.hash_version != DX_HASH_HALF_MD4 && @@ -696,8 +712,11 @@ dx_probe(const struct qstr *d_name, struct inode *dir, frame->entries = entries; frame->at = at; if (!indirect--) return frame; - if (!(bh = ext4_bread (NULL,dir, dx_get_block(at), 0, err))) + if (!(bh = ext4_bread(NULL, dir, dx_get_block(at), 0, err))) { + if (!(*err)) + *err = ERR_BAD_DX_DIR; goto fail2; + } at = entries = ((struct dx_node *) bh->b_data)->entries; if (!buffer_verified(bh) && @@ -807,8 +826,15 @@ static int ext4_htree_next_block(struct inode *dir, __u32 hash, */ while (num_frames--) { if (!(bh = ext4_bread(NULL, dir, dx_get_block(p->at), - 0, &err))) + 0, &err))) { + if (!err) { + ext4_error(dir->i_sb, + "Directory hole detected on inode %lu\n", + dir->i_ino); + return -EIO; + } return err; /* Failure */ + } if (!buffer_verified(bh) && !ext4_dx_csum_verify(dir, @@ -839,12 +865,19 @@ static int htree_dirblock_to_tree(struct file *dir_file, { struct buffer_head *bh; struct ext4_dir_entry_2 *de, *top; - int err, count = 0; + int err = 0, count = 0; dxtrace(printk(KERN_INFO "In htree dirblock_to_tree: block %lu\n", (unsigned long)block)); - if (!(bh = ext4_bread (NULL, dir, block, 0, &err))) + if (!(bh = ext4_bread(NULL, dir, block, 0, &err))) { + if (!err) { + err = -EIO; + ext4_error(dir->i_sb, + "Directory hole detected on inode %lu\n", + dir->i_ino); + } return err; + } if (!buffer_verified(bh) && !ext4_dirent_csum_verify(dir, (struct ext4_dir_entry *)bh->b_data)) @@ -1267,8 +1300,15 @@ static struct buffer_head * ext4_dx_find_entry(struct inode *dir, const struct q return NULL; do { block = dx_get_block(frame->at); - if (!(bh = ext4_bread(NULL, dir, block, 0, err))) + if (!(bh = ext4_bread(NULL, dir, block, 0, err))) { + if (!(*err)) { + *err = -EIO; + ext4_error(dir->i_sb, + "Directory hole detected on inode %lu\n", + dir->i_ino); + } goto errout; + } if (!buffer_verified(bh) && !ext4_dirent_csum_verify(dir, @@ -1801,9 +1841,15 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry, } blocks = dir->i_size >> sb->s_blocksize_bits; for (block = 0; block < blocks; block++) { - bh = ext4_bread(handle, dir, block, 0, &retval); - if(!bh) + if (!(bh = ext4_bread(handle, dir, block, 0, &retval))) { + if (!retval) { + retval = -EIO; + ext4_error(inode->i_sb, + "Directory hole detected on inode %lu\n", + inode->i_ino); + } return retval; + } if (!buffer_verified(bh) && !ext4_dirent_csum_verify(dir, (struct ext4_dir_entry *)bh->b_data)) @@ -1860,8 +1906,15 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, entries = frame->entries; at = frame->at; - if (!(bh = ext4_bread(handle,dir, dx_get_block(frame->at), 0, &err))) + if (!(bh = ext4_bread(handle, dir, dx_get_block(frame->at), 0, &err))) { + if (!err) { + err = -EIO; + ext4_error(dir->i_sb, + "Directory hole detected on inode %lu\n", + dir->i_ino); + } goto cleanup; + } if (!buffer_verified(bh) && !ext4_dirent_csum_verify(dir, (struct ext4_dir_entry *)bh->b_data)) @@ -2149,9 +2202,7 @@ retry: err = PTR_ERR(inode); if (!IS_ERR(inode)) { init_special_inode(inode, inode->i_mode, rdev); -#ifdef CONFIG_EXT4_FS_XATTR inode->i_op = &ext4_special_inode_operations; -#endif err = ext4_add_nondir(handle, dentry, inode); } ext4_journal_stop(handle); @@ -2199,9 +2250,15 @@ retry: inode->i_op = &ext4_dir_inode_operations; inode->i_fop = &ext4_dir_operations; inode->i_size = EXT4_I(inode)->i_disksize = inode->i_sb->s_blocksize; - dir_block = ext4_bread(handle, inode, 0, 1, &err); - if (!dir_block) + if (!(dir_block = ext4_bread(handle, inode, 0, 1, &err))) { + if (!err) { + err = -EIO; + ext4_error(inode->i_sb, + "Directory hole detected on inode %lu\n", + inode->i_ino); + } goto out_clear_inode; + } BUFFER_TRACE(dir_block, "get_write_access"); err = ext4_journal_get_write_access(handle, dir_block); if (err) @@ -2318,6 +2375,11 @@ static int empty_dir(struct inode *inode) EXT4_ERROR_INODE(inode, "error %d reading directory " "lblock %u", err, lblock); + else + ext4_warning(inode->i_sb, + "bad directory (dir #%lu) - no data block", + inode->i_ino); + offset += sb->s_blocksize; continue; } @@ -2362,7 +2424,7 @@ int ext4_orphan_add(handle_t *handle, struct inode *inode) struct ext4_iloc iloc; int err = 0, rc; - if (!ext4_handle_valid(handle)) + if (!EXT4_SB(sb)->s_journal) return 0; mutex_lock(&EXT4_SB(sb)->s_orphan_lock); @@ -2436,8 +2498,7 @@ int ext4_orphan_del(handle_t *handle, struct inode *inode) struct ext4_iloc iloc; int err = 0; - /* ext4_handle_valid() assumes a valid handle_t pointer */ - if (handle && !ext4_handle_valid(handle)) + if (!EXT4_SB(inode->i_sb)->s_journal) return 0; mutex_lock(&EXT4_SB(inode->i_sb)->s_orphan_lock); @@ -2456,7 +2517,7 @@ int ext4_orphan_del(handle_t *handle, struct inode *inode) * transaction handle with which to update the orphan list on * disk, but we still need to remove the inode from the linked * list in memory. */ - if (sbi->s_journal && !handle) + if (!handle) goto out; err = ext4_reserve_inode_write(handle, inode, &iloc); @@ -2826,9 +2887,15 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, goto end_rename; } retval = -EIO; - dir_bh = ext4_bread(handle, old_inode, 0, 0, &retval); - if (!dir_bh) + if (!(dir_bh = ext4_bread(handle, old_inode, 0, 0, &retval))) { + if (!retval) { + retval = -EIO; + ext4_error(old_inode->i_sb, + "Directory hole detected on inode %lu\n", + old_inode->i_ino); + } goto end_rename; + } if (!buffer_verified(dir_bh) && !ext4_dirent_csum_verify(old_inode, (struct ext4_dir_entry *)dir_bh->b_data)) diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index dcdeef169a69..68e896e12a67 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -71,6 +71,9 @@ void ext4_free_io_end(ext4_io_end_t *io) int i; BUG_ON(!io); + BUG_ON(!list_empty(&io->list)); + BUG_ON(io->flag & EXT4_IO_END_UNWRITTEN); + if (io->page) put_page(io->page); for (i = 0; i < io->num_io_pages; i++) @@ -81,13 +84,8 @@ void ext4_free_io_end(ext4_io_end_t *io) kmem_cache_free(io_end_cachep, io); } -/* - * check a range of space and convert unwritten extents to written. - * - * Called with inode->i_mutex; we depend on this when we manipulate - * io->flag, since we could otherwise race with ext4_flush_completed_IO() - */ -int ext4_end_io_nolock(ext4_io_end_t *io) +/* check a range of space and convert unwritten extents to written. */ +static int ext4_end_io(ext4_io_end_t *io) { struct inode *inode = io->inode; loff_t offset = io->offset; @@ -106,63 +104,136 @@ int ext4_end_io_nolock(ext4_io_end_t *io) "(inode %lu, offset %llu, size %zd, error %d)", inode->i_ino, offset, size, ret); } - if (io->iocb) aio_complete(io->iocb, io->result, 0); if (io->flag & EXT4_IO_END_DIRECT) inode_dio_done(inode); /* Wake up anyone waiting on unwritten extent conversion */ - if (atomic_dec_and_test(&EXT4_I(inode)->i_aiodio_unwritten)) + if (atomic_dec_and_test(&EXT4_I(inode)->i_unwritten)) wake_up_all(ext4_ioend_wq(io->inode)); return ret; } -/* - * work on completed aio dio IO, to convert unwritten extents to extents - */ -static void ext4_end_io_work(struct work_struct *work) +static void dump_completed_IO(struct inode *inode) +{ +#ifdef EXT4FS_DEBUG + struct list_head *cur, *before, *after; + ext4_io_end_t *io, *io0, *io1; + unsigned long flags; + + if (list_empty(&EXT4_I(inode)->i_completed_io_list)) { + ext4_debug("inode %lu completed_io list is empty\n", + inode->i_ino); + return; + } + + ext4_debug("Dump inode %lu completed_io list\n", inode->i_ino); + list_for_each_entry(io, &EXT4_I(inode)->i_completed_io_list, list) { + cur = &io->list; + before = cur->prev; + io0 = container_of(before, ext4_io_end_t, list); + after = cur->next; + io1 = container_of(after, ext4_io_end_t, list); + + ext4_debug("io 0x%p from inode %lu,prev 0x%p,next 0x%p\n", + io, inode->i_ino, io0, io1); + } +#endif +} + +/* Add the io_end to per-inode completed end_io list. */ +void ext4_add_complete_io(ext4_io_end_t *io_end) { - ext4_io_end_t *io = container_of(work, ext4_io_end_t, work); - struct inode *inode = io->inode; - struct ext4_inode_info *ei = EXT4_I(inode); - unsigned long flags; + struct ext4_inode_info *ei = EXT4_I(io_end->inode); + struct workqueue_struct *wq; + unsigned long flags; + + BUG_ON(!(io_end->flag & EXT4_IO_END_UNWRITTEN)); + wq = EXT4_SB(io_end->inode->i_sb)->dio_unwritten_wq; spin_lock_irqsave(&ei->i_completed_io_lock, flags); - if (io->flag & EXT4_IO_END_IN_FSYNC) - goto requeue; - if (list_empty(&io->list)) { - spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); - goto free; + if (list_empty(&ei->i_completed_io_list)) { + io_end->flag |= EXT4_IO_END_QUEUED; + queue_work(wq, &io_end->work); } + list_add_tail(&io_end->list, &ei->i_completed_io_list); + spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); +} - if (!mutex_trylock(&inode->i_mutex)) { - bool was_queued; -requeue: - was_queued = !!(io->flag & EXT4_IO_END_QUEUED); - io->flag |= EXT4_IO_END_QUEUED; - spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); - /* - * Requeue the work instead of waiting so that the work - * items queued after this can be processed. - */ - queue_work(EXT4_SB(inode->i_sb)->dio_unwritten_wq, &io->work); - /* - * To prevent the ext4-dio-unwritten thread from keeping - * requeueing end_io requests and occupying cpu for too long, - * yield the cpu if it sees an end_io request that has already - * been requeued. - */ - if (was_queued) - yield(); - return; +static int ext4_do_flush_completed_IO(struct inode *inode, + ext4_io_end_t *work_io) +{ + ext4_io_end_t *io; + struct list_head unwritten, complete, to_free; + unsigned long flags; + struct ext4_inode_info *ei = EXT4_I(inode); + int err, ret = 0; + + INIT_LIST_HEAD(&complete); + INIT_LIST_HEAD(&to_free); + + spin_lock_irqsave(&ei->i_completed_io_lock, flags); + dump_completed_IO(inode); + list_replace_init(&ei->i_completed_io_list, &unwritten); + spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); + + while (!list_empty(&unwritten)) { + io = list_entry(unwritten.next, ext4_io_end_t, list); + BUG_ON(!(io->flag & EXT4_IO_END_UNWRITTEN)); + list_del_init(&io->list); + + err = ext4_end_io(io); + if (unlikely(!ret && err)) + ret = err; + + list_add_tail(&io->list, &complete); + } + spin_lock_irqsave(&ei->i_completed_io_lock, flags); + while (!list_empty(&complete)) { + io = list_entry(complete.next, ext4_io_end_t, list); + io->flag &= ~EXT4_IO_END_UNWRITTEN; + /* end_io context can not be destroyed now because it still + * used by queued worker. Worker thread will destroy it later */ + if (io->flag & EXT4_IO_END_QUEUED) + list_del_init(&io->list); + else + list_move(&io->list, &to_free); + } + /* If we are called from worker context, it is time to clear queued + * flag, and destroy it's end_io if it was converted already */ + if (work_io) { + work_io->flag &= ~EXT4_IO_END_QUEUED; + if (!(work_io->flag & EXT4_IO_END_UNWRITTEN)) + list_add_tail(&work_io->list, &to_free); } - list_del_init(&io->list); spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); - (void) ext4_end_io_nolock(io); - mutex_unlock(&inode->i_mutex); -free: - ext4_free_io_end(io); + + while (!list_empty(&to_free)) { + io = list_entry(to_free.next, ext4_io_end_t, list); + list_del_init(&io->list); + ext4_free_io_end(io); + } + return ret; +} + +/* + * work on completed aio dio IO, to convert unwritten extents to extents + */ +static void ext4_end_io_work(struct work_struct *work) +{ + ext4_io_end_t *io = container_of(work, ext4_io_end_t, work); + ext4_do_flush_completed_IO(io->inode, io); +} + +int ext4_flush_unwritten_io(struct inode *inode) +{ + int ret; + WARN_ON_ONCE(!mutex_is_locked(&inode->i_mutex) && + !(inode->i_state & I_FREEING)); + ret = ext4_do_flush_completed_IO(inode, NULL); + ext4_unwritten_wait(inode); + return ret; } ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags) @@ -195,9 +266,7 @@ static void buffer_io_error(struct buffer_head *bh) static void ext4_end_bio(struct bio *bio, int error) { ext4_io_end_t *io_end = bio->bi_private; - struct workqueue_struct *wq; struct inode *inode; - unsigned long flags; int i; sector_t bi_sector = bio->bi_sector; @@ -255,14 +324,7 @@ static void ext4_end_bio(struct bio *bio, int error) return; } - /* Add the io_end to per-inode completed io list*/ - spin_lock_irqsave(&EXT4_I(inode)->i_completed_io_lock, flags); - list_add_tail(&io_end->list, &EXT4_I(inode)->i_completed_io_list); - spin_unlock_irqrestore(&EXT4_I(inode)->i_completed_io_lock, flags); - - wq = EXT4_SB(inode->i_sb)->dio_unwritten_wq; - /* queue the work to convert unwritten extents to written */ - queue_work(wq, &io_end->work); + ext4_add_complete_io(io_end); } void ext4_io_submit(struct ext4_io_submit *io) diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 41f6ef68e2e1..47bf06a2765d 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -45,6 +45,28 @@ void ext4_resize_end(struct super_block *sb) smp_mb__after_clear_bit(); } +static ext4_group_t ext4_meta_bg_first_group(struct super_block *sb, + ext4_group_t group) { + return (group >> EXT4_DESC_PER_BLOCK_BITS(sb)) << + EXT4_DESC_PER_BLOCK_BITS(sb); +} + +static ext4_fsblk_t ext4_meta_bg_first_block_no(struct super_block *sb, + ext4_group_t group) { + group = ext4_meta_bg_first_group(sb, group); + return ext4_group_first_block_no(sb, group); +} + +static ext4_grpblk_t ext4_group_overhead_blocks(struct super_block *sb, + ext4_group_t group) { + ext4_grpblk_t overhead; + overhead = ext4_bg_num_gdb(sb, group); + if (ext4_bg_has_super(sb, group)) + overhead += 1 + + le16_to_cpu(EXT4_SB(sb)->s_es->s_reserved_gdt_blocks); + return overhead; +} + #define outside(b, first, last) ((b) < (first) || (b) >= (last)) #define inside(b, first, last) ((b) >= (first) && (b) < (last)) @@ -57,9 +79,7 @@ static int verify_group_input(struct super_block *sb, ext4_fsblk_t end = start + input->blocks_count; ext4_group_t group = input->group; ext4_fsblk_t itend = input->inode_table + sbi->s_itb_per_group; - unsigned overhead = ext4_bg_has_super(sb, group) ? - (1 + ext4_bg_num_gdb(sb, group) + - le16_to_cpu(es->s_reserved_gdt_blocks)) : 0; + unsigned overhead = ext4_group_overhead_blocks(sb, group); ext4_fsblk_t metaend = start + overhead; struct buffer_head *bh = NULL; ext4_grpblk_t free_blocks_count, offset; @@ -200,13 +220,15 @@ static void free_flex_gd(struct ext4_new_flex_group_data *flex_gd) * be a partial of a flex group. * * @sb: super block of fs to which the groups belongs + * + * Returns 0 on a successful allocation of the metadata blocks in the + * block group. */ -static void ext4_alloc_group_tables(struct super_block *sb, +static int ext4_alloc_group_tables(struct super_block *sb, struct ext4_new_flex_group_data *flex_gd, int flexbg_size) { struct ext4_new_group_data *group_data = flex_gd->groups; - struct ext4_super_block *es = EXT4_SB(sb)->s_es; ext4_fsblk_t start_blk; ext4_fsblk_t last_blk; ext4_group_t src_group; @@ -226,23 +248,24 @@ static void ext4_alloc_group_tables(struct super_block *sb, (last_group & ~(flexbg_size - 1)))); next_group: group = group_data[0].group; + if (src_group >= group_data[0].group + flex_gd->count) + return -ENOSPC; start_blk = ext4_group_first_block_no(sb, src_group); last_blk = start_blk + group_data[src_group - group].blocks_count; - overhead = ext4_bg_has_super(sb, src_group) ? - (1 + ext4_bg_num_gdb(sb, src_group) + - le16_to_cpu(es->s_reserved_gdt_blocks)) : 0; + overhead = ext4_group_overhead_blocks(sb, src_group); start_blk += overhead; - BUG_ON(src_group >= group_data[0].group + flex_gd->count); /* We collect contiguous blocks as much as possible. */ src_group++; - for (; src_group <= last_group; src_group++) - if (!ext4_bg_has_super(sb, src_group)) + for (; src_group <= last_group; src_group++) { + overhead = ext4_group_overhead_blocks(sb, src_group); + if (overhead != 0) last_blk += group_data[src_group - group].blocks_count; else break; + } /* Allocate block bitmaps */ for (; bb_index < flex_gd->count; bb_index++) { @@ -300,6 +323,7 @@ next_group: group_data[i].free_blocks_count); } } + return 0; } static struct buffer_head *bclean(handle_t *handle, struct super_block *sb, @@ -433,11 +457,13 @@ static int setup_new_flex_group_blocks(struct super_block *sb, ext4_group_t group, count; struct buffer_head *bh = NULL; int reserved_gdb, i, j, err = 0, err2; + int meta_bg; BUG_ON(!flex_gd->count || !group_data || group_data[0].group != sbi->s_groups_count); reserved_gdb = le16_to_cpu(es->s_reserved_gdt_blocks); + meta_bg = EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_META_BG); /* This transaction may be extended/restarted along the way */ handle = ext4_journal_start_sb(sb, EXT4_MAX_TRANS_DATA); @@ -447,12 +473,25 @@ static int setup_new_flex_group_blocks(struct super_block *sb, group = group_data[0].group; for (i = 0; i < flex_gd->count; i++, group++) { unsigned long gdblocks; + ext4_grpblk_t overhead; gdblocks = ext4_bg_num_gdb(sb, group); start = ext4_group_first_block_no(sb, group); + if (meta_bg == 0 && !ext4_bg_has_super(sb, group)) + goto handle_itb; + + if (meta_bg == 1) { + ext4_group_t first_group; + first_group = ext4_meta_bg_first_group(sb, group); + if (first_group != group + 1 && + first_group != group + EXT4_DESC_PER_BLOCK(sb) - 1) + goto handle_itb; + } + + block = start + ext4_bg_has_super(sb, group); /* Copy all of the GDT blocks into the backup in this group */ - for (j = 0, block = start + 1; j < gdblocks; j++, block++) { + for (j = 0; j < gdblocks; j++, block++) { struct buffer_head *gdb; ext4_debug("update backup group %#04llx\n", block); @@ -493,6 +532,7 @@ static int setup_new_flex_group_blocks(struct super_block *sb, goto out; } +handle_itb: /* Initialize group tables of the grop @group */ if (!(bg_flags[i] & EXT4_BG_INODE_ZEROED)) goto handle_bb; @@ -521,11 +561,11 @@ handle_bb: err = PTR_ERR(bh); goto out; } - if (ext4_bg_has_super(sb, group)) { + overhead = ext4_group_overhead_blocks(sb, group); + if (overhead != 0) { ext4_debug("mark backup superblock %#04llx (+0)\n", start); - ext4_set_bits(bh->b_data, 0, gdblocks + reserved_gdb + - 1); + ext4_set_bits(bh->b_data, 0, overhead); } ext4_mark_bitmap_end(group_data[i].blocks_count, sb->s_blocksize * 8, bh->b_data); @@ -822,6 +862,45 @@ exit_bh: } /* + * add_new_gdb_meta_bg is the sister of add_new_gdb. + */ +static int add_new_gdb_meta_bg(struct super_block *sb, + handle_t *handle, ext4_group_t group) { + ext4_fsblk_t gdblock; + struct buffer_head *gdb_bh; + struct buffer_head **o_group_desc, **n_group_desc; + unsigned long gdb_num = group / EXT4_DESC_PER_BLOCK(sb); + int err; + + gdblock = ext4_meta_bg_first_block_no(sb, group) + + ext4_bg_has_super(sb, group); + gdb_bh = sb_bread(sb, gdblock); + if (!gdb_bh) + return -EIO; + n_group_desc = ext4_kvmalloc((gdb_num + 1) * + sizeof(struct buffer_head *), + GFP_NOFS); + if (!n_group_desc) { + err = -ENOMEM; + ext4_warning(sb, "not enough memory for %lu groups", + gdb_num + 1); + return err; + } + + o_group_desc = EXT4_SB(sb)->s_group_desc; + memcpy(n_group_desc, o_group_desc, + EXT4_SB(sb)->s_gdb_count * sizeof(struct buffer_head *)); + n_group_desc[gdb_num] = gdb_bh; + EXT4_SB(sb)->s_group_desc = n_group_desc; + EXT4_SB(sb)->s_gdb_count++; + ext4_kvfree(o_group_desc); + err = ext4_journal_get_write_access(handle, gdb_bh); + if (unlikely(err)) + brelse(gdb_bh); + return err; +} + +/* * Called when we are adding a new group which has a backup copy of each of * the GDT blocks (i.e. sparse group) and there are reserved GDT blocks. * We need to add these reserved backup GDT blocks to the resize inode, so @@ -949,16 +1028,16 @@ exit_free: * do not copy the full number of backups at this time. The resize * which changed s_groups_count will backup again. */ -static void update_backups(struct super_block *sb, - int blk_off, char *data, int size) +static void update_backups(struct super_block *sb, int blk_off, char *data, + int size, int meta_bg) { struct ext4_sb_info *sbi = EXT4_SB(sb); - const ext4_group_t last = sbi->s_groups_count; + ext4_group_t last; const int bpg = EXT4_BLOCKS_PER_GROUP(sb); unsigned three = 1; unsigned five = 5; unsigned seven = 7; - ext4_group_t group; + ext4_group_t group = 0; int rest = sb->s_blocksize - size; handle_t *handle; int err = 0, err2; @@ -970,10 +1049,17 @@ static void update_backups(struct super_block *sb, goto exit_err; } - ext4_superblock_csum_set(sb, (struct ext4_super_block *)data); + if (meta_bg == 0) { + group = ext4_list_backups(sb, &three, &five, &seven); + last = sbi->s_groups_count; + } else { + group = ext4_meta_bg_first_group(sb, group) + 1; + last = (ext4_group_t)(group + EXT4_DESC_PER_BLOCK(sb) - 2); + } - while ((group = ext4_list_backups(sb, &three, &five, &seven)) < last) { + while (group < sbi->s_groups_count) { struct buffer_head *bh; + ext4_fsblk_t backup_block; /* Out of journal space, and can't get more - abort - so sad */ if (ext4_handle_valid(handle) && @@ -982,13 +1068,20 @@ static void update_backups(struct super_block *sb, (err = ext4_journal_restart(handle, EXT4_MAX_TRANS_DATA))) break; - bh = sb_getblk(sb, group * bpg + blk_off); + if (meta_bg == 0) + backup_block = group * bpg + blk_off; + else + backup_block = (ext4_group_first_block_no(sb, group) + + ext4_bg_has_super(sb, group)); + + bh = sb_getblk(sb, backup_block); if (!bh) { err = -EIO; break; } - ext4_debug("update metadata backup %#04lx\n", - (unsigned long)bh->b_blocknr); + ext4_debug("update metadata backup %llu(+%llu)\n", + backup_block, backup_block - + ext4_group_first_block_no(sb, group)); if ((err = ext4_journal_get_write_access(handle, bh))) break; lock_buffer(bh); @@ -1001,6 +1094,13 @@ static void update_backups(struct super_block *sb, if (unlikely(err)) ext4_std_error(sb, err); brelse(bh); + + if (meta_bg == 0) + group = ext4_list_backups(sb, &three, &five, &seven); + else if (group == last) + break; + else + group = last; } if ((err2 = ext4_journal_stop(handle)) && !err) err = err2; @@ -1043,7 +1143,9 @@ static int ext4_add_new_descs(handle_t *handle, struct super_block *sb, struct ext4_super_block *es = sbi->s_es; struct buffer_head *gdb_bh; int i, gdb_off, gdb_num, err = 0; + int meta_bg; + meta_bg = EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_META_BG); for (i = 0; i < count; i++, group++) { int reserved_gdb = ext4_bg_has_super(sb, group) ? le16_to_cpu(es->s_reserved_gdt_blocks) : 0; @@ -1063,8 +1165,11 @@ static int ext4_add_new_descs(handle_t *handle, struct super_block *sb, if (!err && reserved_gdb && ext4_bg_num_gdb(sb, group)) err = reserve_backup_gdb(handle, resize_inode, group); - } else + } else if (meta_bg != 0) { + err = add_new_gdb_meta_bg(sb, handle, group); + } else { err = add_new_gdb(handle, resize_inode, group); + } if (err) break; } @@ -1076,17 +1181,12 @@ static struct buffer_head *ext4_get_bitmap(struct super_block *sb, __u64 block) struct buffer_head *bh = sb_getblk(sb, block); if (!bh) return NULL; - - if (bitmap_uptodate(bh)) - return bh; - - lock_buffer(bh); - if (bh_submit_read(bh) < 0) { - unlock_buffer(bh); - brelse(bh); - return NULL; + if (!bh_uptodate_or_lock(bh)) { + if (bh_submit_read(bh) < 0) { + brelse(bh); + return NULL; + } } - unlock_buffer(bh); return bh; } @@ -1112,8 +1212,7 @@ static int ext4_set_bitmap_checksums(struct super_block *sb, bh = ext4_get_bitmap(sb, group_data->block_bitmap); if (!bh) return -EIO; - ext4_block_bitmap_csum_set(sb, group, gdp, bh, - EXT4_BLOCKS_PER_GROUP(sb) / 8); + ext4_block_bitmap_csum_set(sb, group, gdp, bh); brelse(bh); return 0; @@ -1161,6 +1260,9 @@ static int ext4_setup_new_descs(handle_t *handle, struct super_block *sb, ext4_free_group_clusters_set(sb, gdp, EXT4_B2C(sbi, group_data->free_blocks_count)); ext4_free_inodes_set(sb, gdp, EXT4_INODES_PER_GROUP(sb)); + if (ext4_has_group_desc_csum(sb)) + ext4_itable_unused_set(sb, gdp, + EXT4_INODES_PER_GROUP(sb)); gdp->bg_flags = cpu_to_le16(*bg_flags); ext4_group_desc_csum_set(sb, group, gdp); @@ -1216,7 +1318,7 @@ static void ext4_update_super(struct super_block *sb, } reserved_blocks = ext4_r_blocks_count(es) * 100; - do_div(reserved_blocks, ext4_blocks_count(es)); + reserved_blocks = div64_u64(reserved_blocks, ext4_blocks_count(es)); reserved_blocks *= blocks_count; do_div(reserved_blocks, 100); @@ -1227,6 +1329,7 @@ static void ext4_update_super(struct super_block *sb, le32_add_cpu(&es->s_free_inodes_count, EXT4_INODES_PER_GROUP(sb) * flex_gd->count); + ext4_debug("free blocks count %llu", ext4_free_blocks_count(es)); /* * We need to protect s_groups_count against other CPUs seeing * inconsistent state in the superblock. @@ -1261,6 +1364,8 @@ static void ext4_update_super(struct super_block *sb, percpu_counter_add(&sbi->s_freeinodes_counter, EXT4_INODES_PER_GROUP(sb) * flex_gd->count); + ext4_debug("free blocks count %llu", + percpu_counter_read(&sbi->s_freeclusters_counter)); if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG) && sbi->s_log_groups_per_flex) { @@ -1349,16 +1454,24 @@ exit_journal: err = err2; if (!err) { - int i; + int gdb_num = group / EXT4_DESC_PER_BLOCK(sb); + int gdb_num_end = ((group + flex_gd->count - 1) / + EXT4_DESC_PER_BLOCK(sb)); + int meta_bg = EXT4_HAS_INCOMPAT_FEATURE(sb, + EXT4_FEATURE_INCOMPAT_META_BG); + sector_t old_gdb = 0; + update_backups(sb, sbi->s_sbh->b_blocknr, (char *)es, - sizeof(struct ext4_super_block)); - for (i = 0; i < flex_gd->count; i++, group++) { + sizeof(struct ext4_super_block), 0); + for (; gdb_num <= gdb_num_end; gdb_num++) { struct buffer_head *gdb_bh; - int gdb_num; - gdb_num = group / EXT4_BLOCKS_PER_GROUP(sb); + gdb_bh = sbi->s_group_desc[gdb_num]; + if (old_gdb == gdb_bh->b_blocknr) + continue; update_backups(sb, gdb_bh->b_blocknr, gdb_bh->b_data, - gdb_bh->b_size); + gdb_bh->b_size, meta_bg); + old_gdb = gdb_bh->b_blocknr; } } exit: @@ -1402,9 +1515,7 @@ static int ext4_setup_next_flex_gd(struct super_block *sb, group_data[i].group = group + i; group_data[i].blocks_count = blocks_per_group; - overhead = ext4_bg_has_super(sb, group + i) ? - (1 + ext4_bg_num_gdb(sb, group + i) + - le16_to_cpu(es->s_reserved_gdt_blocks)) : 0; + overhead = ext4_group_overhead_blocks(sb, group + i); group_data[i].free_blocks_count = blocks_per_group - overhead; if (ext4_has_group_desc_csum(sb)) flex_gd->bg_flags[i] = EXT4_BG_BLOCK_UNINIT | @@ -1492,6 +1603,14 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) if (err) goto out; + err = ext4_alloc_flex_bg_array(sb, input->group + 1); + if (err) + return err; + + err = ext4_mb_alloc_groupinfo(sb, input->group + 1); + if (err) + goto out; + flex_gd.count = 1; flex_gd.groups = input; flex_gd.bg_flags = &bg_flags; @@ -1544,11 +1663,13 @@ errout: err = err2; if (!err) { + ext4_fsblk_t first_block; + first_block = ext4_group_first_block_no(sb, 0); if (test_opt(sb, DEBUG)) printk(KERN_DEBUG "EXT4-fs: extended group to %llu " "blocks\n", ext4_blocks_count(es)); - update_backups(sb, EXT4_SB(sb)->s_sbh->b_blocknr, (char *)es, - sizeof(struct ext4_super_block)); + update_backups(sb, EXT4_SB(sb)->s_sbh->b_blocknr - first_block, + (char *)es, sizeof(struct ext4_super_block), 0); } return err; } @@ -1631,6 +1752,94 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, return err; } /* ext4_group_extend */ + +static int num_desc_blocks(struct super_block *sb, ext4_group_t groups) +{ + return (groups + EXT4_DESC_PER_BLOCK(sb) - 1) / EXT4_DESC_PER_BLOCK(sb); +} + +/* + * Release the resize inode and drop the resize_inode feature if there + * are no more reserved gdt blocks, and then convert the file system + * to enable meta_bg + */ +static int ext4_convert_meta_bg(struct super_block *sb, struct inode *inode) +{ + handle_t *handle; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + struct ext4_inode_info *ei = EXT4_I(inode); + ext4_fsblk_t nr; + int i, ret, err = 0; + int credits = 1; + + ext4_msg(sb, KERN_INFO, "Converting file system to meta_bg"); + if (inode) { + if (es->s_reserved_gdt_blocks) { + ext4_error(sb, "Unexpected non-zero " + "s_reserved_gdt_blocks"); + return -EPERM; + } + + /* Do a quick sanity check of the resize inode */ + if (inode->i_blocks != 1 << (inode->i_blkbits - 9)) + goto invalid_resize_inode; + for (i = 0; i < EXT4_N_BLOCKS; i++) { + if (i == EXT4_DIND_BLOCK) { + if (ei->i_data[i]) + continue; + else + goto invalid_resize_inode; + } + if (ei->i_data[i]) + goto invalid_resize_inode; + } + credits += 3; /* block bitmap, bg descriptor, resize inode */ + } + + handle = ext4_journal_start_sb(sb, credits); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + err = ext4_journal_get_write_access(handle, sbi->s_sbh); + if (err) + goto errout; + + EXT4_CLEAR_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_RESIZE_INODE); + EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_META_BG); + sbi->s_es->s_first_meta_bg = + cpu_to_le32(num_desc_blocks(sb, sbi->s_groups_count)); + + err = ext4_handle_dirty_super(handle, sb); + if (err) { + ext4_std_error(sb, err); + goto errout; + } + + if (inode) { + nr = le32_to_cpu(ei->i_data[EXT4_DIND_BLOCK]); + ext4_free_blocks(handle, inode, NULL, nr, 1, + EXT4_FREE_BLOCKS_METADATA | + EXT4_FREE_BLOCKS_FORGET); + ei->i_data[EXT4_DIND_BLOCK] = 0; + inode->i_blocks = 0; + + err = ext4_mark_inode_dirty(handle, inode); + if (err) + ext4_std_error(sb, err); + } + +errout: + ret = ext4_journal_stop(handle); + if (!err) + err = ret; + return ret; + +invalid_resize_inode: + ext4_error(sb, "corrupted/inconsistent resize inode"); + return -EINVAL; +} + /* * ext4_resize_fs() resizes a fs to new size specified by @n_blocks_count * @@ -1643,21 +1852,31 @@ int ext4_resize_fs(struct super_block *sb, ext4_fsblk_t n_blocks_count) struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_super_block *es = sbi->s_es; struct buffer_head *bh; - struct inode *resize_inode; - ext4_fsblk_t o_blocks_count; - ext4_group_t o_group; - ext4_group_t n_group; - ext4_grpblk_t offset, add; + struct inode *resize_inode = NULL; + ext4_grpblk_t add, offset; unsigned long n_desc_blocks; unsigned long o_desc_blocks; - unsigned long desc_blocks; - int err = 0, flexbg_size = 1; + ext4_group_t o_group; + ext4_group_t n_group; + ext4_fsblk_t o_blocks_count; + ext4_fsblk_t n_blocks_count_retry = 0; + unsigned long last_update_time = 0; + int err = 0, flexbg_size = 1 << sbi->s_log_groups_per_flex; + int meta_bg; + /* See if the device is actually as big as what was requested */ + bh = sb_bread(sb, n_blocks_count - 1); + if (!bh) { + ext4_warning(sb, "can't read last block, resize aborted"); + return -ENOSPC; + } + brelse(bh); + +retry: o_blocks_count = ext4_blocks_count(es); - if (test_opt(sb, DEBUG)) - ext4_msg(sb, KERN_DEBUG, "resizing filesystem from %llu " - "to %llu blocks", o_blocks_count, n_blocks_count); + ext4_msg(sb, KERN_INFO, "resizing filesystem from %llu " + "to %llu blocks", o_blocks_count, n_blocks_count); if (n_blocks_count < o_blocks_count) { /* On-line shrinking not supported */ @@ -1672,32 +1891,49 @@ int ext4_resize_fs(struct super_block *sb, ext4_fsblk_t n_blocks_count) ext4_get_group_no_and_offset(sb, n_blocks_count - 1, &n_group, &offset); ext4_get_group_no_and_offset(sb, o_blocks_count - 1, &o_group, &offset); - n_desc_blocks = (n_group + EXT4_DESC_PER_BLOCK(sb)) / - EXT4_DESC_PER_BLOCK(sb); - o_desc_blocks = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) / - EXT4_DESC_PER_BLOCK(sb); - desc_blocks = n_desc_blocks - o_desc_blocks; + n_desc_blocks = num_desc_blocks(sb, n_group + 1); + o_desc_blocks = num_desc_blocks(sb, sbi->s_groups_count); - if (desc_blocks && - (!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_RESIZE_INODE) || - le16_to_cpu(es->s_reserved_gdt_blocks) < desc_blocks)) { - ext4_warning(sb, "No reserved GDT blocks, can't resize"); - return -EPERM; - } + meta_bg = EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_META_BG); - resize_inode = ext4_iget(sb, EXT4_RESIZE_INO); - if (IS_ERR(resize_inode)) { - ext4_warning(sb, "Error opening resize inode"); - return PTR_ERR(resize_inode); + if (EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_RESIZE_INODE)) { + if (meta_bg) { + ext4_error(sb, "resize_inode and meta_bg enabled " + "simultaneously"); + return -EINVAL; + } + if (n_desc_blocks > o_desc_blocks + + le16_to_cpu(es->s_reserved_gdt_blocks)) { + n_blocks_count_retry = n_blocks_count; + n_desc_blocks = o_desc_blocks + + le16_to_cpu(es->s_reserved_gdt_blocks); + n_group = n_desc_blocks * EXT4_DESC_PER_BLOCK(sb); + n_blocks_count = n_group * EXT4_BLOCKS_PER_GROUP(sb); + n_group--; /* set to last group number */ + } + + if (!resize_inode) + resize_inode = ext4_iget(sb, EXT4_RESIZE_INO); + if (IS_ERR(resize_inode)) { + ext4_warning(sb, "Error opening resize inode"); + return PTR_ERR(resize_inode); + } } - /* See if the device is actually as big as what was requested */ - bh = sb_bread(sb, n_blocks_count - 1); - if (!bh) { - ext4_warning(sb, "can't read last block, resize aborted"); - return -ENOSPC; + if ((!resize_inode && !meta_bg) || n_blocks_count == o_blocks_count) { + err = ext4_convert_meta_bg(sb, resize_inode); + if (err) + goto out; + if (resize_inode) { + iput(resize_inode); + resize_inode = NULL; + } + if (n_blocks_count_retry) { + n_blocks_count = n_blocks_count_retry; + n_blocks_count_retry = 0; + goto retry; + } } - brelse(bh); /* extend the last group */ if (n_group == o_group) @@ -1710,12 +1946,15 @@ int ext4_resize_fs(struct super_block *sb, ext4_fsblk_t n_blocks_count) goto out; } - if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG) && - es->s_log_groups_per_flex) - flexbg_size = 1 << es->s_log_groups_per_flex; + if (ext4_blocks_count(es) == n_blocks_count) + goto out; - o_blocks_count = ext4_blocks_count(es); - if (o_blocks_count == n_blocks_count) + err = ext4_alloc_flex_bg_array(sb, n_group + 1); + if (err) + return err; + + err = ext4_mb_alloc_groupinfo(sb, n_group + 1); + if (err) goto out; flex_gd = alloc_flex_gd(flexbg_size); @@ -1729,19 +1968,33 @@ int ext4_resize_fs(struct super_block *sb, ext4_fsblk_t n_blocks_count) */ while (ext4_setup_next_flex_gd(sb, flex_gd, n_blocks_count, flexbg_size)) { - ext4_alloc_group_tables(sb, flex_gd, flexbg_size); + if (jiffies - last_update_time > HZ * 10) { + if (last_update_time) + ext4_msg(sb, KERN_INFO, + "resized to %llu blocks", + ext4_blocks_count(es)); + last_update_time = jiffies; + } + if (ext4_alloc_group_tables(sb, flex_gd, flexbg_size) != 0) + break; err = ext4_flex_group_add(sb, resize_inode, flex_gd); if (unlikely(err)) break; } + if (!err && n_blocks_count_retry) { + n_blocks_count = n_blocks_count_retry; + n_blocks_count_retry = 0; + free_flex_gd(flex_gd); + flex_gd = NULL; + goto retry; + } + out: if (flex_gd) free_flex_gd(flex_gd); - - iput(resize_inode); - if (test_opt(sb, DEBUG)) - ext4_msg(sb, KERN_DEBUG, "resized filesystem from %llu " - "upto %llu blocks", o_blocks_count, n_blocks_count); + if (resize_inode != NULL) + iput(resize_inode); + ext4_msg(sb, KERN_INFO, "resized filesystem to %llu", n_blocks_count); return err; } diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 69c55d4e4626..80928f716850 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -143,9 +143,10 @@ int ext4_superblock_csum_verify(struct super_block *sb, return es->s_checksum == ext4_superblock_csum(sb, es); } -void ext4_superblock_csum_set(struct super_block *sb, - struct ext4_super_block *es) +void ext4_superblock_csum_set(struct super_block *sb) { + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) return; @@ -420,7 +421,7 @@ static void __save_error_info(struct super_block *sb, const char *func, */ if (!es->s_error_count) mod_timer(&EXT4_SB(sb)->s_err_report, jiffies + 24*60*60*HZ); - es->s_error_count = cpu_to_le32(le32_to_cpu(es->s_error_count) + 1); + le32_add_cpu(&es->s_error_count, 1); } static void save_error_info(struct super_block *sb, const char *func, @@ -850,7 +851,6 @@ static void ext4_put_super(struct super_block *sb) flush_workqueue(sbi->dio_unwritten_wq); destroy_workqueue(sbi->dio_unwritten_wq); - lock_super(sb); if (sbi->s_journal) { err = jbd2_journal_destroy(sbi->s_journal); sbi->s_journal = NULL; @@ -917,7 +917,6 @@ static void ext4_put_super(struct super_block *sb) * Now that we are completely done shutting down the * superblock, we need to actually destroy the kobject. */ - unlock_super(sb); kobject_put(&sbi->s_kobj); wait_for_completion(&sbi->s_kobj_unregister); if (sbi->s_chksum_driver) @@ -956,11 +955,10 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) ei->jinode = NULL; INIT_LIST_HEAD(&ei->i_completed_io_list); spin_lock_init(&ei->i_completed_io_lock); - ei->cur_aio_dio = NULL; ei->i_sync_tid = 0; ei->i_datasync_tid = 0; atomic_set(&ei->i_ioend_count, 0); - atomic_set(&ei->i_aiodio_unwritten, 0); + atomic_set(&ei->i_unwritten, 0); return &ei->vfs_inode; } @@ -1224,6 +1222,7 @@ enum { Opt_inode_readahead_blks, Opt_journal_ioprio, Opt_dioread_nolock, Opt_dioread_lock, Opt_discard, Opt_nodiscard, Opt_init_itable, Opt_noinit_itable, + Opt_max_dir_size_kb, }; static const match_table_t tokens = { @@ -1297,6 +1296,7 @@ static const match_table_t tokens = { {Opt_init_itable, "init_itable=%u"}, {Opt_init_itable, "init_itable"}, {Opt_noinit_itable, "noinit_itable"}, + {Opt_max_dir_size_kb, "max_dir_size_kb=%u"}, {Opt_removed, "check=none"}, /* mount option from ext2/3 */ {Opt_removed, "nocheck"}, /* mount option from ext2/3 */ {Opt_removed, "reservation"}, /* mount option from ext2/3 */ @@ -1477,6 +1477,7 @@ static const struct mount_opts { {Opt_jqfmt_vfsold, QFMT_VFS_OLD, MOPT_QFMT}, {Opt_jqfmt_vfsv0, QFMT_VFS_V0, MOPT_QFMT}, {Opt_jqfmt_vfsv1, QFMT_VFS_V1, MOPT_QFMT}, + {Opt_max_dir_size_kb, 0, MOPT_GTE0}, {Opt_err, 0, 0} }; @@ -1592,6 +1593,8 @@ static int handle_mount_opt(struct super_block *sb, char *opt, int token, if (!args->from) arg = EXT4_DEF_LI_WAIT_MULT; sbi->s_li_wait_mult = arg; + } else if (token == Opt_max_dir_size_kb) { + sbi->s_max_dir_size_kb = arg; } else if (token == Opt_stripe) { sbi->s_stripe = arg; } else if (m->flags & MOPT_DATAJ) { @@ -1664,7 +1667,7 @@ static int parse_options(char *options, struct super_block *sb, * Initialize args struct so we know whether arg was * found; some options take optional arguments. */ - args[0].to = args[0].from = 0; + args[0].to = args[0].from = NULL; token = match_token(p, tokens, args); if (handle_mount_opt(sb, p, token, args, journal_devnum, journal_ioprio, is_remount) < 0) @@ -1740,7 +1743,7 @@ static inline void ext4_show_quota_options(struct seq_file *seq, static const char *token2str(int token) { - static const struct match_token *t; + const struct match_token *t; for (t = tokens; t->token != Opt_err; t++) if (t->token == token && !strchr(t->pattern, '=')) @@ -1823,6 +1826,8 @@ static int _ext4_show_options(struct seq_file *seq, struct super_block *sb, if (nodefs || (test_opt(sb, INIT_INODE_TABLE) && (sbi->s_li_wait_mult != EXT4_DEF_LI_WAIT_MULT))) SEQ_OPTS_PRINT("init_itable=%u", sbi->s_li_wait_mult); + if (nodefs || sbi->s_max_dir_size_kb) + SEQ_OPTS_PRINT("max_dir_size_kb=%u", sbi->s_max_dir_size_kb); ext4_show_quota_options(seq, sb); return 0; @@ -1914,34 +1919,56 @@ done: return res; } +int ext4_alloc_flex_bg_array(struct super_block *sb, ext4_group_t ngroup) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct flex_groups *new_groups; + int size; + + if (!sbi->s_log_groups_per_flex) + return 0; + + size = ext4_flex_group(sbi, ngroup - 1) + 1; + if (size <= sbi->s_flex_groups_allocated) + return 0; + + size = roundup_pow_of_two(size * sizeof(struct flex_groups)); + new_groups = ext4_kvzalloc(size, GFP_KERNEL); + if (!new_groups) { + ext4_msg(sb, KERN_ERR, "not enough memory for %d flex groups", + size / (int) sizeof(struct flex_groups)); + return -ENOMEM; + } + + if (sbi->s_flex_groups) { + memcpy(new_groups, sbi->s_flex_groups, + (sbi->s_flex_groups_allocated * + sizeof(struct flex_groups))); + ext4_kvfree(sbi->s_flex_groups); + } + sbi->s_flex_groups = new_groups; + sbi->s_flex_groups_allocated = size / sizeof(struct flex_groups); + return 0; +} + static int ext4_fill_flex_info(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_group_desc *gdp = NULL; - ext4_group_t flex_group_count; ext4_group_t flex_group; unsigned int groups_per_flex = 0; - size_t size; - int i; + int i, err; sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex; if (sbi->s_log_groups_per_flex < 1 || sbi->s_log_groups_per_flex > 31) { sbi->s_log_groups_per_flex = 0; return 1; } - groups_per_flex = 1 << sbi->s_log_groups_per_flex; - - /* We allocate both existing and potentially added groups */ - flex_group_count = ((sbi->s_groups_count + groups_per_flex - 1) + - ((le16_to_cpu(sbi->s_es->s_reserved_gdt_blocks) + 1) << - EXT4_DESC_PER_BLOCK_BITS(sb))) / groups_per_flex; - size = flex_group_count * sizeof(struct flex_groups); - sbi->s_flex_groups = ext4_kvzalloc(size, GFP_KERNEL); - if (sbi->s_flex_groups == NULL) { - ext4_msg(sb, KERN_ERR, "not enough memory for %u flex groups", - flex_group_count); + groups_per_flex = 1U << sbi->s_log_groups_per_flex; + + err = ext4_alloc_flex_bg_array(sb, sbi->s_groups_count); + if (err) goto failed; - } for (i = 0; i < sbi->s_groups_count; i++) { gdp = ext4_get_group_desc(sb, i, NULL); @@ -2144,10 +2171,12 @@ static void ext4_orphan_cleanup(struct super_block *sb, } if (EXT4_SB(sb)->s_mount_state & EXT4_ERROR_FS) { - if (es->s_last_orphan) + /* don't clear list on RO mount w/ errors */ + if (es->s_last_orphan && !(s_flags & MS_RDONLY)) { jbd_debug(1, "Errors on filesystem, " "clearing orphan list.\n"); - es->s_last_orphan = 0; + es->s_last_orphan = 0; + } jbd_debug(1, "Skipping orphan recovery on fs with errors.\n"); return; } @@ -2528,6 +2557,7 @@ EXT4_RW_ATTR_SBI_UI(mb_order2_req, s_mb_order2_reqs); EXT4_RW_ATTR_SBI_UI(mb_stream_req, s_mb_stream_request); EXT4_RW_ATTR_SBI_UI(mb_group_prealloc, s_mb_group_prealloc); EXT4_RW_ATTR_SBI_UI(max_writeback_mb_bump, s_max_writeback_mb_bump); +EXT4_RW_ATTR_SBI_UI(extent_max_zeroout_kb, s_extent_max_zeroout_kb); EXT4_ATTR(trigger_fs_error, 0200, NULL, trigger_test_error); static struct attribute *ext4_attrs[] = { @@ -2543,6 +2573,7 @@ static struct attribute *ext4_attrs[] = { ATTR_LIST(mb_stream_req), ATTR_LIST(mb_group_prealloc), ATTR_LIST(max_writeback_mb_bump), + ATTR_LIST(extent_max_zeroout_kb), ATTR_LIST(trigger_fs_error), NULL, }; @@ -2550,10 +2581,12 @@ static struct attribute *ext4_attrs[] = { /* Features this copy of ext4 supports */ EXT4_INFO_ATTR(lazy_itable_init); EXT4_INFO_ATTR(batched_discard); +EXT4_INFO_ATTR(meta_bg_resize); static struct attribute *ext4_feat_attrs[] = { ATTR_LIST(lazy_itable_init), ATTR_LIST(batched_discard), + ATTR_LIST(meta_bg_resize), NULL, }; @@ -3374,7 +3407,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) * enable delayed allocation by default * Use -o nodelalloc to turn it off */ - if (!IS_EXT3_SB(sb) && + if (!IS_EXT3_SB(sb) && !IS_EXT2_SB(sb) && ((def_mount_opts & EXT4_DEFM_NODELALLOC) == 0)) set_opt(sb, DELALLOC); @@ -3743,6 +3776,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) sbi->s_stripe = ext4_get_stripe_size(sbi); sbi->s_max_writeback_mb_bump = 128; + sbi->s_extent_max_zeroout_kb = 32; /* * set up enough so that it can read an inode @@ -4348,7 +4382,7 @@ static int ext4_commit_super(struct super_block *sb, int sync) cpu_to_le32(percpu_counter_sum_positive( &EXT4_SB(sb)->s_freeinodes_counter)); BUFFER_TRACE(sbh, "marking dirty"); - ext4_superblock_csum_set(sb, es); + ext4_superblock_csum_set(sb); mark_buffer_dirty(sbh); if (sync) { error = sync_dirty_buffer(sbh); @@ -4519,11 +4553,9 @@ static int ext4_unfreeze(struct super_block *sb) if (sb->s_flags & MS_RDONLY) return 0; - lock_super(sb); /* Reset the needs_recovery flag before the fs is unlocked. */ EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); ext4_commit_super(sb, 1); - unlock_super(sb); return 0; } @@ -4559,7 +4591,6 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data) char *orig_data = kstrdup(data, GFP_KERNEL); /* Store the original options */ - lock_super(sb); old_sb_flags = sb->s_flags; old_opts.s_mount_opt = sbi->s_mount_opt; old_opts.s_mount_opt2 = sbi->s_mount_opt2; @@ -4701,7 +4732,6 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data) if (sbi->s_journal == NULL) ext4_commit_super(sb, 1); - unlock_super(sb); #ifdef CONFIG_QUOTA /* Release old quota file names */ for (i = 0; i < MAXQUOTAS; i++) @@ -4714,10 +4744,8 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data) else if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA)) { err = ext4_enable_quotas(sb); - if (err) { - lock_super(sb); + if (err) goto restore_opts; - } } } #endif @@ -4744,7 +4772,6 @@ restore_opts: sbi->s_qf_names[i] = old_opts.s_qf_names[i]; } #endif - unlock_super(sb); kfree(orig_data); return err; } @@ -5269,8 +5296,10 @@ static int __init ext4_init_fs(void) if (err) goto out6; ext4_kset = kset_create_and_add("ext4", NULL, fs_kobj); - if (!ext4_kset) + if (!ext4_kset) { + err = -ENOMEM; goto out5; + } ext4_proc_root = proc_mkdir("fs/ext4", NULL); err = ext4_init_feat_adverts(); diff --git a/fs/fat/Makefile b/fs/fat/Makefile index e06190322c1c..964b634f6667 100644 --- a/fs/fat/Makefile +++ b/fs/fat/Makefile @@ -6,6 +6,6 @@ obj-$(CONFIG_FAT_FS) += fat.o obj-$(CONFIG_VFAT_FS) += vfat.o obj-$(CONFIG_MSDOS_FS) += msdos.o -fat-y := cache.o dir.o fatent.o file.o inode.o misc.o +fat-y := cache.o dir.o fatent.o file.o inode.o misc.o nfs.o vfat-y := namei_vfat.o msdos-y := namei_msdos.o diff --git a/fs/fat/cache.c b/fs/fat/cache.c index 1cc7038e273d..91ad9e1c9441 100644 --- a/fs/fat/cache.c +++ b/fs/fat/cache.c @@ -190,7 +190,8 @@ static void __fat_cache_inval_inode(struct inode *inode) struct fat_cache *cache; while (!list_empty(&i->cache_lru)) { - cache = list_entry(i->cache_lru.next, struct fat_cache, cache_list); + cache = list_entry(i->cache_lru.next, + struct fat_cache, cache_list); list_del_init(&cache->cache_list); i->nr_caches--; fat_cache_free(cache); @@ -261,9 +262,10 @@ int fat_get_cluster(struct inode *inode, int cluster, int *fclus, int *dclus) if (nr < 0) goto out; else if (nr == FAT_ENT_FREE) { - fat_fs_error_ratelimit(sb, "%s: invalid cluster chain" - " (i_pos %lld)", __func__, - MSDOS_I(inode)->i_pos); + fat_fs_error_ratelimit(sb, + "%s: invalid cluster chain (i_pos %lld)", + __func__, + MSDOS_I(inode)->i_pos); nr = -EIO; goto out; } else if (nr == FAT_ENT_EOF) { diff --git a/fs/fat/dir.c b/fs/fat/dir.c index dc49ed2cbffa..2a182342442e 100644 --- a/fs/fat/dir.c +++ b/fs/fat/dir.c @@ -18,7 +18,7 @@ #include <linux/time.h> #include <linux/buffer_head.h> #include <linux/compat.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> #include <linux/kernel.h> #include "fat.h" @@ -123,7 +123,8 @@ static inline int fat_get_entry(struct inode *dir, loff_t *pos, { /* Fast stuff first */ if (*bh && *de && - (*de - (struct msdos_dir_entry *)(*bh)->b_data) < MSDOS_SB(dir->i_sb)->dir_per_block - 1) { + (*de - (struct msdos_dir_entry *)(*bh)->b_data) < + MSDOS_SB(dir->i_sb)->dir_per_block - 1) { *pos += sizeof(struct msdos_dir_entry); (*de)++; return 0; @@ -155,7 +156,8 @@ static int uni16_to_x8(struct super_block *sb, unsigned char *ascii, while (*ip && ((len - NLS_MAX_CHARSET_SIZE) > 0)) { ec = *ip++; - if ((charlen = nls->uni2char(ec, op, NLS_MAX_CHARSET_SIZE)) > 0) { + charlen = nls->uni2char(ec, op, NLS_MAX_CHARSET_SIZE); + if (charlen > 0) { op += charlen; len -= charlen; } else { @@ -172,12 +174,12 @@ static int uni16_to_x8(struct super_block *sb, unsigned char *ascii, } if (unlikely(*ip)) { - fat_msg(sb, KERN_WARNING, "filename was truncated while " - "converting."); + fat_msg(sb, KERN_WARNING, + "filename was truncated while converting."); } *op = 0; - return (op - ascii); + return op - ascii; } static inline int fat_uni_to_x8(struct super_block *sb, const wchar_t *uni, @@ -205,7 +207,8 @@ fat_short2uni(struct nls_table *t, unsigned char *c, int clen, wchar_t *uni) } static inline int -fat_short2lower_uni(struct nls_table *t, unsigned char *c, int clen, wchar_t *uni) +fat_short2lower_uni(struct nls_table *t, unsigned char *c, + int clen, wchar_t *uni) { int charlen; wchar_t wc; @@ -220,7 +223,8 @@ fat_short2lower_uni(struct nls_table *t, unsigned char *c, int clen, wchar_t *un if (!nc) nc = *c; - if ( (charlen = t->char2uni(&nc, 1, uni)) < 0) { + charlen = t->char2uni(&nc, 1, uni); + if (charlen < 0) { *uni = 0x003f; /* a question mark */ charlen = 1; } @@ -537,7 +541,6 @@ end_of_dir: return err; } - EXPORT_SYMBOL_GPL(fat_search_long); struct fat_ioctl_filldir_callback { @@ -568,13 +571,14 @@ static int __fat_readdir(struct inode *inode, struct file *filp, void *dirent, int short_len = 0, fill_len = 0; int ret = 0; - lock_super(sb); + mutex_lock(&sbi->s_lock); cpos = filp->f_pos; /* Fake . and .. for the root directory. */ if (inode->i_ino == MSDOS_ROOT_INO) { while (cpos < 2) { - if (filldir(dirent, "..", cpos+1, cpos, MSDOS_ROOT_INO, DT_DIR) < 0) + if (filldir(dirent, "..", cpos+1, cpos, + MSDOS_ROOT_INO, DT_DIR) < 0) goto out; cpos++; filp->f_pos++; @@ -689,7 +693,7 @@ fill_failed: if (unicode) __putname(unicode); out: - unlock_super(sb); + mutex_unlock(&sbi->s_lock); return ret; } @@ -872,25 +876,26 @@ static int fat_get_short_entry(struct inode *dir, loff_t *pos, } /* - * The ".." entry can not provide the "struct fat_slot_info" informations - * for inode. So, this function provide the some informations only. + * The ".." entry can not provide the "struct fat_slot_info" information + * for inode, nor a usable i_pos. So, this function provides some information + * only. + * + * Since this function walks through the on-disk inodes within a directory, + * callers are responsible for taking any locks necessary to prevent the + * directory from changing. */ int fat_get_dotdot_entry(struct inode *dir, struct buffer_head **bh, - struct msdos_dir_entry **de, loff_t *i_pos) + struct msdos_dir_entry **de) { - loff_t offset; + loff_t offset = 0; - offset = 0; - *bh = NULL; + *de = NULL; while (fat_get_short_entry(dir, &offset, bh, de) >= 0) { - if (!strncmp((*de)->name, MSDOS_DOTDOT, MSDOS_NAME)) { - *i_pos = fat_make_i_pos(dir->i_sb, *bh, *de); + if (!strncmp((*de)->name, MSDOS_DOTDOT, MSDOS_NAME)) return 0; - } } return -ENOENT; } - EXPORT_SYMBOL_GPL(fat_get_dotdot_entry); /* See if directory is empty */ @@ -913,7 +918,6 @@ int fat_dir_empty(struct inode *dir) brelse(bh); return result; } - EXPORT_SYMBOL_GPL(fat_dir_empty); /* @@ -959,7 +963,6 @@ int fat_scan(struct inode *dir, const unsigned char *name, } return -ENOENT; } - EXPORT_SYMBOL_GPL(fat_scan); static int __fat_remove_entries(struct inode *dir, loff_t pos, int nr_slots) @@ -1047,7 +1050,6 @@ int fat_remove_entries(struct inode *dir, struct fat_slot_info *sinfo) return 0; } - EXPORT_SYMBOL_GPL(fat_remove_entries); static int fat_zeroed_cluster(struct inode *dir, sector_t blknr, int nr_used, @@ -1141,10 +1143,8 @@ int fat_alloc_new_dir(struct inode *dir, struct timespec *ts) de[0].ctime_cs = de[1].ctime_cs = 0; de[0].adate = de[0].cdate = de[1].adate = de[1].cdate = 0; } - de[0].start = cpu_to_le16(cluster); - de[0].starthi = cpu_to_le16(cluster >> 16); - de[1].start = cpu_to_le16(MSDOS_I(dir)->i_logstart); - de[1].starthi = cpu_to_le16(MSDOS_I(dir)->i_logstart >> 16); + fat_set_start(&de[0], cluster); + fat_set_start(&de[1], MSDOS_I(dir)->i_logstart); de[0].size = de[1].size = 0; memset(de + 2, 0, sb->s_blocksize - 2 * sizeof(*de)); set_buffer_uptodate(bhs[0]); @@ -1161,7 +1161,6 @@ error_free: error: return err; } - EXPORT_SYMBOL_GPL(fat_alloc_new_dir); static int fat_add_new_entries(struct inode *dir, void *slots, int nr_slots, @@ -1377,5 +1376,4 @@ error_remove: __fat_remove_entries(dir, pos, free_slots); return err; } - EXPORT_SYMBOL_GPL(fat_add_entries); diff --git a/fs/fat/fat.h b/fs/fat/fat.h index 7d8e0dcac5d5..623f36f0423b 100644 --- a/fs/fat/fat.h +++ b/fs/fat/fat.h @@ -5,6 +5,7 @@ #include <linux/string.h> #include <linux/nls.h> #include <linux/fs.h> +#include <linux/hash.h> #include <linux/mutex.h> #include <linux/ratelimit.h> #include <linux/msdos_fs.h> @@ -27,26 +28,27 @@ struct fat_mount_options { kgid_t fs_gid; unsigned short fs_fmask; unsigned short fs_dmask; - unsigned short codepage; /* Codepage for shortname conversions */ - char *iocharset; /* Charset used for filename input/display */ - unsigned short shortname; /* flags for shortname display/create rule */ - unsigned char name_check; /* r = relaxed, n = normal, s = strict */ - unsigned char errors; /* On error: continue, panic, remount-ro */ + unsigned short codepage; /* Codepage for shortname conversions */ + char *iocharset; /* Charset used for filename input/display */ + unsigned short shortname; /* flags for shortname display/create rule */ + unsigned char name_check; /* r = relaxed, n = normal, s = strict */ + unsigned char errors; /* On error: continue, panic, remount-ro */ unsigned short allow_utime;/* permission for setting the [am]time */ - unsigned quiet:1, /* set = fake successful chmods and chowns */ - showexec:1, /* set = only set x bit for com/exe/bat */ - sys_immutable:1, /* set = system files are immutable */ - dotsOK:1, /* set = hidden and system files are named '.filename' */ - isvfat:1, /* 0=no vfat long filename support, 1=vfat support */ - utf8:1, /* Use of UTF-8 character set (Default) */ - unicode_xlate:1, /* create escape sequences for unhandled Unicode */ - numtail:1, /* Does first alias have a numeric '~1' type tail? */ - flush:1, /* write things quickly */ - nocase:1, /* Does this need case conversion? 0=need case conversion*/ - usefree:1, /* Use free_clusters for FAT32 */ - tz_utc:1, /* Filesystem timestamps are in UTC */ - rodir:1, /* allow ATTR_RO for directory */ - discard:1; /* Issue discard requests on deletions */ + unsigned quiet:1, /* set = fake successful chmods and chowns */ + showexec:1, /* set = only set x bit for com/exe/bat */ + sys_immutable:1, /* set = system files are immutable */ + dotsOK:1, /* set = hidden and system files are named '.filename' */ + isvfat:1, /* 0=no vfat long filename support, 1=vfat support */ + utf8:1, /* Use of UTF-8 character set (Default) */ + unicode_xlate:1, /* create escape sequences for unhandled Unicode */ + numtail:1, /* Does first alias have a numeric '~1' type tail? */ + flush:1, /* write things quickly */ + nocase:1, /* Does this need case conversion? 0=need case conversion*/ + usefree:1, /* Use free_clusters for FAT32 */ + tz_utc:1, /* Filesystem timestamps are in UTC */ + rodir:1, /* allow ATTR_RO for directory */ + discard:1, /* Issue discard requests on deletions */ + nfs:1; /* Do extra work needed for NFS export */ }; #define FAT_HASH_BITS 8 @@ -56,28 +58,29 @@ struct fat_mount_options { * MS-DOS file system in-core superblock data */ struct msdos_sb_info { - unsigned short sec_per_clus; /* sectors/cluster */ - unsigned short cluster_bits; /* log2(cluster_size) */ - unsigned int cluster_size; /* cluster size */ - unsigned char fats,fat_bits; /* number of FATs, FAT bits (12 or 16) */ + unsigned short sec_per_clus; /* sectors/cluster */ + unsigned short cluster_bits; /* log2(cluster_size) */ + unsigned int cluster_size; /* cluster size */ + unsigned char fats, fat_bits; /* number of FATs, FAT bits (12 or 16) */ unsigned short fat_start; - unsigned long fat_length; /* FAT start & length (sec.) */ + unsigned long fat_length; /* FAT start & length (sec.) */ unsigned long dir_start; - unsigned short dir_entries; /* root dir start & entries */ - unsigned long data_start; /* first data sector */ - unsigned long max_cluster; /* maximum cluster number */ - unsigned long root_cluster; /* first cluster of the root directory */ - unsigned long fsinfo_sector; /* sector number of FAT32 fsinfo */ + unsigned short dir_entries; /* root dir start & entries */ + unsigned long data_start; /* first data sector */ + unsigned long max_cluster; /* maximum cluster number */ + unsigned long root_cluster; /* first cluster of the root directory */ + unsigned long fsinfo_sector; /* sector number of FAT32 fsinfo */ struct mutex fat_lock; + struct mutex s_lock; unsigned int prev_free; /* previously allocated cluster number */ unsigned int free_clusters; /* -1 if undefined */ unsigned int free_clus_valid; /* is free_clusters valid? */ struct fat_mount_options options; - struct nls_table *nls_disk; /* Codepage used on disk */ - struct nls_table *nls_io; /* Charset used for input and display */ - const void *dir_ops; /* Opaque; default directory operations */ - int dir_per_block; /* dir entries per block */ - int dir_per_block_bits; /* log2(dir_per_block) */ + struct nls_table *nls_disk; /* Codepage used on disk */ + struct nls_table *nls_io; /* Charset used for input and display */ + const void *dir_ops; /* Opaque; default directory operations */ + int dir_per_block; /* dir entries per block */ + int dir_per_block_bits; /* log2(dir_per_block) */ int fatent_shift; struct fatent_operations *fatent_ops; @@ -88,6 +91,9 @@ struct msdos_sb_info { spinlock_t inode_hash_lock; struct hlist_head inode_hashtable[FAT_HASH_SIZE]; + + spinlock_t dir_hash_lock; + struct hlist_head dir_hashtable[FAT_HASH_SIZE]; }; #define FAT_CACHE_VALID 0 /* special case for valid cache */ @@ -110,6 +116,7 @@ struct msdos_inode_info { int i_attrs; /* unused attribute bits */ loff_t i_pos; /* on-disk position of directory entry or 0 */ struct hlist_node i_fat_hash; /* hash by i_location */ + struct hlist_node i_dir_hash; /* hash by i_logstart */ struct rw_semaphore truncate_lock; /* protect bmap against truncate */ struct inode vfs_inode; }; @@ -262,7 +269,7 @@ extern int fat_subdirs(struct inode *dir); extern int fat_scan(struct inode *dir, const unsigned char *name, struct fat_slot_info *sinfo); extern int fat_get_dotdot_entry(struct inode *dir, struct buffer_head **bh, - struct msdos_dir_entry **de, loff_t *i_pos); + struct msdos_dir_entry **de); extern int fat_alloc_new_dir(struct inode *dir, struct timespec *ts); extern int fat_add_entries(struct inode *dir, void *slots, int nr_slots, struct fat_slot_info *sinfo); @@ -322,7 +329,7 @@ extern long fat_generic_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); extern const struct file_operations fat_file_operations; extern const struct inode_operations fat_file_inode_operations; -extern int fat_setattr(struct dentry * dentry, struct iattr * attr); +extern int fat_setattr(struct dentry *dentry, struct iattr *attr); extern void fat_truncate_blocks(struct inode *inode, loff_t offset); extern int fat_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); @@ -340,7 +347,12 @@ extern int fat_fill_super(struct super_block *sb, void *data, int silent, int isvfat, void (*setup)(struct super_block *)); extern int fat_flush_inodes(struct super_block *sb, struct inode *i1, - struct inode *i2); + struct inode *i2); +static inline unsigned long fat_dir_hash(int logstart) +{ + return hash_32(logstart, FAT_HASH_BITS); +} + /* fat/misc.c */ extern __printf(3, 4) __cold void __fat_fs_error(struct super_block *sb, int report, const char *fmt, ...); @@ -366,6 +378,14 @@ extern int fat_sync_bhs(struct buffer_head **bhs, int nr_bhs); int fat_cache_init(void); void fat_cache_destroy(void); +/* fat/nfs.c */ +struct fid; +extern struct dentry *fat_fh_to_dentry(struct super_block *sb, struct fid *fid, + int fh_len, int fh_type); +extern struct dentry *fat_fh_to_parent(struct super_block *sb, struct fid *fid, + int fh_len, int fh_type); +extern struct dentry *fat_get_parent(struct dentry *child_dir); + /* helper for printk */ typedef unsigned long long llu; diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c index 31f08ab62c56..260705c58062 100644 --- a/fs/fat/fatent.c +++ b/fs/fat/fatent.c @@ -186,9 +186,6 @@ static void fat16_ent_put(struct fat_entry *fatent, int new) static void fat32_ent_put(struct fat_entry *fatent, int new) { - if (new == FAT_ENT_EOF) - new = EOF_FAT32; - WARN_ON(new & 0xf0000000); new |= le32_to_cpu(*fatent->u.ent32_p) & ~0x0fffffff; *fatent->u.ent32_p = cpu_to_le32(new); @@ -203,15 +200,18 @@ static int fat12_ent_next(struct fat_entry *fatent) fatent->entry++; if (fatent->nr_bhs == 1) { - WARN_ON(ent12_p[0] > (u8 *)(bhs[0]->b_data + (bhs[0]->b_size - 2))); - WARN_ON(ent12_p[1] > (u8 *)(bhs[0]->b_data + (bhs[0]->b_size - 1))); + WARN_ON(ent12_p[0] > (u8 *)(bhs[0]->b_data + + (bhs[0]->b_size - 2))); + WARN_ON(ent12_p[1] > (u8 *)(bhs[0]->b_data + + (bhs[0]->b_size - 1))); if (nextp < (u8 *)(bhs[0]->b_data + (bhs[0]->b_size - 1))) { ent12_p[0] = nextp - 1; ent12_p[1] = nextp; return 1; } } else { - WARN_ON(ent12_p[0] != (u8 *)(bhs[0]->b_data + (bhs[0]->b_size - 1))); + WARN_ON(ent12_p[0] != (u8 *)(bhs[0]->b_data + + (bhs[0]->b_size - 1))); WARN_ON(ent12_p[1] != (u8 *)bhs[1]->b_data); ent12_p[0] = nextp - 1; ent12_p[1] = nextp; @@ -631,7 +631,6 @@ error: return err; } - EXPORT_SYMBOL_GPL(fat_free_clusters); /* 128kb is the whole sectors for FAT12 and FAT16 */ diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 4e5a6ac54ebd..5bafaad00530 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -281,15 +281,42 @@ static inline unsigned long fat_hash(loff_t i_pos) return hash_32(i_pos, FAT_HASH_BITS); } +static void dir_hash_init(struct super_block *sb) +{ + struct msdos_sb_info *sbi = MSDOS_SB(sb); + int i; + + spin_lock_init(&sbi->dir_hash_lock); + for (i = 0; i < FAT_HASH_SIZE; i++) + INIT_HLIST_HEAD(&sbi->dir_hashtable[i]); +} + void fat_attach(struct inode *inode, loff_t i_pos) { struct msdos_sb_info *sbi = MSDOS_SB(inode->i_sb); - struct hlist_head *head = sbi->inode_hashtable + fat_hash(i_pos); - spin_lock(&sbi->inode_hash_lock); - MSDOS_I(inode)->i_pos = i_pos; - hlist_add_head(&MSDOS_I(inode)->i_fat_hash, head); - spin_unlock(&sbi->inode_hash_lock); + if (inode->i_ino != MSDOS_ROOT_INO) { + struct hlist_head *head = sbi->inode_hashtable + + fat_hash(i_pos); + + spin_lock(&sbi->inode_hash_lock); + MSDOS_I(inode)->i_pos = i_pos; + hlist_add_head(&MSDOS_I(inode)->i_fat_hash, head); + spin_unlock(&sbi->inode_hash_lock); + } + + /* If NFS support is enabled, cache the mapping of start cluster + * to directory inode. This is used during reconnection of + * dentries to the filesystem root. + */ + if (S_ISDIR(inode->i_mode) && sbi->options.nfs) { + struct hlist_head *d_head = sbi->dir_hashtable; + d_head += fat_dir_hash(MSDOS_I(inode)->i_logstart); + + spin_lock(&sbi->dir_hash_lock); + hlist_add_head(&MSDOS_I(inode)->i_dir_hash, d_head); + spin_unlock(&sbi->dir_hash_lock); + } } EXPORT_SYMBOL_GPL(fat_attach); @@ -300,6 +327,12 @@ void fat_detach(struct inode *inode) MSDOS_I(inode)->i_pos = 0; hlist_del_init(&MSDOS_I(inode)->i_fat_hash); spin_unlock(&sbi->inode_hash_lock); + + if (S_ISDIR(inode->i_mode) && sbi->options.nfs) { + spin_lock(&sbi->dir_hash_lock); + hlist_del_init(&MSDOS_I(inode)->i_dir_hash); + spin_unlock(&sbi->dir_hash_lock); + } } EXPORT_SYMBOL_GPL(fat_detach); @@ -504,6 +537,7 @@ static void init_once(void *foo) ei->cache_valid_id = FAT_CACHE_VALID + 1; INIT_LIST_HEAD(&ei->cache_lru); INIT_HLIST_NODE(&ei->i_fat_hash); + INIT_HLIST_NODE(&ei->i_dir_hash); inode_init_once(&ei->vfs_inode); } @@ -639,9 +673,9 @@ static int fat_write_inode(struct inode *inode, struct writeback_control *wbc) if (inode->i_ino == MSDOS_FSINFO_INO) { struct super_block *sb = inode->i_sb; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); err = fat_clusters_flush(sb); - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); } else err = __fat_write_inode(inode, wbc->sync_mode == WB_SYNC_ALL); @@ -668,125 +702,9 @@ static const struct super_operations fat_sops = { .show_options = fat_show_options, }; -/* - * a FAT file handle with fhtype 3 is - * 0/ i_ino - for fast, reliable lookup if still in the cache - * 1/ i_generation - to see if i_ino is still valid - * bit 0 == 0 iff directory - * 2/ i_pos(8-39) - if ino has changed, but still in cache - * 3/ i_pos(4-7)|i_logstart - to semi-verify inode found at i_pos - * 4/ i_pos(0-3)|parent->i_logstart - maybe used to hunt for the file on disc - * - * Hack for NFSv2: Maximum FAT entry number is 28bits and maximum - * i_pos is 40bits (blocknr(32) + dir offset(8)), so two 4bits - * of i_logstart is used to store the directory entry offset. - */ - -static struct dentry *fat_fh_to_dentry(struct super_block *sb, - struct fid *fid, int fh_len, int fh_type) -{ - struct inode *inode = NULL; - u32 *fh = fid->raw; - - if (fh_len < 5 || fh_type != 3) - return NULL; - - inode = ilookup(sb, fh[0]); - if (!inode || inode->i_generation != fh[1]) { - if (inode) - iput(inode); - inode = NULL; - } - if (!inode) { - loff_t i_pos; - int i_logstart = fh[3] & 0x0fffffff; - - i_pos = (loff_t)fh[2] << 8; - i_pos |= ((fh[3] >> 24) & 0xf0) | (fh[4] >> 28); - - /* try 2 - see if i_pos is in F-d-c - * require i_logstart to be the same - * Will fail if you truncate and then re-write - */ - - inode = fat_iget(sb, i_pos); - if (inode && MSDOS_I(inode)->i_logstart != i_logstart) { - iput(inode); - inode = NULL; - } - } - - /* - * For now, do nothing if the inode is not found. - * - * What we could do is: - * - * - follow the file starting at fh[4], and record the ".." entry, - * and the name of the fh[2] entry. - * - then follow the ".." file finding the next step up. - * - * This way we build a path to the root of the tree. If this works, we - * lookup the path and so get this inode into the cache. Finally try - * the fat_iget lookup again. If that fails, then we are totally out - * of luck. But all that is for another day - */ - return d_obtain_alias(inode); -} - -static int -fat_encode_fh(struct inode *inode, __u32 *fh, int *lenp, struct inode *parent) -{ - int len = *lenp; - struct msdos_sb_info *sbi = MSDOS_SB(inode->i_sb); - loff_t i_pos; - - if (len < 5) { - *lenp = 5; - return 255; /* no room */ - } - - i_pos = fat_i_pos_read(sbi, inode); - *lenp = 5; - fh[0] = inode->i_ino; - fh[1] = inode->i_generation; - fh[2] = i_pos >> 8; - fh[3] = ((i_pos & 0xf0) << 24) | MSDOS_I(inode)->i_logstart; - fh[4] = (i_pos & 0x0f) << 28; - if (parent) - fh[4] |= MSDOS_I(parent)->i_logstart; - return 3; -} - -static struct dentry *fat_get_parent(struct dentry *child) -{ - struct super_block *sb = child->d_sb; - struct buffer_head *bh; - struct msdos_dir_entry *de; - loff_t i_pos; - struct dentry *parent; - struct inode *inode; - int err; - - lock_super(sb); - - err = fat_get_dotdot_entry(child->d_inode, &bh, &de, &i_pos); - if (err) { - parent = ERR_PTR(err); - goto out; - } - inode = fat_build_inode(sb, de, i_pos); - brelse(bh); - - parent = d_obtain_alias(inode); -out: - unlock_super(sb); - - return parent; -} - static const struct export_operations fat_export_ops = { - .encode_fh = fat_encode_fh, .fh_to_dentry = fat_fh_to_dentry, + .fh_to_parent = fat_fh_to_parent, .get_parent = fat_get_parent, }; @@ -836,6 +754,8 @@ static int fat_show_options(struct seq_file *m, struct dentry *root) seq_puts(m, ",usefree"); if (opts->quiet) seq_puts(m, ",quiet"); + if (opts->nfs) + seq_puts(m, ",nfs"); if (opts->showexec) seq_puts(m, ",showexec"); if (opts->sys_immutable) @@ -880,7 +800,7 @@ enum { Opt_shortname_winnt, Opt_shortname_mixed, Opt_utf8_no, Opt_utf8_yes, Opt_uni_xl_no, Opt_uni_xl_yes, Opt_nonumtail_no, Opt_nonumtail_yes, Opt_obsolete, Opt_flush, Opt_tz_utc, Opt_rodir, Opt_err_cont, - Opt_err_panic, Opt_err_ro, Opt_discard, Opt_err, + Opt_err_panic, Opt_err_ro, Opt_discard, Opt_nfs, Opt_err, }; static const match_table_t fat_tokens = { @@ -909,6 +829,7 @@ static const match_table_t fat_tokens = { {Opt_err_panic, "errors=panic"}, {Opt_err_ro, "errors=remount-ro"}, {Opt_discard, "discard"}, + {Opt_nfs, "nfs"}, {Opt_obsolete, "conv=binary"}, {Opt_obsolete, "conv=text"}, {Opt_obsolete, "conv=auto"}, @@ -989,6 +910,7 @@ static int parse_options(struct super_block *sb, char *options, int is_vfat, opts->numtail = 1; opts->usefree = opts->nocase = 0; opts->tz_utc = 0; + opts->nfs = 0; opts->errors = FAT_ERRORS_RO; *debug = 0; @@ -1153,6 +1075,9 @@ static int parse_options(struct super_block *sb, char *options, int is_vfat, case Opt_discard: opts->discard = 1; break; + case Opt_nfs: + opts->nfs = 1; + break; /* obsolete mount options */ case Opt_obsolete: @@ -1343,6 +1268,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, int isvfat, b = (struct fat_boot_sector *) bh->b_data; } + mutex_init(&sbi->s_lock); sbi->cluster_size = sb->s_blocksize * sbi->sec_per_clus; sbi->cluster_bits = ffs(sbi->cluster_size) - 1; sbi->fats = b->fats; @@ -1443,6 +1369,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, int isvfat, /* set up enough so that it can read an inode */ fat_hash_init(sb); + dir_hash_init(sb); fat_ent_access_init(sb); /* @@ -1497,6 +1424,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, int isvfat, } error = -ENOMEM; insert_inode_hash(root_inode); + fat_attach(root_inode, 0); sb->s_root = d_make_root(root_inode); if (!sb->s_root) { fat_msg(sb, KERN_ERR, "get root inode failed"); @@ -1536,18 +1464,14 @@ static int writeback_inode(struct inode *inode) { int ret; - struct address_space *mapping = inode->i_mapping; - struct writeback_control wbc = { - .sync_mode = WB_SYNC_NONE, - .nr_to_write = 0, - }; - /* if we used WB_SYNC_ALL, sync_inode waits for the io for the - * inode to finish. So WB_SYNC_NONE is sent down to sync_inode + + /* if we used wait=1, sync_inode_metadata waits for the io for the + * inode to finish. So wait=0 is sent down to sync_inode_metadata * and filemap_fdatawrite is used for the data blocks */ - ret = sync_inode(inode, &wbc); + ret = sync_inode_metadata(inode, 0); if (!ret) - ret = filemap_fdatawrite(mapping); + ret = filemap_fdatawrite(inode->i_mapping); return ret; } diff --git a/fs/fat/namei_msdos.c b/fs/fat/namei_msdos.c index b0e12bf9f4a1..e2cfda94a28d 100644 --- a/fs/fat/namei_msdos.c +++ b/fs/fat/namei_msdos.c @@ -208,7 +208,7 @@ static struct dentry *msdos_lookup(struct inode *dir, struct dentry *dentry, struct inode *inode; int err; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); err = msdos_find(dir, dentry->d_name.name, dentry->d_name.len, &sinfo); switch (err) { case -ENOENT: @@ -221,7 +221,7 @@ static struct dentry *msdos_lookup(struct inode *dir, struct dentry *dentry, default: inode = ERR_PTR(err); } - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); return d_splice_alias(inode, dentry); } @@ -273,7 +273,7 @@ static int msdos_create(struct inode *dir, struct dentry *dentry, umode_t mode, unsigned char msdos_name[MSDOS_NAME]; int err, is_hid; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); err = msdos_format_name(dentry->d_name.name, dentry->d_name.len, msdos_name, &MSDOS_SB(sb)->options); @@ -302,7 +302,7 @@ static int msdos_create(struct inode *dir, struct dentry *dentry, umode_t mode, d_instantiate(dentry, inode); out: - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); if (!err) err = fat_flush_inodes(sb, dir, inode); return err; @@ -316,7 +316,7 @@ static int msdos_rmdir(struct inode *dir, struct dentry *dentry) struct fat_slot_info sinfo; int err; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); /* * Check whether the directory is not in use, then check * whether it is empty. @@ -337,7 +337,7 @@ static int msdos_rmdir(struct inode *dir, struct dentry *dentry) inode->i_ctime = CURRENT_TIME_SEC; fat_detach(inode); out: - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); if (!err) err = fat_flush_inodes(sb, dir, inode); @@ -354,7 +354,7 @@ static int msdos_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) struct timespec ts; int err, is_hid, cluster; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); err = msdos_format_name(dentry->d_name.name, dentry->d_name.len, msdos_name, &MSDOS_SB(sb)->options); @@ -392,14 +392,14 @@ static int msdos_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) d_instantiate(dentry, inode); - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); fat_flush_inodes(sb, dir, inode); return 0; out_free: fat_free_clusters(dir, cluster); out: - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); return err; } @@ -407,11 +407,11 @@ out: static int msdos_unlink(struct inode *dir, struct dentry *dentry) { struct inode *inode = dentry->d_inode; - struct super_block *sb= inode->i_sb; + struct super_block *sb = inode->i_sb; struct fat_slot_info sinfo; int err; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); err = msdos_find(dir, dentry->d_name.name, dentry->d_name.len, &sinfo); if (err) goto out; @@ -423,7 +423,7 @@ static int msdos_unlink(struct inode *dir, struct dentry *dentry) inode->i_ctime = CURRENT_TIME_SEC; fat_detach(inode); out: - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); if (!err) err = fat_flush_inodes(sb, dir, inode); @@ -440,7 +440,7 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name, struct inode *old_inode, *new_inode; struct fat_slot_info old_sinfo, sinfo; struct timespec ts; - loff_t dotdot_i_pos, new_i_pos; + loff_t new_i_pos; int err, old_attrs, is_dir, update_dotdot, corrupt = 0; old_sinfo.bh = sinfo.bh = dotdot_bh = NULL; @@ -456,8 +456,7 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name, is_dir = S_ISDIR(old_inode->i_mode); update_dotdot = (is_dir && old_dir != new_dir); if (update_dotdot) { - if (fat_get_dotdot_entry(old_inode, &dotdot_bh, &dotdot_de, - &dotdot_i_pos) < 0) { + if (fat_get_dotdot_entry(old_inode, &dotdot_bh, &dotdot_de)) { err = -EIO; goto out; } @@ -607,7 +606,7 @@ static int msdos_rename(struct inode *old_dir, struct dentry *old_dentry, unsigned char old_msdos_name[MSDOS_NAME], new_msdos_name[MSDOS_NAME]; int err, is_hid; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); err = msdos_format_name(old_dentry->d_name.name, old_dentry->d_name.len, old_msdos_name, @@ -626,7 +625,7 @@ static int msdos_rename(struct inode *old_dir, struct dentry *old_dentry, err = do_msdos_rename(old_dir, old_msdos_name, old_dentry, new_dir, new_msdos_name, new_dentry, is_hid); out: - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); if (!err) err = fat_flush_inodes(sb, old_dir, new_dir); return err; diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c index 6a6d8c0715a1..ac959d655e7d 100644 --- a/fs/fat/namei_vfat.c +++ b/fs/fat/namei_vfat.c @@ -721,7 +721,7 @@ static struct dentry *vfat_lookup(struct inode *dir, struct dentry *dentry, struct dentry *alias; int err; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); err = vfat_find(dir, &dentry->d_name, &sinfo); if (err) { @@ -752,13 +752,13 @@ static struct dentry *vfat_lookup(struct inode *dir, struct dentry *dentry, if (!S_ISDIR(inode->i_mode)) d_move(alias, dentry); iput(inode); - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); return alias; } else dput(alias); out: - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); dentry->d_time = dentry->d_parent->d_inode->i_version; dentry = d_splice_alias(inode, dentry); if (dentry) @@ -766,7 +766,7 @@ out: return dentry; error: - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); return ERR_PTR(err); } @@ -779,7 +779,7 @@ static int vfat_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct timespec ts; int err; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); ts = CURRENT_TIME_SEC; err = vfat_add_entry(dir, &dentry->d_name, 0, 0, &ts, &sinfo); @@ -800,7 +800,7 @@ static int vfat_create(struct inode *dir, struct dentry *dentry, umode_t mode, dentry->d_time = dentry->d_parent->d_inode->i_version; d_instantiate(dentry, inode); out: - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); return err; } @@ -811,7 +811,7 @@ static int vfat_rmdir(struct inode *dir, struct dentry *dentry) struct fat_slot_info sinfo; int err; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); err = fat_dir_empty(inode); if (err) @@ -829,7 +829,7 @@ static int vfat_rmdir(struct inode *dir, struct dentry *dentry) inode->i_mtime = inode->i_atime = CURRENT_TIME_SEC; fat_detach(inode); out: - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); return err; } @@ -841,7 +841,7 @@ static int vfat_unlink(struct inode *dir, struct dentry *dentry) struct fat_slot_info sinfo; int err; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); err = vfat_find(dir, &dentry->d_name, &sinfo); if (err) @@ -854,7 +854,7 @@ static int vfat_unlink(struct inode *dir, struct dentry *dentry) inode->i_mtime = inode->i_atime = CURRENT_TIME_SEC; fat_detach(inode); out: - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); return err; } @@ -867,7 +867,7 @@ static int vfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) struct timespec ts; int err, cluster; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); ts = CURRENT_TIME_SEC; cluster = fat_alloc_new_dir(dir, &ts); @@ -896,13 +896,13 @@ static int vfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) dentry->d_time = dentry->d_parent->d_inode->i_version; d_instantiate(dentry, inode); - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); return 0; out_free: fat_free_clusters(dir, cluster); out: - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); return err; } @@ -914,14 +914,14 @@ static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *old_inode, *new_inode; struct fat_slot_info old_sinfo, sinfo; struct timespec ts; - loff_t dotdot_i_pos, new_i_pos; + loff_t new_i_pos; int err, is_dir, update_dotdot, corrupt = 0; struct super_block *sb = old_dir->i_sb; old_sinfo.bh = sinfo.bh = dotdot_bh = NULL; old_inode = old_dentry->d_inode; new_inode = new_dentry->d_inode; - lock_super(sb); + mutex_lock(&MSDOS_SB(sb)->s_lock); err = vfat_find(old_dir, &old_dentry->d_name, &old_sinfo); if (err) goto out; @@ -929,8 +929,7 @@ static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry, is_dir = S_ISDIR(old_inode->i_mode); update_dotdot = (is_dir && old_dir != new_dir); if (update_dotdot) { - if (fat_get_dotdot_entry(old_inode, &dotdot_bh, &dotdot_de, - &dotdot_i_pos) < 0) { + if (fat_get_dotdot_entry(old_inode, &dotdot_bh, &dotdot_de)) { err = -EIO; goto out; } @@ -997,7 +996,7 @@ out: brelse(sinfo.bh); brelse(dotdot_bh); brelse(old_sinfo.bh); - unlock_super(sb); + mutex_unlock(&MSDOS_SB(sb)->s_lock); return err; diff --git a/fs/fat/nfs.c b/fs/fat/nfs.c new file mode 100644 index 000000000000..ef4b5faba87b --- /dev/null +++ b/fs/fat/nfs.c @@ -0,0 +1,101 @@ +/* fs/fat/nfs.c + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ + +#include <linux/exportfs.h> +#include "fat.h" + +/** + * Look up a directory inode given its starting cluster. + */ +static struct inode *fat_dget(struct super_block *sb, int i_logstart) +{ + struct msdos_sb_info *sbi = MSDOS_SB(sb); + struct hlist_head *head; + struct hlist_node *_p; + struct msdos_inode_info *i; + struct inode *inode = NULL; + + head = sbi->dir_hashtable + fat_dir_hash(i_logstart); + spin_lock(&sbi->dir_hash_lock); + hlist_for_each_entry(i, _p, head, i_dir_hash) { + BUG_ON(i->vfs_inode.i_sb != sb); + if (i->i_logstart != i_logstart) + continue; + inode = igrab(&i->vfs_inode); + if (inode) + break; + } + spin_unlock(&sbi->dir_hash_lock); + return inode; +} + +static struct inode *fat_nfs_get_inode(struct super_block *sb, + u64 ino, u32 generation) +{ + struct inode *inode; + + if ((ino < MSDOS_ROOT_INO) || (ino == MSDOS_FSINFO_INO)) + return NULL; + + inode = ilookup(sb, ino); + if (inode && generation && (inode->i_generation != generation)) { + iput(inode); + inode = NULL; + } + + return inode; +} + +/** + * Map a NFS file handle to a corresponding dentry. + * The dentry may or may not be connected to the filesystem root. + */ +struct dentry *fat_fh_to_dentry(struct super_block *sb, struct fid *fid, + int fh_len, int fh_type) +{ + return generic_fh_to_dentry(sb, fid, fh_len, fh_type, + fat_nfs_get_inode); +} + +/* + * Find the parent for a file specified by NFS handle. + * This requires that the handle contain the i_ino of the parent. + */ +struct dentry *fat_fh_to_parent(struct super_block *sb, struct fid *fid, + int fh_len, int fh_type) +{ + return generic_fh_to_parent(sb, fid, fh_len, fh_type, + fat_nfs_get_inode); +} + +/* + * Find the parent for a directory that is not currently connected to + * the filesystem root. + * + * On entry, the caller holds child_dir->d_inode->i_mutex. + */ +struct dentry *fat_get_parent(struct dentry *child_dir) +{ + struct super_block *sb = child_dir->d_sb; + struct buffer_head *bh = NULL; + struct msdos_dir_entry *de; + struct inode *parent_inode = NULL; + + if (!fat_get_dotdot_entry(child_dir->d_inode, &bh, &de)) { + int parent_logstart = fat_get_start(MSDOS_SB(sb), de); + parent_inode = fat_dget(sb, parent_logstart); + } + brelse(bh); + + return d_obtain_alias(parent_inode); +} diff --git a/fs/fcntl.c b/fs/fcntl.c index 8f704291d4ed..71a600a19f06 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -258,7 +258,7 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, err = f_dupfd(arg, filp, 0); break; case F_DUPFD_CLOEXEC: - err = f_dupfd(arg, filp, FD_CLOEXEC); + err = f_dupfd(arg, filp, O_CLOEXEC); break; case F_GETFD: err = get_close_on_exec(fd) ? FD_CLOEXEC : 0; diff --git a/fs/file.c b/fs/file.c index 0f1bda4bebfa..d3b5fa80b71b 100644 --- a/fs/file.c +++ b/fs/file.c @@ -922,6 +922,9 @@ SYSCALL_DEFINE3(dup3, unsigned int, oldfd, unsigned int, newfd, int, flags) if ((flags & ~O_CLOEXEC) != 0) return -EINVAL; + if (unlikely(oldfd == newfd)) + return -EINVAL; + if (newfd >= rlimit(RLIMIT_NOFILE)) return -EMFILE; diff --git a/fs/file_table.c b/fs/file_table.c index dac67923330f..a72bf9ddd0d2 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -36,7 +36,7 @@ struct files_stat_struct files_stat = { .max_files = NR_FILE }; -DEFINE_LGLOCK(files_lglock); +DEFINE_STATIC_LGLOCK(files_lglock); /* SLAB cache for file structures */ static struct kmem_cache *filp_cachep __read_mostly; diff --git a/fs/filesystems.c b/fs/filesystems.c index 96f24286667a..da165f6adcbf 100644 --- a/fs/filesystems.c +++ b/fs/filesystems.c @@ -124,7 +124,7 @@ EXPORT_SYMBOL(unregister_filesystem); static int fs_index(const char __user * __name) { struct file_system_type * tmp; - char * name; + struct filename *name; int err, index; name = getname(__name); @@ -135,7 +135,7 @@ static int fs_index(const char __user * __name) err = -EINVAL; read_lock(&file_systems_lock); for (tmp=file_systems, index=0 ; tmp ; tmp=tmp->next, index++) { - if (strcmp(tmp->name,name) == 0) { + if (strcmp(tmp->name, name->name) == 0) { err = index; break; } diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 6d46c0d78338..51ea267d444c 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -63,6 +63,7 @@ int writeback_in_progress(struct backing_dev_info *bdi) { return test_bit(BDI_writeback_running, &bdi->state); } +EXPORT_SYMBOL(writeback_in_progress); static inline struct backing_dev_info *inode_to_bdi(struct inode *inode) { @@ -248,7 +249,7 @@ static bool inode_dirtied_after(struct inode *inode, unsigned long t) } /* - * Move expired (dirtied after work->older_than_this) dirty inodes from + * Move expired (dirtied before work->older_than_this) dirty inodes from * @delaying_queue to @dispatch_queue. */ static int move_expired_inodes(struct list_head *delaying_queue, @@ -438,8 +439,7 @@ static void requeue_inode(struct inode *inode, struct bdi_writeback *wb, * setting I_SYNC flag and calling inode_sync_complete() to clear it. */ static int -__writeback_single_inode(struct inode *inode, struct bdi_writeback *wb, - struct writeback_control *wbc) +__writeback_single_inode(struct inode *inode, struct writeback_control *wbc) { struct address_space *mapping = inode->i_mapping; long nr_to_write = wbc->nr_to_write; @@ -526,7 +526,7 @@ writeback_single_inode(struct inode *inode, struct bdi_writeback *wb, inode->i_state |= I_SYNC; spin_unlock(&inode->i_lock); - ret = __writeback_single_inode(inode, wb, wbc); + ret = __writeback_single_inode(inode, wbc); spin_lock(&wb->list_lock); spin_lock(&inode->i_lock); @@ -669,7 +669,7 @@ static long writeback_sb_inodes(struct super_block *sb, * We use I_SYNC to pin the inode in memory. While it is set * evict_inode() will wait so the inode cannot be freed. */ - __writeback_single_inode(inode, wb, &wbc); + __writeback_single_inode(inode, &wbc); work->nr_pages -= write_chunk - wbc.nr_to_write; wrote += write_chunk - wbc.nr_to_write; diff --git a/fs/fuse/file.c b/fs/fuse/file.c index aba15f1b7ad2..78d2837bc940 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -1379,6 +1379,7 @@ static const struct vm_operations_struct fuse_file_vm_ops = { .close = fuse_vma_close, .fault = filemap_fault, .page_mkwrite = fuse_page_mkwrite, + .remap_pages = generic_file_remap_pages, }; static int fuse_file_mmap(struct file *file, struct vm_area_struct *vma) diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c index e8ed6d4a6181..4767774a5f3e 100644 --- a/fs/gfs2/export.c +++ b/fs/gfs2/export.c @@ -161,6 +161,8 @@ static struct dentry *gfs2_fh_to_dentry(struct super_block *sb, struct fid *fid, case GFS2_SMALL_FH_SIZE: case GFS2_LARGE_FH_SIZE: case GFS2_OLD_FH_SIZE: + if (fh_len < GFS2_SMALL_FH_SIZE) + return NULL; this.no_formal_ino = ((u64)be32_to_cpu(fh[0])) << 32; this.no_formal_ino |= be32_to_cpu(fh[1]); this.no_addr = ((u64)be32_to_cpu(fh[2])) << 32; @@ -180,6 +182,8 @@ static struct dentry *gfs2_fh_to_parent(struct super_block *sb, struct fid *fid, switch (fh_type) { case GFS2_LARGE_FH_SIZE: case GFS2_OLD_FH_SIZE: + if (fh_len < GFS2_LARGE_FH_SIZE) + return NULL; parent.no_formal_ino = ((u64)be32_to_cpu(fh[4])) << 32; parent.no_formal_ino |= be32_to_cpu(fh[5]); parent.no_addr = ((u64)be32_to_cpu(fh[6])) << 32; diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index 30e21997a1a1..0def0504afc1 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -492,6 +492,7 @@ out: static const struct vm_operations_struct gfs2_vm_ops = { .fault = filemap_fault, .page_mkwrite = gfs2_page_mkwrite, + .remap_pages = generic_file_remap_pages, }; /** @@ -526,7 +527,6 @@ static int gfs2_mmap(struct file *file, struct vm_area_struct *vma) return error; } vma->vm_ops = &gfs2_vm_ops; - vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } diff --git a/fs/hostfs/hostfs.h b/fs/hostfs/hostfs.h index 1fe731337f07..9c88da0e855a 100644 --- a/fs/hostfs/hostfs.h +++ b/fs/hostfs/hostfs.h @@ -1,7 +1,7 @@ #ifndef __UM_FS_HOSTFS #define __UM_FS_HOSTFS -#include "os.h" +#include <os.h> /* * These are exactly the same definitions as in fs.h, but the names are diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 6c9f3a9d5e21..457addc5c91f 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -16,8 +16,8 @@ #include <linux/mount.h> #include <linux/namei.h> #include "hostfs.h" -#include "init.h" -#include "kern.h" +#include <init.h> +#include <kern.h> struct hostfs_inode_info { int fd; @@ -848,9 +848,11 @@ int hostfs_setattr(struct dentry *dentry, struct iattr *attr) attr->ia_size != i_size_read(inode)) { int error; - error = vmtruncate(inode, attr->ia_size); - if (err) - return err; + error = inode_newsize_ok(inode, attr->ia_size); + if (error) + return error; + + truncate_setsize(inode, attr->ia_size); } setattr_copy(inode, attr); diff --git a/fs/hostfs/hostfs_user.c b/fs/hostfs/hostfs_user.c index a74ad0d371c2..67838f3aa20a 100644 --- a/fs/hostfs/hostfs_user.c +++ b/fs/hostfs/hostfs_user.c @@ -15,7 +15,6 @@ #include <sys/types.h> #include <sys/vfs.h> #include "hostfs.h" -#include "os.h" #include <utime.h> static void stat64_to_hostfs(const struct stat64 *buf, struct hostfs_stat *p) diff --git a/fs/hpfs/anode.c b/fs/hpfs/anode.c index 4bae4a4a60b1..2d5b254ad9e2 100644 --- a/fs/hpfs/anode.c +++ b/fs/hpfs/anode.c @@ -102,7 +102,7 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi return -1; } if (hpfs_alloc_if_possible(s, se = le32_to_cpu(btree->u.external[n].disk_secno) + le32_to_cpu(btree->u.external[n].length))) { - btree->u.external[n].length = cpu_to_le32(le32_to_cpu(btree->u.external[n].length) + 1); + le32_add_cpu(&btree->u.external[n].length, 1); mark_buffer_dirty(bh); brelse(bh); return se; @@ -153,7 +153,7 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi btree = &anode->btree; } btree->n_free_nodes--; n = btree->n_used_nodes++; - btree->first_free = cpu_to_le16(le16_to_cpu(btree->first_free) + 12); + le16_add_cpu(&btree->first_free, 12); btree->u.external[n].disk_secno = cpu_to_le32(se); btree->u.external[n].file_secno = cpu_to_le32(fs); btree->u.external[n].length = cpu_to_le32(1); @@ -174,7 +174,7 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi } if (btree->n_free_nodes) { btree->n_free_nodes--; n = btree->n_used_nodes++; - btree->first_free = cpu_to_le16(le16_to_cpu(btree->first_free) + 8); + le16_add_cpu(&btree->first_free, 8); btree->u.internal[n].file_secno = cpu_to_le32(-1); btree->u.internal[n].down = cpu_to_le32(na); btree->u.internal[n-1].file_secno = cpu_to_le32(fs); diff --git a/fs/hpfs/dnode.c b/fs/hpfs/dnode.c index 3228c524ebe5..4364b2a02c5d 100644 --- a/fs/hpfs/dnode.c +++ b/fs/hpfs/dnode.c @@ -145,10 +145,10 @@ static void set_last_pointer(struct super_block *s, struct dnode *d, dnode_secno } } if (ptr) { - d->first_free = cpu_to_le32(le32_to_cpu(d->first_free) + 4); + le32_add_cpu(&d->first_free, 4); if (le32_to_cpu(d->first_free) > 2048) { hpfs_error(s, "set_last_pointer: too long dnode %08x", le32_to_cpu(d->self)); - d->first_free = cpu_to_le32(le32_to_cpu(d->first_free) - 4); + le32_add_cpu(&d->first_free, -4); return; } de->length = cpu_to_le16(36); @@ -184,7 +184,7 @@ struct hpfs_dirent *hpfs_add_de(struct super_block *s, struct dnode *d, de->not_8x3 = hpfs_is_name_long(name, namelen); de->namelen = namelen; memcpy(de->name, name, namelen); - d->first_free = cpu_to_le32(le32_to_cpu(d->first_free) + d_size); + le32_add_cpu(&d->first_free, d_size); return de; } @@ -314,7 +314,7 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno, set_last_pointer(i->i_sb, ad, de->down ? de_down_pointer(de) : 0); de = de_next_de(de); memmove((char *)nd + 20, de, le32_to_cpu(nd->first_free) + (char *)nd - (char *)de); - nd->first_free = cpu_to_le32(le32_to_cpu(nd->first_free) - ((char *)de - (char *)nd - 20)); + le32_add_cpu(&nd->first_free, -((char *)de - (char *)nd - 20)); memcpy(d, nd, le32_to_cpu(nd->first_free)); for_all_poss(i, hpfs_pos_del, (loff_t)dno << 4, pos); fix_up_ptrs(i->i_sb, ad); @@ -474,8 +474,8 @@ static secno move_to_top(struct inode *i, dnode_secno from, dnode_secno to) hpfs_brelse4(&qbh); return 0; } - dnode->first_free = cpu_to_le32(le32_to_cpu(dnode->first_free) - 4); - de->length = cpu_to_le16(le16_to_cpu(de->length) - 4); + le32_add_cpu(&dnode->first_free, -4); + le16_add_cpu(&de->length, -4); de->down = 0; hpfs_mark_4buffers_dirty(&qbh); dno = up; @@ -570,8 +570,8 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno) for_all_poss(i, hpfs_pos_subst, ((loff_t)dno << 4) | 1, ((loff_t)up << 4) | p); if (!down) { de->down = 0; - de->length = cpu_to_le16(le16_to_cpu(de->length) - 4); - dnode->first_free = cpu_to_le32(le32_to_cpu(dnode->first_free) - 4); + le16_add_cpu(&de->length, -4); + le32_add_cpu(&dnode->first_free, -4); memmove(de_next_de(de), (char *)de_next_de(de) + 4, (char *)dnode + le32_to_cpu(dnode->first_free) - (char *)de_next_de(de)); } else { @@ -647,14 +647,14 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno) printk("HPFS: warning: unbalanced dnode tree, see hpfs.txt 4 more info\n"); printk("HPFS: warning: goin'on\n"); } - del->length = cpu_to_le16(le16_to_cpu(del->length) + 4); + le16_add_cpu(&del->length, 4); del->down = 1; - d1->first_free = cpu_to_le32(le32_to_cpu(d1->first_free) + 4); + le32_add_cpu(&d1->first_free, 4); } if (dlp && !down) { - del->length = cpu_to_le16(le16_to_cpu(del->length) - 4); + le16_add_cpu(&del->length, -4); del->down = 0; - d1->first_free = cpu_to_le32(le32_to_cpu(d1->first_free) - 4); + le32_add_cpu(&d1->first_free, -4); } else if (down) *(__le32 *) ((void *) del + le16_to_cpu(del->length) - 4) = cpu_to_le32(down); } else goto endm; @@ -668,9 +668,9 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno) memcpy(de_cp, de_prev, le16_to_cpu(de_prev->length)); hpfs_delete_de(i->i_sb, dnode, de_prev); if (!de_prev->down) { - de_prev->length = cpu_to_le16(le16_to_cpu(de_prev->length) + 4); + le16_add_cpu(&de_prev->length, 4); de_prev->down = 1; - dnode->first_free = cpu_to_le32(le32_to_cpu(dnode->first_free) + 4); + le32_add_cpu(&dnode->first_free, 4); } *(__le32 *) ((void *) de_prev + le16_to_cpu(de_prev->length) - 4) = cpu_to_le32(ndown); hpfs_mark_4buffers_dirty(&qbh); diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index bc28bf077a6a..a3076228523d 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -398,7 +398,6 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data) *flags |= MS_NOATIME; hpfs_lock(s); - lock_super(s); uid = sbi->sb_uid; gid = sbi->sb_gid; umask = 0777 & ~sbi->sb_mode; lowercase = sbi->sb_lowercase; @@ -431,12 +430,10 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data) replace_mount_options(s, new_opts); - unlock_super(s); hpfs_unlock(s); return 0; out_err: - unlock_super(s); hpfs_unlock(s); kfree(new_opts); return -EINVAL; diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c index c1dffe47fde2..78f21f8dc2ec 100644 --- a/fs/hppfs/hppfs.c +++ b/fs/hppfs/hppfs.c @@ -18,7 +18,7 @@ #include <linux/pid_namespace.h> #include <linux/namei.h> #include <asm/uaccess.h> -#include "os.h" +#include <os.h> static struct inode *get_inode(struct super_block *, struct dentry *); @@ -674,7 +674,7 @@ static struct inode *get_inode(struct super_block *sb, struct dentry *dentry) if (!inode) { dput(dentry); - return ERR_PTR(-ENOMEM); + return NULL; } if (S_ISDIR(dentry->d_inode->i_mode)) { diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 9460120a5170..c5bc355d8243 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -110,7 +110,7 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) * way when do_mmap_pgoff unwinds (may be important on powerpc * and ia64). */ - vma->vm_flags |= VM_HUGETLB | VM_RESERVED; + vma->vm_flags |= VM_HUGETLB | VM_DONTEXPAND | VM_DONTDUMP; vma->vm_ops = &hugetlb_vm_ops; if (vma->vm_pgoff & (~huge_page_mask(h) >> PAGE_SHIFT)) @@ -397,17 +397,16 @@ static void hugetlbfs_evict_inode(struct inode *inode) } static inline void -hugetlb_vmtruncate_list(struct prio_tree_root *root, pgoff_t pgoff) +hugetlb_vmtruncate_list(struct rb_root *root, pgoff_t pgoff) { struct vm_area_struct *vma; - struct prio_tree_iter iter; - vma_prio_tree_foreach(vma, &iter, root, pgoff, ULONG_MAX) { + vma_interval_tree_foreach(vma, root, pgoff, ULONG_MAX) { unsigned long v_offset; /* * Can the expression below overflow on 32-bit arches? - * No, because the prio_tree returns us only those vmas + * No, because the interval tree returns us only those vmas * which overlap the truncated area starting at pgoff, * and no vma on a 32-bit arch can span beyond the 4GB. */ @@ -432,7 +431,7 @@ static int hugetlb_vmtruncate(struct inode *inode, loff_t offset) i_size_write(inode, offset); mutex_lock(&mapping->i_mmap_mutex); - if (!prio_tree_empty(&mapping->i_mmap)) + if (!RB_EMPTY_ROOT(&mapping->i_mmap)) hugetlb_vmtruncate_list(&mapping->i_mmap, pgoff); mutex_unlock(&mapping->i_mmap_mutex); truncate_hugepages(inode, offset); diff --git a/fs/inode.c b/fs/inode.c index ac8d904b3f16..b03c71957246 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -348,7 +348,7 @@ void address_space_init_once(struct address_space *mapping) mutex_init(&mapping->i_mmap_mutex); INIT_LIST_HEAD(&mapping->private_list); spin_lock_init(&mapping->private_lock); - INIT_RAW_PRIO_TREE_ROOT(&mapping->i_mmap); + mapping->i_mmap = RB_ROOT; INIT_LIST_HEAD(&mapping->i_mmap_nonlinear); } EXPORT_SYMBOL(address_space_init_once); diff --git a/fs/internal.h b/fs/internal.h index 371bcc4b1697..916b7cbf3e3e 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -97,8 +97,8 @@ struct open_flags { int acc_mode; int intent; }; -extern struct file *do_filp_open(int dfd, const char *pathname, - const struct open_flags *op, int lookup_flags); +extern struct file *do_filp_open(int dfd, struct filename *pathname, + const struct open_flags *op, int flags); extern struct file *do_file_open_root(struct dentry *, struct vfsmount *, const char *, const struct open_flags *, int lookup_flags); diff --git a/fs/isofs/export.c b/fs/isofs/export.c index 1d3804492aa7..2b4f2358eadb 100644 --- a/fs/isofs/export.c +++ b/fs/isofs/export.c @@ -175,7 +175,7 @@ static struct dentry *isofs_fh_to_parent(struct super_block *sb, { struct isofs_fid *ifid = (struct isofs_fid *)fid; - if (fh_type != 2) + if (fh_len < 2 || fh_type != 2) return NULL; return isofs_export_iget(sb, diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index af5280fb579b..3091d42992f0 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -1014,17 +1014,35 @@ restart_loop: * there's no point in keeping a checkpoint record for * it. */ - /* A buffer which has been freed while still being - * journaled by a previous transaction may end up still - * being dirty here, but we want to avoid writing back - * that buffer in the future after the "add to orphan" - * operation been committed, That's not only a performance - * gain, it also stops aliasing problems if the buffer is - * left behind for writeback and gets reallocated for another - * use in a different page. */ - if (buffer_freed(bh) && !jh->b_next_transaction) { - clear_buffer_freed(bh); - clear_buffer_jbddirty(bh); + /* + * A buffer which has been freed while still being journaled by + * a previous transaction. + */ + if (buffer_freed(bh)) { + /* + * If the running transaction is the one containing + * "add to orphan" operation (b_next_transaction != + * NULL), we have to wait for that transaction to + * commit before we can really get rid of the buffer. + * So just clear b_modified to not confuse transaction + * credit accounting and refile the buffer to + * BJ_Forget of the running transaction. If the just + * committed transaction contains "add to orphan" + * operation, we can completely invalidate the buffer + * now. We are rather through in that since the + * buffer may be still accessible when blocksize < + * pagesize and it is attached to the last partial + * page. + */ + jh->b_modified = 0; + if (!jh->b_next_transaction) { + clear_buffer_freed(bh); + clear_buffer_jbddirty(bh); + clear_buffer_mapped(bh); + clear_buffer_new(bh); + clear_buffer_req(bh); + bh->b_bdev = NULL; + } } if (buffer_jbddirty(bh)) { diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index e149b99a7ffb..484b8d1c6cb6 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1354,6 +1354,11 @@ static void jbd2_mark_journal_empty(journal_t *journal) BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex)); read_lock(&journal->j_state_lock); + /* Is it already empty? */ + if (sb->s_start == 0) { + read_unlock(&journal->j_state_lock); + return; + } jbd_debug(1, "JBD2: Marking journal as empty (seq %d)\n", journal->j_tail_sequence); diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index 0131e4362534..626846bac32f 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -289,8 +289,11 @@ int jbd2_journal_recover(journal_t *journal) if (!err) err = err2; /* Make sure all replayed data is on permanent storage */ - if (journal->j_flags & JBD2_BARRIER) - blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL); + if (journal->j_flags & JBD2_BARRIER) { + err2 = blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL); + if (!err) + err = err2; + } return err; } diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index fb1ab9533b67..a74ba4659549 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -1841,15 +1841,16 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction) * We're outside-transaction here. Either or both of j_running_transaction * and j_committing_transaction may be NULL. */ -static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) +static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh, + int partial_page) { transaction_t *transaction; struct journal_head *jh; int may_free = 1; - int ret; BUFFER_TRACE(bh, "entry"); +retry: /* * It is safe to proceed here without the j_list_lock because the * buffers cannot be stolen by try_to_free_buffers as long as we are @@ -1878,10 +1879,18 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) * clear the buffer dirty bit at latest at the moment when the * transaction marking the buffer as freed in the filesystem * structures is committed because from that moment on the - * buffer can be reallocated and used by a different page. + * block can be reallocated and used by a different page. * Since the block hasn't been freed yet but the inode has * already been added to orphan list, it is safe for us to add * the buffer to BJ_Forget list of the newest transaction. + * + * Also we have to clear buffer_mapped flag of a truncated buffer + * because the buffer_head may be attached to the page straddling + * i_size (can happen only when blocksize < pagesize) and thus the + * buffer_head can be reused when the file is extended again. So we end + * up keeping around invalidated buffers attached to transactions' + * BJ_Forget list just to stop checkpointing code from cleaning up + * the transaction this buffer was modified in. */ transaction = jh->b_transaction; if (transaction == NULL) { @@ -1908,13 +1917,9 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) * committed, the buffer won't be needed any * longer. */ JBUFFER_TRACE(jh, "checkpointed: add to BJ_Forget"); - ret = __dispose_buffer(jh, + may_free = __dispose_buffer(jh, journal->j_running_transaction); - jbd2_journal_put_journal_head(jh); - spin_unlock(&journal->j_list_lock); - jbd_unlock_bh_state(bh); - write_unlock(&journal->j_state_lock); - return ret; + goto zap_buffer; } else { /* There is no currently-running transaction. So the * orphan record which we wrote for this file must have @@ -1922,13 +1927,9 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) * the committing transaction, if it exists. */ if (journal->j_committing_transaction) { JBUFFER_TRACE(jh, "give to committing trans"); - ret = __dispose_buffer(jh, + may_free = __dispose_buffer(jh, journal->j_committing_transaction); - jbd2_journal_put_journal_head(jh); - spin_unlock(&journal->j_list_lock); - jbd_unlock_bh_state(bh); - write_unlock(&journal->j_state_lock); - return ret; + goto zap_buffer; } else { /* The orphan record's transaction has * committed. We can cleanse this buffer */ @@ -1940,10 +1941,24 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) JBUFFER_TRACE(jh, "on committing transaction"); /* * The buffer is committing, we simply cannot touch - * it. So we just set j_next_transaction to the - * running transaction (if there is one) and mark - * buffer as freed so that commit code knows it should - * clear dirty bits when it is done with the buffer. + * it. If the page is straddling i_size we have to wait + * for commit and try again. + */ + if (partial_page) { + tid_t tid = journal->j_committing_transaction->t_tid; + + jbd2_journal_put_journal_head(jh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + write_unlock(&journal->j_state_lock); + jbd2_log_wait_commit(journal, tid); + goto retry; + } + /* + * OK, buffer won't be reachable after truncate. We just set + * j_next_transaction to the running transaction (if there is + * one) and mark buffer as freed so that commit code knows it + * should clear dirty bits when it is done with the buffer. */ set_buffer_freed(bh); if (journal->j_running_transaction && buffer_jbddirty(bh)) @@ -1966,6 +1981,15 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) } zap_buffer: + /* + * This is tricky. Although the buffer is truncated, it may be reused + * if blocksize < pagesize and it is attached to the page straddling + * EOF. Since the buffer might have been added to BJ_Forget list of the + * running transaction, journal_get_write_access() won't clear + * b_modified and credit accounting gets confused. So clear b_modified + * here. + */ + jh->b_modified = 0; jbd2_journal_put_journal_head(jh); zap_buffer_no_jh: spin_unlock(&journal->j_list_lock); @@ -2017,7 +2041,8 @@ void jbd2_journal_invalidatepage(journal_t *journal, if (offset <= curr_off) { /* This block is wholly outside the truncation point */ lock_buffer(bh); - may_free &= journal_unmap_buffer(journal, bh); + may_free &= journal_unmap_buffer(journal, bh, + offset > 0); unlock_buffer(bh); } curr_off = next_off; diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c index 1ea349fff68b..ae81b01e6fd7 100644 --- a/fs/jffs2/readinode.c +++ b/fs/jffs2/readinode.c @@ -394,8 +394,11 @@ static int jffs2_add_tn_to_tree(struct jffs2_sb_info *c, } /* Trivial function to remove the last node in the tree. Which by definition - has no right-hand -- so can be removed just by making its only child (if - any) take its place under its parent. */ + has no right-hand child — so can be removed just by making its left-hand + child (if any) take its place under its parent. Since this is only done + when we're consuming the whole tree, there's no need to use rb_erase() + and let it worry about adjusting colours and balancing the tree. That + would just be a waste of time. */ static void eat_last(struct rb_root *root, struct rb_node *node) { struct rb_node *parent = rb_parent(node); @@ -412,12 +415,12 @@ static void eat_last(struct rb_root *root, struct rb_node *node) link = &parent->rb_right; *link = node->rb_left; - /* Colour doesn't matter now. Only the parent pointer. */ if (node->rb_left) - node->rb_left->rb_parent_color = node->rb_parent_color; + node->rb_left->__rb_parent_color = node->__rb_parent_color; } -/* We put this in reverse order, so we can just use eat_last */ +/* We put the version tree in reverse order, so we can use the same eat_last() + function that we use to consume the tmpnode tree (tn_root). */ static void ver_insert(struct rb_root *ver_root, struct jffs2_tmp_dnode_info *tn) { struct rb_node **link = &ver_root->rb_node; diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index ff487954cd96..d3d8799e2187 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -100,6 +100,10 @@ static int jffs2_sync_fs(struct super_block *sb, int wait) { struct jffs2_sb_info *c = JFFS2_SB_INFO(sb); +#ifdef CONFIG_JFFS2_FS_WRITEBUFFER + cancel_delayed_work_sync(&c->wbuf_dwork); +#endif + mutex_lock(&c->alloc_sem); jffs2_flush_wbuf_pad(c); mutex_unlock(&c->alloc_sem); diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c index 6f4529d3697f..a6597d60d76d 100644 --- a/fs/jffs2/wbuf.c +++ b/fs/jffs2/wbuf.c @@ -1044,10 +1044,10 @@ int jffs2_check_oob_empty(struct jffs2_sb_info *c, ops.datbuf = NULL; ret = mtd_read_oob(c->mtd, jeb->offset, &ops); - if (ret || ops.oobretlen != ops.ooblen) { + if ((ret && !mtd_is_bitflip(ret)) || ops.oobretlen != ops.ooblen) { pr_err("cannot read OOB for EB at %08x, requested %zd bytes, read %zd bytes, error %d\n", jeb->offset, ops.ooblen, ops.oobretlen, ret); - if (!ret) + if (!ret || mtd_is_bitflip(ret)) ret = -EIO; return ret; } @@ -1086,10 +1086,10 @@ int jffs2_check_nand_cleanmarker(struct jffs2_sb_info *c, ops.datbuf = NULL; ret = mtd_read_oob(c->mtd, jeb->offset, &ops); - if (ret || ops.oobretlen != ops.ooblen) { + if ((ret && !mtd_is_bitflip(ret)) || ops.oobretlen != ops.ooblen) { pr_err("cannot read OOB for EB at %08x, requested %zd bytes, read %zd bytes, error %d\n", jeb->offset, ops.ooblen, ops.oobretlen, ret); - if (!ret) + if (!ret || mtd_is_bitflip(ret)) ret = -EIO; return ret; } diff --git a/fs/jfs/jfs_discard.c b/fs/jfs/jfs_discard.c index 9947563e4175..dfcd50304559 100644 --- a/fs/jfs/jfs_discard.c +++ b/fs/jfs/jfs_discard.c @@ -83,7 +83,7 @@ int jfs_ioc_trim(struct inode *ip, struct fstrim_range *range) struct bmap *bmp = JFS_SBI(ip->i_sb)->bmap; struct super_block *sb = ipbmap->i_sb; int agno, agno_end; - s64 start, end, minlen; + u64 start, end, minlen; u64 trimmed = 0; /** @@ -93,15 +93,19 @@ int jfs_ioc_trim(struct inode *ip, struct fstrim_range *range) * minlen: minimum extent length in Bytes */ start = range->start >> sb->s_blocksize_bits; - if (start < 0) - start = 0; end = start + (range->len >> sb->s_blocksize_bits) - 1; - if (end >= bmp->db_mapsize) - end = bmp->db_mapsize - 1; minlen = range->minlen >> sb->s_blocksize_bits; - if (minlen <= 0) + if (minlen == 0) minlen = 1; + if (minlen > bmp->db_agsize || + start >= bmp->db_mapsize || + range->len < sb->s_blocksize) + return -EINVAL; + + if (end >= bmp->db_mapsize) + end = bmp->db_mapsize - 1; + /** * we trim all ag's within the range */ diff --git a/fs/lockd/clntxdr.c b/fs/lockd/clntxdr.c index d269ada7670e..982d2676e1f8 100644 --- a/fs/lockd/clntxdr.c +++ b/fs/lockd/clntxdr.c @@ -223,7 +223,7 @@ static void encode_nlm_stat(struct xdr_stream *xdr, { __be32 *p; - BUG_ON(be32_to_cpu(stat) > NLM_LCK_DENIED_GRACE_PERIOD); + WARN_ON_ONCE(be32_to_cpu(stat) > NLM_LCK_DENIED_GRACE_PERIOD); p = xdr_reserve_space(xdr, 4); *p = stat; } diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c index 7ef14b3c5bee..3d7e09bcc0e9 100644 --- a/fs/lockd/mon.c +++ b/fs/lockd/mon.c @@ -7,7 +7,6 @@ */ #include <linux/types.h> -#include <linux/utsname.h> #include <linux/kernel.h> #include <linux/ktime.h> #include <linux/slab.h> @@ -19,6 +18,8 @@ #include <asm/unaligned.h> +#include "netns.h" + #define NLMDBG_FACILITY NLMDBG_MONITOR #define NSM_PROGRAM 100024 #define NSM_VERSION 1 @@ -40,6 +41,7 @@ struct nsm_args { u32 proc; char *mon_name; + char *nodename; }; struct nsm_res { @@ -70,7 +72,7 @@ static struct rpc_clnt *nsm_create(struct net *net) }; struct rpc_create_args args = { .net = net, - .protocol = XPRT_TRANSPORT_UDP, + .protocol = XPRT_TRANSPORT_TCP, .address = (struct sockaddr *)&sin, .addrsize = sizeof(sin), .servername = "rpc.statd", @@ -83,10 +85,61 @@ static struct rpc_clnt *nsm_create(struct net *net) return rpc_create(&args); } +static struct rpc_clnt *nsm_client_set(struct lockd_net *ln, + struct rpc_clnt *clnt) +{ + spin_lock(&ln->nsm_clnt_lock); + if (ln->nsm_users == 0) { + if (clnt == NULL) + goto out; + ln->nsm_clnt = clnt; + } + clnt = ln->nsm_clnt; + ln->nsm_users++; +out: + spin_unlock(&ln->nsm_clnt_lock); + return clnt; +} + +static struct rpc_clnt *nsm_client_get(struct net *net) +{ + struct rpc_clnt *clnt, *new; + struct lockd_net *ln = net_generic(net, lockd_net_id); + + clnt = nsm_client_set(ln, NULL); + if (clnt != NULL) + goto out; + + clnt = new = nsm_create(net); + if (IS_ERR(clnt)) + goto out; + + clnt = nsm_client_set(ln, new); + if (clnt != new) + rpc_shutdown_client(new); +out: + return clnt; +} + +static void nsm_client_put(struct net *net) +{ + struct lockd_net *ln = net_generic(net, lockd_net_id); + struct rpc_clnt *clnt = NULL; + + spin_lock(&ln->nsm_clnt_lock); + ln->nsm_users--; + if (ln->nsm_users == 0) { + clnt = ln->nsm_clnt; + ln->nsm_clnt = NULL; + } + spin_unlock(&ln->nsm_clnt_lock); + if (clnt != NULL) + rpc_shutdown_client(clnt); +} + static int nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res, - struct net *net) + struct rpc_clnt *clnt) { - struct rpc_clnt *clnt; int status; struct nsm_args args = { .priv = &nsm->sm_priv, @@ -94,31 +147,24 @@ static int nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res, .vers = 3, .proc = NLMPROC_NSM_NOTIFY, .mon_name = nsm->sm_mon_name, + .nodename = clnt->cl_nodename, }; struct rpc_message msg = { .rpc_argp = &args, .rpc_resp = res, }; - clnt = nsm_create(net); - if (IS_ERR(clnt)) { - status = PTR_ERR(clnt); - dprintk("lockd: failed to create NSM upcall transport, " - "status=%d\n", status); - goto out; - } + BUG_ON(clnt == NULL); memset(res, 0, sizeof(*res)); msg.rpc_proc = &clnt->cl_procinfo[proc]; - status = rpc_call_sync(clnt, &msg, 0); + status = rpc_call_sync(clnt, &msg, RPC_TASK_SOFTCONN); if (status < 0) dprintk("lockd: NSM upcall RPC failed, status=%d\n", status); else status = 0; - rpc_shutdown_client(clnt); - out: return status; } @@ -138,6 +184,7 @@ int nsm_monitor(const struct nlm_host *host) struct nsm_handle *nsm = host->h_nsmhandle; struct nsm_res res; int status; + struct rpc_clnt *clnt; dprintk("lockd: nsm_monitor(%s)\n", nsm->sm_name); @@ -150,7 +197,15 @@ int nsm_monitor(const struct nlm_host *host) */ nsm->sm_mon_name = nsm_use_hostnames ? nsm->sm_name : nsm->sm_addrbuf; - status = nsm_mon_unmon(nsm, NSMPROC_MON, &res, host->net); + clnt = nsm_client_get(host->net); + if (IS_ERR(clnt)) { + status = PTR_ERR(clnt); + dprintk("lockd: failed to create NSM upcall transport, " + "status=%d, net=%p\n", status, host->net); + return status; + } + + status = nsm_mon_unmon(nsm, NSMPROC_MON, &res, clnt); if (unlikely(res.status != 0)) status = -EIO; if (unlikely(status < 0)) { @@ -182,9 +237,11 @@ void nsm_unmonitor(const struct nlm_host *host) if (atomic_read(&nsm->sm_count) == 1 && nsm->sm_monitored && !nsm->sm_sticky) { + struct lockd_net *ln = net_generic(host->net, lockd_net_id); + dprintk("lockd: nsm_unmonitor(%s)\n", nsm->sm_name); - status = nsm_mon_unmon(nsm, NSMPROC_UNMON, &res, host->net); + status = nsm_mon_unmon(nsm, NSMPROC_UNMON, &res, ln->nsm_clnt); if (res.status != 0) status = -EIO; if (status < 0) @@ -192,6 +249,8 @@ void nsm_unmonitor(const struct nlm_host *host) nsm->sm_name); else nsm->sm_monitored = 0; + + nsm_client_put(host->net); } } @@ -430,7 +489,7 @@ static void encode_my_id(struct xdr_stream *xdr, const struct nsm_args *argp) { __be32 *p; - encode_nsm_string(xdr, utsname()->nodename); + encode_nsm_string(xdr, argp->nodename); p = xdr_reserve_space(xdr, 4 + 4 + 4); *p++ = cpu_to_be32(argp->prog); *p++ = cpu_to_be32(argp->vers); diff --git a/fs/lockd/netns.h b/fs/lockd/netns.h index 4eee248ba96e..5010b55628b4 100644 --- a/fs/lockd/netns.h +++ b/fs/lockd/netns.h @@ -12,6 +12,10 @@ struct lockd_net { struct delayed_work grace_period_end; struct lock_manager lockd_manager; struct list_head grace_list; + + spinlock_t nsm_clnt_lock; + unsigned int nsm_users; + struct rpc_clnt *nsm_clnt; }; extern int lockd_net_id; diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index 31a63f87b806..a2aa97d45670 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -126,7 +126,7 @@ static void restart_grace(void) static int lockd(void *vrqstp) { - int err = 0, preverr = 0; + int err = 0; struct svc_rqst *rqstp = vrqstp; /* try_to_freeze() is called from svc_recv() */ @@ -165,21 +165,8 @@ lockd(void *vrqstp) * recvfrom routine. */ err = svc_recv(rqstp, timeout); - if (err == -EAGAIN || err == -EINTR) { - preverr = err; + if (err == -EAGAIN || err == -EINTR) continue; - } - if (err < 0) { - if (err != preverr) { - printk(KERN_WARNING "%s: unexpected error " - "from svc_recv (%d)\n", __func__, err); - preverr = err; - } - schedule_timeout_interruptible(HZ); - continue; - } - preverr = err; - dprintk("lockd: request from %s\n", svc_print_addr(rqstp, buf, sizeof(buf))); @@ -596,6 +583,7 @@ static int lockd_init_net(struct net *net) INIT_DELAYED_WORK(&ln->grace_period_end, grace_ender); INIT_LIST_HEAD(&ln->grace_list); + spin_lock_init(&ln->nsm_clnt_lock); return 0; } diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 3009a365e082..21171f0c6477 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -68,7 +68,8 @@ nlmsvc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain file pointer. Not used by FREE_ALL call. */ if (filp != NULL) { - if ((error = nlm_lookup_file(rqstp, &file, &lock->fh)) != 0) + error = cast_status(nlm_lookup_file(rqstp, &file, &lock->fh)); + if (error != 0) goto no_locks; *filp = file; diff --git a/fs/locks.c b/fs/locks.c index abc7dc6c490b..a94e331a52a2 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -1289,7 +1289,7 @@ EXPORT_SYMBOL(__break_lease); void lease_get_mtime(struct inode *inode, struct timespec *time) { struct file_lock *flock = inode->i_flock; - if (flock && IS_LEASE(flock) && (flock->fl_type & F_WRLCK)) + if (flock && IS_LEASE(flock) && (flock->fl_type == F_WRLCK)) *time = current_fs_time(inode->i_sb); else *time = inode->i_mtime; @@ -2185,8 +2185,8 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl, } else { seq_printf(f, "%s ", (lease_breaking(fl)) - ? (fl->fl_type & F_UNLCK) ? "UNLCK" : "READ " - : (fl->fl_type & F_WRLCK) ? "WRITE" : "READ "); + ? (fl->fl_type == F_UNLCK) ? "UNLCK" : "READ " + : (fl->fl_type == F_WRLCK) ? "WRITE" : "READ "); } if (inode) { #ifdef WE_CAN_BREAK_LSLK_NOW diff --git a/fs/namei.c b/fs/namei.c index aa30d19e9edd..937f9d50c84b 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -117,18 +117,70 @@ * POSIX.1 2.4: an empty pathname is invalid (ENOENT). * PATH_MAX includes the nul terminator --RR. */ -static char *getname_flags(const char __user *filename, int flags, int *empty) +void final_putname(struct filename *name) { - char *result = __getname(), *err; + if (name->separate) { + __putname(name->name); + kfree(name); + } else { + __putname(name); + } +} + +#define EMBEDDED_NAME_MAX (PATH_MAX - sizeof(struct filename)) + +static struct filename * +getname_flags(const char __user *filename, int flags, int *empty) +{ + struct filename *result, *err; int len; + long max; + char *kname; + result = audit_reusename(filename); + if (result) + return result; + + result = __getname(); if (unlikely(!result)) return ERR_PTR(-ENOMEM); - len = strncpy_from_user(result, filename, PATH_MAX); - err = ERR_PTR(len); - if (unlikely(len < 0)) + /* + * First, try to embed the struct filename inside the names_cache + * allocation + */ + kname = (char *)result + sizeof(*result); + result->name = kname; + result->separate = false; + max = EMBEDDED_NAME_MAX; + +recopy: + len = strncpy_from_user(kname, filename, max); + if (unlikely(len < 0)) { + err = ERR_PTR(len); goto error; + } + + /* + * Uh-oh. We have a name that's approaching PATH_MAX. Allocate a + * separate struct filename so we can dedicate the entire + * names_cache allocation for the pathname, and re-do the copy from + * userland. + */ + if (len == EMBEDDED_NAME_MAX && max == EMBEDDED_NAME_MAX) { + kname = (char *)result; + + result = kzalloc(sizeof(*result), GFP_KERNEL); + if (!result) { + err = ERR_PTR(-ENOMEM); + result = (struct filename *)kname; + goto error; + } + result->name = kname; + result->separate = true; + max = PATH_MAX; + goto recopy; + } /* The empty path is special. */ if (unlikely(!len)) { @@ -140,30 +192,32 @@ static char *getname_flags(const char __user *filename, int flags, int *empty) } err = ERR_PTR(-ENAMETOOLONG); - if (likely(len < PATH_MAX)) { - audit_getname(result); - return result; - } + if (unlikely(len >= PATH_MAX)) + goto error; + + result->uptr = filename; + audit_getname(result); + return result; error: - __putname(result); + final_putname(result); return err; } -char *getname(const char __user * filename) +struct filename * +getname(const char __user * filename) { return getname_flags(filename, 0, NULL); } +EXPORT_SYMBOL(getname); #ifdef CONFIG_AUDITSYSCALL -void putname(const char *name) +void putname(struct filename *name) { if (unlikely(!audit_dummy_context())) - audit_putname(name); - else - __putname(name); + return audit_putname(name); + final_putname(name); } -EXPORT_SYMBOL(putname); #endif static int check_acl(struct inode *inode, int mask) @@ -651,8 +705,8 @@ static inline void put_link(struct nameidata *nd, struct path *link, void *cooki path_put(link); } -int sysctl_protected_symlinks __read_mostly = 1; -int sysctl_protected_hardlinks __read_mostly = 1; +int sysctl_protected_symlinks __read_mostly = 0; +int sysctl_protected_hardlinks __read_mostly = 0; /** * may_follow_link - Check symlink following for unsafe situations @@ -692,9 +746,9 @@ static inline int may_follow_link(struct path *link, struct nameidata *nd) if (uid_eq(parent->i_uid, inode->i_uid)) return 0; + audit_log_link_denied("follow_link", link); path_put_conditional(link, nd); path_put(&nd->path); - audit_log_link_denied("follow_link", link); return -EACCES; } @@ -810,6 +864,7 @@ follow_link(struct path *link, struct nameidata *nd, void **p) return error; out_put_nd_path: + *p = NULL; path_put(&nd->path); path_put(link); return error; @@ -1962,24 +2017,29 @@ static int path_lookupat(int dfd, const char *name, return err; } -static int do_path_lookup(int dfd, const char *name, +static int filename_lookup(int dfd, struct filename *name, unsigned int flags, struct nameidata *nd) { - int retval = path_lookupat(dfd, name, flags | LOOKUP_RCU, nd); + int retval = path_lookupat(dfd, name->name, flags | LOOKUP_RCU, nd); if (unlikely(retval == -ECHILD)) - retval = path_lookupat(dfd, name, flags, nd); + retval = path_lookupat(dfd, name->name, flags, nd); if (unlikely(retval == -ESTALE)) - retval = path_lookupat(dfd, name, flags | LOOKUP_REVAL, nd); + retval = path_lookupat(dfd, name->name, + flags | LOOKUP_REVAL, nd); - if (likely(!retval)) { - if (unlikely(!audit_dummy_context())) { - if (nd->path.dentry && nd->inode) - audit_inode(name, nd->path.dentry); - } - } + if (likely(!retval)) + audit_inode(name, nd->path.dentry, flags & LOOKUP_PARENT); return retval; } +static int do_path_lookup(int dfd, const char *name, + unsigned int flags, struct nameidata *nd) +{ + struct filename filename = { .name = name }; + + return filename_lookup(dfd, &filename, flags, nd); +} + /* does lookup, returns the object with parent locked */ struct dentry *kern_path_locked(const char *name, struct path *path) { @@ -2097,13 +2157,13 @@ int user_path_at_empty(int dfd, const char __user *name, unsigned flags, struct path *path, int *empty) { struct nameidata nd; - char *tmp = getname_flags(name, flags, empty); + struct filename *tmp = getname_flags(name, flags, empty); int err = PTR_ERR(tmp); if (!IS_ERR(tmp)) { BUG_ON(flags & LOOKUP_PARENT); - err = do_path_lookup(dfd, tmp, flags, &nd); + err = filename_lookup(dfd, tmp, flags, &nd); putname(tmp); if (!err) *path = nd.path; @@ -2117,22 +2177,28 @@ int user_path_at(int dfd, const char __user *name, unsigned flags, return user_path_at_empty(dfd, name, flags, path, NULL); } -static int user_path_parent(int dfd, const char __user *path, - struct nameidata *nd, char **name) +/* + * NB: most callers don't do anything directly with the reference to the + * to struct filename, but the nd->last pointer points into the name string + * allocated by getname. So we must hold the reference to it until all + * path-walking is complete. + */ +static struct filename * +user_path_parent(int dfd, const char __user *path, struct nameidata *nd) { - char *s = getname(path); + struct filename *s = getname(path); int error; if (IS_ERR(s)) - return PTR_ERR(s); + return s; - error = do_path_lookup(dfd, s, LOOKUP_PARENT, nd); - if (error) + error = filename_lookup(dfd, s, LOOKUP_PARENT, nd); + if (error) { putname(s); - else - *name = s; + return ERR_PTR(error); + } - return error; + return s; } /* @@ -2179,7 +2245,7 @@ static int may_delete(struct inode *dir,struct dentry *victim,int isdir) return -ENOENT; BUG_ON(victim->d_parent->d_inode != dir); - audit_inode_child(victim, dir); + audit_inode_child(dir, victim, AUDIT_TYPE_CHILD_DELETE); error = inode_permission(dir, MAY_WRITE | MAY_EXEC); if (error) @@ -2624,7 +2690,7 @@ out_dput: */ static int do_last(struct nameidata *nd, struct path *path, struct file *file, const struct open_flags *op, - int *opened, const char *pathname) + int *opened, struct filename *name) { struct dentry *dir = nd->path.dentry; int open_flag = op->open_flag; @@ -2651,7 +2717,7 @@ static int do_last(struct nameidata *nd, struct path *path, error = complete_walk(nd); if (error) return error; - audit_inode(pathname, nd->path.dentry); + audit_inode(name, nd->path.dentry, 0); if (open_flag & O_CREAT) { error = -EISDIR; goto out; @@ -2661,7 +2727,7 @@ static int do_last(struct nameidata *nd, struct path *path, error = complete_walk(nd); if (error) return error; - audit_inode(pathname, dir); + audit_inode(name, dir, 0); goto finish_open; } @@ -2690,7 +2756,7 @@ static int do_last(struct nameidata *nd, struct path *path, if (error) return error; - audit_inode(pathname, dir); + audit_inode(name, dir, 0); error = -EISDIR; /* trailing slashes? */ if (nd->last.name[nd->last.len]) @@ -2720,7 +2786,7 @@ retry_lookup: !S_ISREG(file->f_path.dentry->d_inode->i_mode)) will_truncate = false; - audit_inode(pathname, file->f_path.dentry); + audit_inode(name, file->f_path.dentry, 0); goto opened; } @@ -2737,7 +2803,7 @@ retry_lookup: * create/update audit record if it already exists. */ if (path->dentry->d_inode) - audit_inode(pathname, path->dentry); + audit_inode(name, path->dentry, 0); /* * If atomic_open() acquired write access it is dropped now due to @@ -2802,7 +2868,7 @@ finish_lookup: error = -ENOTDIR; if ((nd->flags & LOOKUP_DIRECTORY) && !nd->inode->i_op->lookup) goto out; - audit_inode(pathname, nd->path.dentry); + audit_inode(name, nd->path.dentry, 0); finish_open: if (!S_ISREG(nd->inode->i_mode)) will_truncate = false; @@ -2870,7 +2936,7 @@ stale_open: goto retry_lookup; } -static struct file *path_openat(int dfd, const char *pathname, +static struct file *path_openat(int dfd, struct filename *pathname, struct nameidata *nd, const struct open_flags *op, int flags) { struct file *base = NULL; @@ -2885,12 +2951,12 @@ static struct file *path_openat(int dfd, const char *pathname, file->f_flags = op->open_flag; - error = path_init(dfd, pathname, flags | LOOKUP_PARENT, nd, &base); + error = path_init(dfd, pathname->name, flags | LOOKUP_PARENT, nd, &base); if (unlikely(error)) goto out; current->total_link_count = 0; - error = link_path_walk(pathname, nd); + error = link_path_walk(pathname->name, nd); if (unlikely(error)) goto out; @@ -2936,7 +3002,7 @@ out: return file; } -struct file *do_filp_open(int dfd, const char *pathname, +struct file *do_filp_open(int dfd, struct filename *pathname, const struct open_flags *op, int flags) { struct nameidata nd; @@ -2955,6 +3021,7 @@ struct file *do_file_open_root(struct dentry *dentry, struct vfsmount *mnt, { struct nameidata nd; struct file *file; + struct filename filename = { .name = name }; nd.root.mnt = mnt; nd.root.dentry = dentry; @@ -2964,11 +3031,11 @@ struct file *do_file_open_root(struct dentry *dentry, struct vfsmount *mnt, if (dentry->d_inode->i_op->follow_link && op->intent & LOOKUP_OPEN) return ERR_PTR(-ELOOP); - file = path_openat(-1, name, &nd, op, flags | LOOKUP_RCU); + file = path_openat(-1, &filename, &nd, op, flags | LOOKUP_RCU); if (unlikely(file == ERR_PTR(-ECHILD))) - file = path_openat(-1, name, &nd, op, flags); + file = path_openat(-1, &filename, &nd, op, flags); if (unlikely(file == ERR_PTR(-ESTALE))) - file = path_openat(-1, name, &nd, op, flags | LOOKUP_REVAL); + file = path_openat(-1, &filename, &nd, op, flags | LOOKUP_REVAL); return file; } @@ -3043,11 +3110,11 @@ EXPORT_SYMBOL(done_path_create); struct dentry *user_path_create(int dfd, const char __user *pathname, struct path *path, int is_dir) { - char *tmp = getname(pathname); + struct filename *tmp = getname(pathname); struct dentry *res; if (IS_ERR(tmp)) return ERR_CAST(tmp); - res = kern_path_create(dfd, tmp, path, is_dir); + res = kern_path_create(dfd, tmp->name, path, is_dir); putname(tmp); return res; } @@ -3252,13 +3319,13 @@ out: static long do_rmdir(int dfd, const char __user *pathname) { int error = 0; - char * name; + struct filename *name; struct dentry *dentry; struct nameidata nd; - error = user_path_parent(dfd, pathname, &nd, &name); - if (error) - return error; + name = user_path_parent(dfd, pathname, &nd); + if (IS_ERR(name)) + return PTR_ERR(name); switch(nd.last_type) { case LAST_DOTDOT: @@ -3347,14 +3414,14 @@ int vfs_unlink(struct inode *dir, struct dentry *dentry) static long do_unlinkat(int dfd, const char __user *pathname) { int error; - char *name; + struct filename *name; struct dentry *dentry; struct nameidata nd; struct inode *inode = NULL; - error = user_path_parent(dfd, pathname, &nd, &name); - if (error) - return error; + name = user_path_parent(dfd, pathname, &nd); + if (IS_ERR(name)) + return PTR_ERR(name); error = -EISDIR; if (nd.last_type != LAST_NORM) @@ -3438,7 +3505,7 @@ SYSCALL_DEFINE3(symlinkat, const char __user *, oldname, int, newdfd, const char __user *, newname) { int error; - char *from; + struct filename *from; struct dentry *dentry; struct path path; @@ -3451,9 +3518,9 @@ SYSCALL_DEFINE3(symlinkat, const char __user *, oldname, if (IS_ERR(dentry)) goto out_putname; - error = security_path_symlink(&path, dentry, from); + error = security_path_symlink(&path, dentry, from->name); if (!error) - error = vfs_symlink(path.dentry->d_inode, dentry, from); + error = vfs_symlink(path.dentry->d_inode, dentry, from->name); done_path_create(&path, dentry); out_putname: putname(from); @@ -3733,17 +3800,21 @@ SYSCALL_DEFINE4(renameat, int, olddfd, const char __user *, oldname, struct dentry *old_dentry, *new_dentry; struct dentry *trap; struct nameidata oldnd, newnd; - char *from; - char *to; + struct filename *from; + struct filename *to; int error; - error = user_path_parent(olddfd, oldname, &oldnd, &from); - if (error) + from = user_path_parent(olddfd, oldname, &oldnd); + if (IS_ERR(from)) { + error = PTR_ERR(from); goto exit; + } - error = user_path_parent(newdfd, newname, &newnd, &to); - if (error) + to = user_path_parent(newdfd, newname, &newnd); + if (IS_ERR(to)) { + error = PTR_ERR(to); goto exit1; + } error = -EXDEV; if (oldnd.path.mnt != newnd.path.mnt) @@ -3967,7 +4038,6 @@ EXPORT_SYMBOL(follow_down_one); EXPORT_SYMBOL(follow_down); EXPORT_SYMBOL(follow_up); EXPORT_SYMBOL(get_write_access); /* nfsd */ -EXPORT_SYMBOL(getname); EXPORT_SYMBOL(lock_rename); EXPORT_SYMBOL(lookup_one_len); EXPORT_SYMBOL(page_follow_link_light); diff --git a/fs/namespace.c b/fs/namespace.c index 7bdf7907413f..24960626bb6b 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1640,7 +1640,7 @@ static int do_change_type(struct path *path, int flag) /* * do loopback mount. */ -static int do_loopback(struct path *path, char *old_name, +static int do_loopback(struct path *path, const char *old_name, int recurse) { LIST_HEAD(umount_list); @@ -1764,7 +1764,7 @@ static inline int tree_contains_unbindable(struct mount *mnt) return 0; } -static int do_move_mount(struct path *path, char *old_name) +static int do_move_mount(struct path *path, const char *old_name) { struct path old_path, parent_path; struct mount *p; @@ -1917,8 +1917,8 @@ unlock: * create a new mount for userspace and request it to be added into the * namespace's tree */ -static int do_new_mount(struct path *path, char *type, int flags, - int mnt_flags, char *name, void *data) +static int do_new_mount(struct path *path, const char *type, int flags, + int mnt_flags, const char *name, void *data) { struct vfsmount *mnt; int err; @@ -2191,8 +2191,8 @@ int copy_mount_string(const void __user *data, char **where) * Therefore, if this magic number is present, it carries no information * and must be discarded. */ -long do_mount(char *dev_name, char *dir_name, char *type_page, - unsigned long flags, void *data_page) +long do_mount(const char *dev_name, const char *dir_name, + const char *type_page, unsigned long flags, void *data_page) { struct path path; int retval = 0; @@ -2408,7 +2408,7 @@ SYSCALL_DEFINE5(mount, char __user *, dev_name, char __user *, dir_name, { int ret; char *kernel_type; - char *kernel_dir; + struct filename *kernel_dir; char *kernel_dev; unsigned long data_page; @@ -2430,7 +2430,7 @@ SYSCALL_DEFINE5(mount, char __user *, dev_name, char __user *, dir_name, if (ret < 0) goto out_data; - ret = do_mount(kernel_dev, kernel_dir, kernel_type, flags, + ret = do_mount(kernel_dev, kernel_dir->name, kernel_type, flags, (void *) data_page); free_page(data_page); diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig index db7ad719628a..13ca196385f5 100644 --- a/fs/nfs/Kconfig +++ b/fs/nfs/Kconfig @@ -95,8 +95,8 @@ config NFS_SWAP This option enables swapon to work on files located on NFS mounts. config NFS_V4_1 - bool "NFS client support for NFSv4.1 (EXPERIMENTAL)" - depends on NFS_V4 && EXPERIMENTAL + bool "NFS client support for NFSv4.1" + depends on NFS_V4 select SUNRPC_BACKCHANNEL help This option enables support for minor version 1 of the NFSv4 protocol diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c index dd392ed5f2e2..f1027b06a1a9 100644 --- a/fs/nfs/blocklayout/blocklayout.c +++ b/fs/nfs/blocklayout/blocklayout.c @@ -37,6 +37,7 @@ #include <linux/bio.h> /* struct bio */ #include <linux/buffer_head.h> /* various write calls */ #include <linux/prefetch.h> +#include <linux/pagevec.h> #include "../pnfs.h" #include "../internal.h" @@ -162,25 +163,39 @@ static struct bio *bl_alloc_init_bio(int npg, sector_t isect, return bio; } -static struct bio *bl_add_page_to_bio(struct bio *bio, int npg, int rw, +static struct bio *do_add_page_to_bio(struct bio *bio, int npg, int rw, sector_t isect, struct page *page, struct pnfs_block_extent *be, void (*end_io)(struct bio *, int err), - struct parallel_io *par) + struct parallel_io *par, + unsigned int offset, int len) { + isect = isect + (offset >> SECTOR_SHIFT); + dprintk("%s: npg %d rw %d isect %llu offset %u len %d\n", __func__, + npg, rw, (unsigned long long)isect, offset, len); retry: if (!bio) { bio = bl_alloc_init_bio(npg, isect, be, end_io, par); if (!bio) return ERR_PTR(-ENOMEM); } - if (bio_add_page(bio, page, PAGE_CACHE_SIZE, 0) < PAGE_CACHE_SIZE) { + if (bio_add_page(bio, page, len, offset) < len) { bio = bl_submit_bio(rw, bio); goto retry; } return bio; } +static struct bio *bl_add_page_to_bio(struct bio *bio, int npg, int rw, + sector_t isect, struct page *page, + struct pnfs_block_extent *be, + void (*end_io)(struct bio *, int err), + struct parallel_io *par) +{ + return do_add_page_to_bio(bio, npg, rw, isect, page, be, + end_io, par, 0, PAGE_CACHE_SIZE); +} + /* This is basically copied from mpage_end_io_read */ static void bl_end_io_read(struct bio *bio, int err) { @@ -228,14 +243,6 @@ bl_end_par_io_read(void *data, int unused) schedule_work(&rdata->task.u.tk_work); } -static bool -bl_check_alignment(u64 offset, u32 len, unsigned long blkmask) -{ - if ((offset & blkmask) || (len & blkmask)) - return false; - return true; -} - static enum pnfs_try_status bl_read_pagelist(struct nfs_read_data *rdata) { @@ -246,15 +253,15 @@ bl_read_pagelist(struct nfs_read_data *rdata) sector_t isect, extent_length = 0; struct parallel_io *par; loff_t f_offset = rdata->args.offset; + size_t bytes_left = rdata->args.count; + unsigned int pg_offset, pg_len; struct page **pages = rdata->args.pages; int pg_index = rdata->args.pgbase >> PAGE_CACHE_SHIFT; + const bool is_dio = (header->dreq != NULL); dprintk("%s enter nr_pages %u offset %lld count %u\n", __func__, rdata->pages.npages, f_offset, (unsigned int)rdata->args.count); - if (!bl_check_alignment(f_offset, rdata->args.count, PAGE_CACHE_MASK)) - goto use_mds; - par = alloc_parallel(rdata); if (!par) goto use_mds; @@ -284,36 +291,53 @@ bl_read_pagelist(struct nfs_read_data *rdata) extent_length = min(extent_length, cow_length); } } + + if (is_dio) { + pg_offset = f_offset & ~PAGE_CACHE_MASK; + if (pg_offset + bytes_left > PAGE_CACHE_SIZE) + pg_len = PAGE_CACHE_SIZE - pg_offset; + else + pg_len = bytes_left; + + f_offset += pg_len; + bytes_left -= pg_len; + isect += (pg_offset >> SECTOR_SHIFT); + } else { + pg_offset = 0; + pg_len = PAGE_CACHE_SIZE; + } + hole = is_hole(be, isect); if (hole && !cow_read) { bio = bl_submit_bio(READ, bio); /* Fill hole w/ zeroes w/o accessing device */ dprintk("%s Zeroing page for hole\n", __func__); - zero_user_segment(pages[i], 0, PAGE_CACHE_SIZE); + zero_user_segment(pages[i], pg_offset, pg_len); print_page(pages[i]); SetPageUptodate(pages[i]); } else { struct pnfs_block_extent *be_read; be_read = (hole && cow_read) ? cow_read : be; - bio = bl_add_page_to_bio(bio, rdata->pages.npages - i, + bio = do_add_page_to_bio(bio, rdata->pages.npages - i, READ, isect, pages[i], be_read, - bl_end_io_read, par); + bl_end_io_read, par, + pg_offset, pg_len); if (IS_ERR(bio)) { header->pnfs_error = PTR_ERR(bio); bio = NULL; goto out; } } - isect += PAGE_CACHE_SECTORS; + isect += (pg_len >> SECTOR_SHIFT); extent_length -= PAGE_CACHE_SECTORS; } if ((isect << SECTOR_SHIFT) >= header->inode->i_size) { rdata->res.eof = 1; - rdata->res.count = header->inode->i_size - f_offset; + rdata->res.count = header->inode->i_size - rdata->args.offset; } else { - rdata->res.count = (isect << SECTOR_SHIFT) - f_offset; + rdata->res.count = (isect << SECTOR_SHIFT) - rdata->args.offset; } out: bl_put_extent(be); @@ -461,6 +485,106 @@ map_block(struct buffer_head *bh, sector_t isect, struct pnfs_block_extent *be) return; } +static void +bl_read_single_end_io(struct bio *bio, int error) +{ + struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1; + struct page *page = bvec->bv_page; + + /* Only one page in bvec */ + unlock_page(page); +} + +static int +bl_do_readpage_sync(struct page *page, struct pnfs_block_extent *be, + unsigned int offset, unsigned int len) +{ + struct bio *bio; + struct page *shadow_page; + sector_t isect; + char *kaddr, *kshadow_addr; + int ret = 0; + + dprintk("%s: offset %u len %u\n", __func__, offset, len); + + shadow_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM); + if (shadow_page == NULL) + return -ENOMEM; + + bio = bio_alloc(GFP_NOIO, 1); + if (bio == NULL) + return -ENOMEM; + + isect = (page->index << PAGE_CACHE_SECTOR_SHIFT) + + (offset / SECTOR_SIZE); + + bio->bi_sector = isect - be->be_f_offset + be->be_v_offset; + bio->bi_bdev = be->be_mdev; + bio->bi_end_io = bl_read_single_end_io; + + lock_page(shadow_page); + if (bio_add_page(bio, shadow_page, + SECTOR_SIZE, round_down(offset, SECTOR_SIZE)) == 0) { + unlock_page(shadow_page); + bio_put(bio); + return -EIO; + } + + submit_bio(READ, bio); + wait_on_page_locked(shadow_page); + if (unlikely(!test_bit(BIO_UPTODATE, &bio->bi_flags))) { + ret = -EIO; + } else { + kaddr = kmap_atomic(page); + kshadow_addr = kmap_atomic(shadow_page); + memcpy(kaddr + offset, kshadow_addr + offset, len); + kunmap_atomic(kshadow_addr); + kunmap_atomic(kaddr); + } + __free_page(shadow_page); + bio_put(bio); + + return ret; +} + +static int +bl_read_partial_page_sync(struct page *page, struct pnfs_block_extent *be, + unsigned int dirty_offset, unsigned int dirty_len, + bool full_page) +{ + int ret = 0; + unsigned int start, end; + + if (full_page) { + start = 0; + end = PAGE_CACHE_SIZE; + } else { + start = round_down(dirty_offset, SECTOR_SIZE); + end = round_up(dirty_offset + dirty_len, SECTOR_SIZE); + } + + dprintk("%s: offset %u len %d\n", __func__, dirty_offset, dirty_len); + if (!be) { + zero_user_segments(page, start, dirty_offset, + dirty_offset + dirty_len, end); + if (start == 0 && end == PAGE_CACHE_SIZE && + trylock_page(page)) { + SetPageUptodate(page); + unlock_page(page); + } + return ret; + } + + if (start != dirty_offset) + ret = bl_do_readpage_sync(page, be, start, dirty_offset - start); + + if (!ret && (dirty_offset + dirty_len < end)) + ret = bl_do_readpage_sync(page, be, dirty_offset + dirty_len, + end - dirty_offset - dirty_len); + + return ret; +} + /* Given an unmapped page, zero it or read in page for COW, page is locked * by caller. */ @@ -494,7 +618,6 @@ init_page_for_write(struct page *page, struct pnfs_block_extent *cow_read) SetPageUptodate(page); cleanup: - bl_put_extent(cow_read); if (bh) free_buffer_head(bh); if (ret) { @@ -566,6 +689,7 @@ bl_write_pagelist(struct nfs_write_data *wdata, int sync) struct parallel_io *par = NULL; loff_t offset = wdata->args.offset; size_t count = wdata->args.count; + unsigned int pg_offset, pg_len, saved_len; struct page **pages = wdata->args.pages; struct page *page; pgoff_t index; @@ -574,10 +698,13 @@ bl_write_pagelist(struct nfs_write_data *wdata, int sync) NFS_SERVER(header->inode)->pnfs_blksize >> PAGE_CACHE_SHIFT; dprintk("%s enter, %Zu@%lld\n", __func__, count, offset); - /* Check for alignment first */ - if (!bl_check_alignment(offset, count, PAGE_CACHE_MASK)) - goto out_mds; + if (header->dreq != NULL && + (!IS_ALIGNED(offset, NFS_SERVER(header->inode)->pnfs_blksize) || + !IS_ALIGNED(count, NFS_SERVER(header->inode)->pnfs_blksize))) { + dprintk("pnfsblock nonblock aligned DIO writes. Resend MDS\n"); + goto out_mds; + } /* At this point, wdata->pages is a (sequential) list of nfs_pages. * We want to write each, and if there is an error set pnfs_error * to have it redone using nfs. @@ -674,10 +801,11 @@ next_page: if (!extent_length) { /* We've used up the previous extent */ bl_put_extent(be); + bl_put_extent(cow_read); bio = bl_submit_bio(WRITE, bio); /* Get the next one */ be = bl_find_get_extent(BLK_LSEG2EXT(header->lseg), - isect, NULL); + isect, &cow_read); if (!be || !is_writable(be, isect)) { header->pnfs_error = -EINVAL; goto out; @@ -694,7 +822,26 @@ next_page: extent_length = be->be_length - (isect - be->be_f_offset); } - if (be->be_state == PNFS_BLOCK_INVALID_DATA) { + + dprintk("%s offset %lld count %Zu\n", __func__, offset, count); + pg_offset = offset & ~PAGE_CACHE_MASK; + if (pg_offset + count > PAGE_CACHE_SIZE) + pg_len = PAGE_CACHE_SIZE - pg_offset; + else + pg_len = count; + + saved_len = pg_len; + if (be->be_state == PNFS_BLOCK_INVALID_DATA && + !bl_is_sector_init(be->be_inval, isect)) { + ret = bl_read_partial_page_sync(pages[i], cow_read, + pg_offset, pg_len, true); + if (ret) { + dprintk("%s bl_read_partial_page_sync fail %d\n", + __func__, ret); + header->pnfs_error = ret; + goto out; + } + ret = bl_mark_sectors_init(be->be_inval, isect, PAGE_CACHE_SECTORS); if (unlikely(ret)) { @@ -703,15 +850,35 @@ next_page: header->pnfs_error = ret; goto out; } + + /* Expand to full page write */ + pg_offset = 0; + pg_len = PAGE_CACHE_SIZE; + } else if ((pg_offset & (SECTOR_SIZE - 1)) || + (pg_len & (SECTOR_SIZE - 1))){ + /* ahh, nasty case. We have to do sync full sector + * read-modify-write cycles. + */ + unsigned int saved_offset = pg_offset; + ret = bl_read_partial_page_sync(pages[i], be, pg_offset, + pg_len, false); + pg_offset = round_down(pg_offset, SECTOR_SIZE); + pg_len = round_up(saved_offset + pg_len, SECTOR_SIZE) + - pg_offset; } - bio = bl_add_page_to_bio(bio, wdata->pages.npages - i, WRITE, + + + bio = do_add_page_to_bio(bio, wdata->pages.npages - i, WRITE, isect, pages[i], be, - bl_end_io_write, par); + bl_end_io_write, par, + pg_offset, pg_len); if (IS_ERR(bio)) { header->pnfs_error = PTR_ERR(bio); bio = NULL; goto out; } + offset += saved_len; + count -= saved_len; isect += PAGE_CACHE_SECTORS; last_isect = isect; extent_length -= PAGE_CACHE_SECTORS; @@ -729,17 +896,16 @@ next_page: } write_done: - wdata->res.count = (last_isect << SECTOR_SHIFT) - (offset); - if (count < wdata->res.count) { - wdata->res.count = count; - } + wdata->res.count = wdata->args.count; out: bl_put_extent(be); + bl_put_extent(cow_read); bl_submit_bio(WRITE, bio); put_parallel(par); return PNFS_ATTEMPTED; out_mds: bl_put_extent(be); + bl_put_extent(cow_read); kfree(par); return PNFS_NOT_ATTEMPTED; } @@ -874,7 +1040,7 @@ static void free_blk_mountid(struct block_mount_id *mid) } } -/* This is mostly copied from the filelayout's get_device_info function. +/* This is mostly copied from the filelayout_get_device_info function. * It seems much of this should be at the generic pnfs level. */ static struct pnfs_block_dev * @@ -1011,33 +1177,95 @@ bl_clear_layoutdriver(struct nfs_server *server) return 0; } +static bool +is_aligned_req(struct nfs_page *req, unsigned int alignment) +{ + return IS_ALIGNED(req->wb_offset, alignment) && + IS_ALIGNED(req->wb_bytes, alignment); +} + static void bl_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { - if (!bl_check_alignment(req->wb_offset, req->wb_bytes, PAGE_CACHE_MASK)) + if (pgio->pg_dreq != NULL && + !is_aligned_req(req, SECTOR_SIZE)) nfs_pageio_reset_read_mds(pgio); else pnfs_generic_pg_init_read(pgio, req); } +static bool +bl_pg_test_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, + struct nfs_page *req) +{ + if (pgio->pg_dreq != NULL && + !is_aligned_req(req, SECTOR_SIZE)) + return false; + + return pnfs_generic_pg_test(pgio, prev, req); +} + +/* + * Return the number of contiguous bytes for a given inode + * starting at page frame idx. + */ +static u64 pnfs_num_cont_bytes(struct inode *inode, pgoff_t idx) +{ + struct address_space *mapping = inode->i_mapping; + pgoff_t end; + + /* Optimize common case that writes from 0 to end of file */ + end = DIV_ROUND_UP(i_size_read(inode), PAGE_CACHE_SIZE); + if (end != NFS_I(inode)->npages) { + rcu_read_lock(); + end = radix_tree_next_hole(&mapping->page_tree, idx + 1, ULONG_MAX); + rcu_read_unlock(); + } + + if (!end) + return i_size_read(inode) - (idx << PAGE_CACHE_SHIFT); + else + return (end - idx) << PAGE_CACHE_SHIFT; +} + static void bl_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { - if (!bl_check_alignment(req->wb_offset, req->wb_bytes, PAGE_CACHE_MASK)) + if (pgio->pg_dreq != NULL && + !is_aligned_req(req, PAGE_CACHE_SIZE)) { nfs_pageio_reset_write_mds(pgio); - else - pnfs_generic_pg_init_write(pgio, req); + } else { + u64 wb_size; + if (pgio->pg_dreq == NULL) + wb_size = pnfs_num_cont_bytes(pgio->pg_inode, + req->wb_index); + else + wb_size = nfs_dreq_bytes_left(pgio->pg_dreq); + + pnfs_generic_pg_init_write(pgio, req, wb_size); + } +} + +static bool +bl_pg_test_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, + struct nfs_page *req) +{ + if (pgio->pg_dreq != NULL && + !is_aligned_req(req, PAGE_CACHE_SIZE)) + return false; + + return pnfs_generic_pg_test(pgio, prev, req); } static const struct nfs_pageio_ops bl_pg_read_ops = { .pg_init = bl_pg_init_read, - .pg_test = pnfs_generic_pg_test, + .pg_test = bl_pg_test_read, .pg_doio = pnfs_generic_pg_readpages, }; static const struct nfs_pageio_ops bl_pg_write_ops = { .pg_init = bl_pg_init_write, - .pg_test = pnfs_generic_pg_test, + .pg_test = bl_pg_test_write, .pg_doio = pnfs_generic_pg_writepages, }; diff --git a/fs/nfs/blocklayout/blocklayout.h b/fs/nfs/blocklayout/blocklayout.h index 03350690118e..f4891bde8851 100644 --- a/fs/nfs/blocklayout/blocklayout.h +++ b/fs/nfs/blocklayout/blocklayout.h @@ -41,6 +41,7 @@ #define PAGE_CACHE_SECTORS (PAGE_CACHE_SIZE >> SECTOR_SHIFT) #define PAGE_CACHE_SECTOR_SHIFT (PAGE_CACHE_SHIFT - SECTOR_SHIFT) +#define SECTOR_SIZE (1 << SECTOR_SHIFT) struct block_mount_id { spinlock_t bm_lock; /* protects list */ @@ -172,7 +173,6 @@ struct bl_msg_hdr { /* blocklayoutdev.c */ ssize_t bl_pipe_downcall(struct file *, const char __user *, size_t); void bl_pipe_destroy_msg(struct rpc_pipe_msg *); -struct block_device *nfs4_blkdev_get(dev_t dev); int nfs4_blkdev_put(struct block_device *bdev); struct pnfs_block_dev *nfs4_blk_decode_device(struct nfs_server *server, struct pnfs_device *dev); diff --git a/fs/nfs/blocklayout/blocklayoutdev.c b/fs/nfs/blocklayout/blocklayoutdev.c index c96554245ccf..a86c5bdad9e3 100644 --- a/fs/nfs/blocklayout/blocklayoutdev.c +++ b/fs/nfs/blocklayout/blocklayoutdev.c @@ -53,22 +53,6 @@ static int decode_sector_number(__be32 **rp, sector_t *sp) return 0; } -/* Open a block_device by device number. */ -struct block_device *nfs4_blkdev_get(dev_t dev) -{ - struct block_device *bd; - - dprintk("%s enter\n", __func__); - bd = blkdev_get_by_dev(dev, FMODE_READ, NULL); - if (IS_ERR(bd)) - goto fail; - return bd; -fail: - dprintk("%s failed to open device : %ld\n", - __func__, PTR_ERR(bd)); - return NULL; -} - /* * Release the block device */ @@ -172,11 +156,12 @@ nfs4_blk_decode_device(struct nfs_server *server, goto out; } - bd = nfs4_blkdev_get(MKDEV(reply->major, reply->minor)); + bd = blkdev_get_by_dev(MKDEV(reply->major, reply->minor), + FMODE_READ, NULL); if (IS_ERR(bd)) { - rc = PTR_ERR(bd); - dprintk("%s failed to open device : %d\n", __func__, rc); - rv = ERR_PTR(rc); + dprintk("%s failed to open device : %ld\n", __func__, + PTR_ERR(bd)); + rv = ERR_CAST(bd); goto out; } diff --git a/fs/nfs/blocklayout/extents.c b/fs/nfs/blocklayout/extents.c index 1f9a6032796b..9c3e117c3ed1 100644 --- a/fs/nfs/blocklayout/extents.c +++ b/fs/nfs/blocklayout/extents.c @@ -683,8 +683,7 @@ encode_pnfs_block_layoutupdate(struct pnfs_block_layout *bl, p = xdr_encode_hyper(p, lce->bse_length << SECTOR_SHIFT); p = xdr_encode_hyper(p, 0LL); *p++ = cpu_to_be32(PNFS_BLOCK_READWRITE_DATA); - list_del(&lce->bse_node); - list_add_tail(&lce->bse_node, &bl->bl_committing); + list_move_tail(&lce->bse_node, &bl->bl_committing); bl->bl_count--; count++; } diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index 4c8459e5bdee..5088b57b078a 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -12,6 +12,7 @@ #include <linux/sunrpc/svc.h> #include <linux/sunrpc/svcsock.h> #include <linux/nfs_fs.h> +#include <linux/errno.h> #include <linux/mutex.h> #include <linux/freezer.h> #include <linux/kthread.h> @@ -23,6 +24,7 @@ #include "nfs4_fs.h" #include "callback.h" #include "internal.h" +#include "netns.h" #define NFSDBG_FACILITY NFSDBG_CALLBACK @@ -37,7 +39,32 @@ static struct nfs_callback_data nfs_callback_info[NFS4_MAX_MINOR_VERSION + 1]; static DEFINE_MUTEX(nfs_callback_mutex); static struct svc_program nfs4_callback_program; -unsigned short nfs_callback_tcpport6; +static int nfs4_callback_up_net(struct svc_serv *serv, struct net *net) +{ + int ret; + struct nfs_net *nn = net_generic(net, nfs_net_id); + + ret = svc_create_xprt(serv, "tcp", net, PF_INET, + nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS); + if (ret <= 0) + goto out_err; + nn->nfs_callback_tcpport = ret; + dprintk("NFS: Callback listener port = %u (af %u, net %p)\n", + nn->nfs_callback_tcpport, PF_INET, net); + + ret = svc_create_xprt(serv, "tcp", net, PF_INET6, + nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS); + if (ret > 0) { + nn->nfs_callback_tcpport6 = ret; + dprintk("NFS: Callback listener port = %u (af %u, net %p)\n", + nn->nfs_callback_tcpport6, PF_INET6, net); + } else if (ret != -EAFNOSUPPORT) + goto out_err; + return 0; + +out_err: + return (ret) ? ret : -ENOMEM; +} /* * This is the NFSv4 callback kernel thread. @@ -45,7 +72,7 @@ unsigned short nfs_callback_tcpport6; static int nfs4_callback_svc(void *vrqstp) { - int err, preverr = 0; + int err; struct svc_rqst *rqstp = vrqstp; set_freezable(); @@ -55,20 +82,8 @@ nfs4_callback_svc(void *vrqstp) * Listen for a request on the socket */ err = svc_recv(rqstp, MAX_SCHEDULE_TIMEOUT); - if (err == -EAGAIN || err == -EINTR) { - preverr = err; - continue; - } - if (err < 0) { - if (err != preverr) { - printk(KERN_WARNING "NFS: %s: unexpected error " - "from svc_recv (%d)\n", __func__, err); - preverr = err; - } - schedule_timeout_uninterruptible(HZ); + if (err == -EAGAIN || err == -EINTR) continue; - } - preverr = err; svc_process(rqstp); } return 0; @@ -78,38 +93,23 @@ nfs4_callback_svc(void *vrqstp) * Prepare to bring up the NFSv4 callback service */ static struct svc_rqst * -nfs4_callback_up(struct svc_serv *serv, struct rpc_xprt *xprt) +nfs4_callback_up(struct svc_serv *serv) { - int ret; - - ret = svc_create_xprt(serv, "tcp", &init_net, PF_INET, - nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS); - if (ret <= 0) - goto out_err; - nfs_callback_tcpport = ret; - dprintk("NFS: Callback listener port = %u (af %u)\n", - nfs_callback_tcpport, PF_INET); - - ret = svc_create_xprt(serv, "tcp", &init_net, PF_INET6, - nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS); - if (ret > 0) { - nfs_callback_tcpport6 = ret; - dprintk("NFS: Callback listener port = %u (af %u)\n", - nfs_callback_tcpport6, PF_INET6); - } else if (ret == -EAFNOSUPPORT) - ret = 0; - else - goto out_err; - return svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE); - -out_err: - if (ret == 0) - ret = -ENOMEM; - return ERR_PTR(ret); } #if defined(CONFIG_NFS_V4_1) +static int nfs41_callback_up_net(struct svc_serv *serv, struct net *net) +{ + /* + * Create an svc_sock for the back channel service that shares the + * fore channel connection. + * Returns the input port (0) and sets the svc_serv bc_xprt on success + */ + return svc_create_xprt(serv, "tcp-bc", net, PF_INET, 0, + SVC_SOCK_ANONYMOUS); +} + /* * The callback service for NFSv4.1 callbacks */ @@ -149,28 +149,9 @@ nfs41_callback_svc(void *vrqstp) * Bring up the NFSv4.1 callback service */ static struct svc_rqst * -nfs41_callback_up(struct svc_serv *serv, struct rpc_xprt *xprt) +nfs41_callback_up(struct svc_serv *serv) { struct svc_rqst *rqstp; - int ret; - - /* - * Create an svc_sock for the back channel service that shares the - * fore channel connection. - * Returns the input port (0) and sets the svc_serv bc_xprt on success - */ - ret = svc_create_xprt(serv, "tcp-bc", &init_net, PF_INET, 0, - SVC_SOCK_ANONYMOUS); - if (ret < 0) { - rqstp = ERR_PTR(ret); - goto out; - } - - /* - * Save the svc_serv in the transport so that it can - * be referenced when the session backchannel is initialized - */ - xprt->bc_serv = serv; INIT_LIST_HEAD(&serv->sv_cb_list); spin_lock_init(&serv->sv_cb_lock); @@ -180,90 +161,74 @@ nfs41_callback_up(struct svc_serv *serv, struct rpc_xprt *xprt) svc_xprt_put(serv->sv_bc_xprt); serv->sv_bc_xprt = NULL; } -out: dprintk("--> %s return %ld\n", __func__, IS_ERR(rqstp) ? PTR_ERR(rqstp) : 0); return rqstp; } -static inline int nfs_minorversion_callback_svc_setup(u32 minorversion, - struct svc_serv *serv, struct rpc_xprt *xprt, +static void nfs_minorversion_callback_svc_setup(struct svc_serv *serv, struct svc_rqst **rqstpp, int (**callback_svc)(void *vrqstp)) { - if (minorversion) { - *rqstpp = nfs41_callback_up(serv, xprt); - *callback_svc = nfs41_callback_svc; - } - return minorversion; + *rqstpp = nfs41_callback_up(serv); + *callback_svc = nfs41_callback_svc; } static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt, - struct nfs_callback_data *cb_info) + struct svc_serv *serv) { if (minorversion) - xprt->bc_serv = cb_info->serv; + /* + * Save the svc_serv in the transport so that it can + * be referenced when the session backchannel is initialized + */ + xprt->bc_serv = serv; } #else -static inline int nfs_minorversion_callback_svc_setup(u32 minorversion, - struct svc_serv *serv, struct rpc_xprt *xprt, - struct svc_rqst **rqstpp, int (**callback_svc)(void *vrqstp)) +static int nfs41_callback_up_net(struct svc_serv *serv, struct net *net) { return 0; } +static void nfs_minorversion_callback_svc_setup(struct svc_serv *serv, + struct svc_rqst **rqstpp, int (**callback_svc)(void *vrqstp)) +{ + *rqstpp = ERR_PTR(-ENOTSUPP); + *callback_svc = ERR_PTR(-ENOTSUPP); +} + static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt, - struct nfs_callback_data *cb_info) + struct svc_serv *serv) { } #endif /* CONFIG_NFS_V4_1 */ -/* - * Bring up the callback thread if it is not already up. - */ -int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt) +static int nfs_callback_start_svc(int minorversion, struct rpc_xprt *xprt, + struct svc_serv *serv) { - struct svc_serv *serv = NULL; struct svc_rqst *rqstp; int (*callback_svc)(void *vrqstp); struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion]; char svc_name[12]; - int ret = 0; - int minorversion_setup; - struct net *net = &init_net; + int ret; - mutex_lock(&nfs_callback_mutex); - if (cb_info->users++ || cb_info->task != NULL) { - nfs_callback_bc_serv(minorversion, xprt, cb_info); - goto out; - } - serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, NULL); - if (!serv) { - ret = -ENOMEM; - goto out_err; - } - /* As there is only one thread we need to over-ride the - * default maximum of 80 connections - */ - serv->sv_maxconn = 1024; + nfs_callback_bc_serv(minorversion, xprt, serv); - ret = svc_bind(serv, net); - if (ret < 0) { - printk(KERN_WARNING "NFS: bind callback service failed\n"); - goto out_err; - } + if (cb_info->task) + return 0; - minorversion_setup = nfs_minorversion_callback_svc_setup(minorversion, - serv, xprt, &rqstp, &callback_svc); - if (!minorversion_setup) { + switch (minorversion) { + case 0: /* v4.0 callback setup */ - rqstp = nfs4_callback_up(serv, xprt); + rqstp = nfs4_callback_up(serv); callback_svc = nfs4_callback_svc; + break; + default: + nfs_minorversion_callback_svc_setup(serv, + &rqstp, &callback_svc); } - if (IS_ERR(rqstp)) { - ret = PTR_ERR(rqstp); - goto out_err; - } + if (IS_ERR(rqstp)) + return PTR_ERR(rqstp); svc_sock_update_bufs(serv); @@ -276,41 +241,165 @@ int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt) svc_exit_thread(cb_info->rqst); cb_info->rqst = NULL; cb_info->task = NULL; - goto out_err; + return ret; + } + dprintk("nfs_callback_up: service started\n"); + return 0; +} + +static void nfs_callback_down_net(u32 minorversion, struct svc_serv *serv, struct net *net) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + + if (--nn->cb_users[minorversion]) + return; + + dprintk("NFS: destroy per-net callback data; net=%p\n", net); + svc_shutdown_net(serv, net); +} + +static int nfs_callback_up_net(int minorversion, struct svc_serv *serv, struct net *net) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + int ret; + + if (nn->cb_users[minorversion]++) + return 0; + + dprintk("NFS: create per-net callback data; net=%p\n", net); + + ret = svc_bind(serv, net); + if (ret < 0) { + printk(KERN_WARNING "NFS: bind callback service failed\n"); + goto err_bind; + } + + switch (minorversion) { + case 0: + ret = nfs4_callback_up_net(serv, net); + break; + case 1: + ret = nfs41_callback_up_net(serv, net); + break; + default: + printk(KERN_ERR "NFS: unknown callback version: %d\n", + minorversion); + ret = -EINVAL; + break; + } + + if (ret < 0) { + printk(KERN_ERR "NFS: callback service start failed\n"); + goto err_socks; + } + return 0; + +err_socks: + svc_rpcb_cleanup(serv, net); +err_bind: + dprintk("NFS: Couldn't create callback socket: err = %d; " + "net = %p\n", ret, net); + return ret; +} + +static struct svc_serv *nfs_callback_create_svc(int minorversion) +{ + struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion]; + struct svc_serv *serv; + + /* + * Check whether we're already up and running. + */ + if (cb_info->task) { + /* + * Note: increase service usage, because later in case of error + * svc_destroy() will be called. + */ + svc_get(cb_info->serv); + return cb_info->serv; + } + + /* + * Sanity check: if there's no task, + * we should be the first user ... + */ + if (cb_info->users) + printk(KERN_WARNING "nfs_callback_create_svc: no kthread, %d users??\n", + cb_info->users); + + serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, NULL); + if (!serv) { + printk(KERN_ERR "nfs_callback_create_svc: create service failed\n"); + return ERR_PTR(-ENOMEM); + } + /* As there is only one thread we need to over-ride the + * default maximum of 80 connections + */ + serv->sv_maxconn = 1024; + dprintk("nfs_callback_create_svc: service created\n"); + return serv; +} + +/* + * Bring up the callback thread if it is not already up. + */ +int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt) +{ + struct svc_serv *serv; + struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion]; + int ret; + struct net *net = xprt->xprt_net; + + mutex_lock(&nfs_callback_mutex); + + serv = nfs_callback_create_svc(minorversion); + if (IS_ERR(serv)) { + ret = PTR_ERR(serv); + goto err_create; } -out: + + ret = nfs_callback_up_net(minorversion, serv, net); + if (ret < 0) + goto err_net; + + ret = nfs_callback_start_svc(minorversion, xprt, serv); + if (ret < 0) + goto err_start; + + cb_info->users++; /* * svc_create creates the svc_serv with sv_nrthreads == 1, and then * svc_prepare_thread increments that. So we need to call svc_destroy * on both success and failure so that the refcount is 1 when the * thread exits. */ - if (serv) - svc_destroy(serv); +err_net: + svc_destroy(serv); +err_create: mutex_unlock(&nfs_callback_mutex); return ret; -out_err: - dprintk("NFS: Couldn't create callback socket or server thread; " - "err = %d\n", ret); - cb_info->users--; - if (serv) - svc_shutdown_net(serv, net); - goto out; + +err_start: + nfs_callback_down_net(minorversion, serv, net); + dprintk("NFS: Couldn't create server thread; err = %d\n", ret); + goto err_net; } /* * Kill the callback thread if it's no longer being used. */ -void nfs_callback_down(int minorversion) +void nfs_callback_down(int minorversion, struct net *net) { struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion]; mutex_lock(&nfs_callback_mutex); + nfs_callback_down_net(minorversion, cb_info->serv, net); cb_info->users--; if (cb_info->users == 0 && cb_info->task != NULL) { kthread_stop(cb_info->task); - svc_shutdown_net(cb_info->serv, &init_net); + dprintk("nfs_callback_down: service stopped\n"); svc_exit_thread(cb_info->rqst); + dprintk("nfs_callback_down: service destroyed\n"); cb_info->serv = NULL; cb_info->rqst = NULL; cb_info->task = NULL; diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h index b44d7b128b71..4251c2ae06ad 100644 --- a/fs/nfs/callback.h +++ b/fs/nfs/callback.h @@ -194,7 +194,7 @@ extern __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy, struct cb_process_state *cps); #if IS_ENABLED(CONFIG_NFS_V4) extern int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt); -extern void nfs_callback_down(int minorversion); +extern void nfs_callback_down(int minorversion, struct net *net); extern int nfs4_validate_delegation_stateid(struct nfs_delegation *delegation, const nfs4_stateid *stateid); extern int nfs4_set_callback_sessionid(struct nfs_client *clp); @@ -209,6 +209,5 @@ extern int nfs4_set_callback_sessionid(struct nfs_client *clp); extern unsigned int nfs_callback_set_tcpport; extern unsigned short nfs_callback_tcpport; -extern unsigned short nfs_callback_tcpport6; #endif /* __LINUX_FS_NFS_CALLBACK_H */ diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index 1b5d809a105e..76b4a7a3e559 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -122,7 +122,15 @@ static struct pnfs_layout_hdr * get_layout_by_fh_locked(struct nfs_client *clp, ino = igrab(lo->plh_inode); if (!ino) continue; - get_layout_hdr(lo); + spin_lock(&ino->i_lock); + /* Is this layout in the process of being freed? */ + if (NFS_I(ino)->layout != lo) { + spin_unlock(&ino->i_lock); + iput(ino); + continue; + } + pnfs_get_layout_hdr(lo); + spin_unlock(&ino->i_lock); return lo; } } @@ -158,7 +166,7 @@ static u32 initiate_file_draining(struct nfs_client *clp, ino = lo->plh_inode; spin_lock(&ino->i_lock); if (test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags) || - mark_matching_lsegs_invalid(lo, &free_me_list, + pnfs_mark_matching_lsegs_invalid(lo, &free_me_list, &args->cbl_range)) rv = NFS4ERR_DELAY; else @@ -166,7 +174,7 @@ static u32 initiate_file_draining(struct nfs_client *clp, pnfs_set_layout_stateid(lo, &args->cbl_stateid, true); spin_unlock(&ino->i_lock); pnfs_free_lseg_list(&free_me_list); - put_layout_hdr(lo); + pnfs_put_layout_hdr(lo); iput(ino); return rv; } @@ -196,9 +204,18 @@ static u32 initiate_bulk_draining(struct nfs_client *clp, continue; list_for_each_entry(lo, &server->layouts, plh_layouts) { - if (!igrab(lo->plh_inode)) + ino = igrab(lo->plh_inode); + if (ino) + continue; + spin_lock(&ino->i_lock); + /* Is this layout in the process of being freed? */ + if (NFS_I(ino)->layout != lo) { + spin_unlock(&ino->i_lock); + iput(ino); continue; - get_layout_hdr(lo); + } + pnfs_get_layout_hdr(lo); + spin_unlock(&ino->i_lock); BUG_ON(!list_empty(&lo->plh_bulk_recall)); list_add(&lo->plh_bulk_recall, &recall_list); } @@ -211,12 +228,12 @@ static u32 initiate_bulk_draining(struct nfs_client *clp, ino = lo->plh_inode; spin_lock(&ino->i_lock); set_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags); - if (mark_matching_lsegs_invalid(lo, &free_me_list, &range)) + if (pnfs_mark_matching_lsegs_invalid(lo, &free_me_list, &range)) rv = NFS4ERR_DELAY; list_del_init(&lo->plh_bulk_recall); spin_unlock(&ino->i_lock); pnfs_free_lseg_list(&free_me_list); - put_layout_hdr(lo); + pnfs_put_layout_hdr(lo); iput(ino); } return rv; diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 99694442b93f..8b39a42ac35e 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -93,10 +93,10 @@ static struct nfs_subversion *find_nfs_version(unsigned int version) spin_unlock(&nfs_version_lock); return nfs; } - }; + } spin_unlock(&nfs_version_lock); - return ERR_PTR(-EPROTONOSUPPORT);; + return ERR_PTR(-EPROTONOSUPPORT); } struct nfs_subversion *get_nfs_version(unsigned int version) @@ -498,7 +498,8 @@ nfs_get_client(const struct nfs_client_initdata *cl_init, return nfs_found_client(cl_init, clp); } if (new) { - list_add(&new->cl_share_link, &nn->nfs_client_list); + list_add_tail(&new->cl_share_link, + &nn->nfs_client_list); spin_unlock(&nn->nfs_client_lock); new->cl_flags = cl_init->init_flags; return rpc_ops->init_client(new, timeparms, ip_addr, @@ -668,7 +669,8 @@ int nfs_init_server_rpcclient(struct nfs_server *server, { struct nfs_client *clp = server->nfs_client; - server->client = rpc_clone_client(clp->cl_rpcclient); + server->client = rpc_clone_client_set_auth(clp->cl_rpcclient, + pseudoflavour); if (IS_ERR(server->client)) { dprintk("%s: couldn't create rpc_client!\n", __func__); return PTR_ERR(server->client); @@ -678,16 +680,6 @@ int nfs_init_server_rpcclient(struct nfs_server *server, timeo, sizeof(server->client->cl_timeout_default)); server->client->cl_timeout = &server->client->cl_timeout_default; - - if (pseudoflavour != clp->cl_rpcclient->cl_auth->au_flavor) { - struct rpc_auth *auth; - - auth = rpcauth_create(pseudoflavour, server->client); - if (IS_ERR(auth)) { - dprintk("%s: couldn't create credcache!\n", __func__); - return PTR_ERR(auth); - } - } server->client->cl_softrtry = 0; if (server->flags & NFS_MOUNT_SOFT) server->client->cl_softrtry = 1; @@ -761,6 +753,8 @@ static int nfs_init_server(struct nfs_server *server, data->timeo, data->retrans); if (data->flags & NFS_MOUNT_NORESVPORT) set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags); + if (server->options & NFS_OPTION_MIGRATION) + set_bit(NFS_CS_MIGRATION, &cl_init.init_flags); /* Allocate or find a client reference we can use */ clp = nfs_get_client(&cl_init, &timeparms, NULL, RPC_AUTH_UNIX); @@ -855,7 +849,6 @@ static void nfs_server_set_fsinfo(struct nfs_server *server, if (server->wsize > NFS_MAX_FILE_IO_SIZE) server->wsize = NFS_MAX_FILE_IO_SIZE; server->wpages = (server->wsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - server->pnfs_blksize = fsinfo->blksize; server->wtmult = nfs_block_bits(fsinfo->wtmult, NULL); diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 627f108ede23..ce8cb926526b 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -2072,7 +2072,7 @@ found: nfs_access_free_entry(entry); } -static void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set) +void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set) { struct nfs_access_entry *cache = kmalloc(sizeof(*cache), GFP_KERNEL); if (cache == NULL) @@ -2098,6 +2098,20 @@ static void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *s spin_unlock(&nfs_access_lru_lock); } } +EXPORT_SYMBOL_GPL(nfs_access_add_cache); + +void nfs_access_set_mask(struct nfs_access_entry *entry, u32 access_result) +{ + entry->mask = 0; + if (access_result & NFS4_ACCESS_READ) + entry->mask |= MAY_READ; + if (access_result & + (NFS4_ACCESS_MODIFY | NFS4_ACCESS_EXTEND | NFS4_ACCESS_DELETE)) + entry->mask |= MAY_WRITE; + if (access_result & (NFS4_ACCESS_LOOKUP|NFS4_ACCESS_EXECUTE)) + entry->mask |= MAY_EXEC; +} +EXPORT_SYMBOL_GPL(nfs_access_set_mask); static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask) { diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 1ba385b7c90d..cae26cbd59ee 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -46,6 +46,7 @@ #include <linux/kref.h> #include <linux/slab.h> #include <linux/task_io_accounting_ops.h> +#include <linux/module.h> #include <linux/nfs_fs.h> #include <linux/nfs_page.h> @@ -78,6 +79,7 @@ struct nfs_direct_req { atomic_t io_count; /* i/os we're waiting for */ spinlock_t lock; /* protect completion state */ ssize_t count, /* bytes actually processed */ + bytes_left, /* bytes left to be sent */ error; /* any reported error */ struct completion completion; /* wait for i/o completion */ @@ -190,6 +192,12 @@ static void nfs_direct_req_release(struct nfs_direct_req *dreq) kref_put(&dreq->kref, nfs_direct_req_free); } +ssize_t nfs_dreq_bytes_left(struct nfs_direct_req *dreq) +{ + return dreq->bytes_left; +} +EXPORT_SYMBOL_GPL(nfs_dreq_bytes_left); + /* * Collects and returns the final error value/byte-count. */ @@ -390,6 +398,7 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_pageio_descriptor *de user_addr += req_len; pos += req_len; count -= req_len; + dreq->bytes_left -= req_len; } /* The nfs_page now hold references to these pages */ nfs_direct_release_pages(pagevec, npages); @@ -450,23 +459,28 @@ static ssize_t nfs_direct_read(struct kiocb *iocb, const struct iovec *iov, ssize_t result = -ENOMEM; struct inode *inode = iocb->ki_filp->f_mapping->host; struct nfs_direct_req *dreq; + struct nfs_lock_context *l_ctx; dreq = nfs_direct_req_alloc(); if (dreq == NULL) goto out; dreq->inode = inode; + dreq->bytes_left = iov_length(iov, nr_segs); dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp)); - dreq->l_ctx = nfs_get_lock_context(dreq->ctx); - if (dreq->l_ctx == NULL) + l_ctx = nfs_get_lock_context(dreq->ctx); + if (IS_ERR(l_ctx)) { + result = PTR_ERR(l_ctx); goto out_release; + } + dreq->l_ctx = l_ctx; if (!is_sync_kiocb(iocb)) dreq->iocb = iocb; + NFS_I(inode)->read_io += iov_length(iov, nr_segs); result = nfs_direct_read_schedule_iovec(dreq, iov, nr_segs, pos, uio); if (!result) result = nfs_direct_wait(dreq); - NFS_I(inode)->read_io += result; out_release: nfs_direct_req_release(dreq); out: @@ -706,6 +720,7 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_pageio_descriptor *d user_addr += req_len; pos += req_len; count -= req_len; + dreq->bytes_left -= req_len; } /* The nfs_page now hold references to these pages */ nfs_direct_release_pages(pagevec, npages); @@ -814,6 +829,7 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq, get_dreq(dreq); atomic_inc(&inode->i_dio_count); + NFS_I(dreq->inode)->write_io += iov_length(iov, nr_segs); for (seg = 0; seg < nr_segs; seg++) { const struct iovec *vec = &iov[seg]; result = nfs_direct_write_schedule_segment(&desc, vec, pos, uio); @@ -825,7 +841,6 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq, pos += vec->iov_len; } nfs_pageio_complete(&desc); - NFS_I(dreq->inode)->write_io += desc.pg_bytes_written; /* * If no bytes were started, return the error, and let the @@ -849,16 +864,21 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, const struct iovec *iov, ssize_t result = -ENOMEM; struct inode *inode = iocb->ki_filp->f_mapping->host; struct nfs_direct_req *dreq; + struct nfs_lock_context *l_ctx; dreq = nfs_direct_req_alloc(); if (!dreq) goto out; dreq->inode = inode; + dreq->bytes_left = count; dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp)); - dreq->l_ctx = nfs_get_lock_context(dreq->ctx); - if (dreq->l_ctx == NULL) + l_ctx = nfs_get_lock_context(dreq->ctx); + if (IS_ERR(l_ctx)) { + result = PTR_ERR(l_ctx); goto out_release; + } + dreq->l_ctx = l_ctx; if (!is_sync_kiocb(iocb)) dreq->iocb = iocb; diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 6a7fcab7ecb3..582bb8866131 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -259,7 +259,7 @@ nfs_file_fsync_commit(struct file *file, loff_t start, loff_t end, int datasync) struct dentry *dentry = file->f_path.dentry; struct nfs_open_context *ctx = nfs_file_open_context(file); struct inode *inode = dentry->d_inode; - int have_error, status; + int have_error, do_resend, status; int ret = 0; dprintk("NFS: fsync file(%s/%s) datasync %d\n", @@ -267,15 +267,23 @@ nfs_file_fsync_commit(struct file *file, loff_t start, loff_t end, int datasync) datasync); nfs_inc_stats(inode, NFSIOS_VFSFSYNC); + do_resend = test_and_clear_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags); have_error = test_and_clear_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags); status = nfs_commit_inode(inode, FLUSH_SYNC); - if (status >= 0 && ret < 0) - status = ret; have_error |= test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags); - if (have_error) + if (have_error) { ret = xchg(&ctx->error, 0); - if (!ret && status < 0) + if (ret) + goto out; + } + if (status < 0) { ret = status; + goto out; + } + do_resend |= test_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags); + if (do_resend) + ret = -EAGAIN; +out: return ret; } EXPORT_SYMBOL_GPL(nfs_file_fsync_commit); @@ -286,13 +294,22 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync) int ret; struct inode *inode = file->f_path.dentry->d_inode; - ret = filemap_write_and_wait_range(inode->i_mapping, start, end); - if (ret != 0) - goto out; - mutex_lock(&inode->i_mutex); - ret = nfs_file_fsync_commit(file, start, end, datasync); - mutex_unlock(&inode->i_mutex); -out: + do { + ret = filemap_write_and_wait_range(inode->i_mapping, start, end); + if (ret != 0) + break; + mutex_lock(&inode->i_mutex); + ret = nfs_file_fsync_commit(file, start, end, datasync); + mutex_unlock(&inode->i_mutex); + /* + * If nfs_file_fsync_commit detected a server reboot, then + * resend all dirty pages that might have been covered by + * the NFS_CONTEXT_RESEND_WRITES flag + */ + start = 0; + end = LLONG_MAX; + } while (ret == -EAGAIN); + return ret; } @@ -578,6 +595,7 @@ out: static const struct vm_operations_struct nfs_file_vm_ops = { .fault = filemap_fault, .page_mkwrite = nfs_vm_page_mkwrite, + .remap_pages = generic_file_remap_pages, }; static int nfs_need_sync_write(struct file *filp, struct inode *inode) diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c index 4654ced096a6..033803c36644 100644 --- a/fs/nfs/getroot.c +++ b/fs/nfs/getroot.c @@ -32,6 +32,8 @@ #include <asm/uaccess.h> +#include "internal.h" + #define NFSDBG_FACILITY NFSDBG_CLIENT /* diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c index a850079467d8..9cc4a3fbf4b0 100644 --- a/fs/nfs/idmap.c +++ b/fs/nfs/idmap.c @@ -55,18 +55,19 @@ static const struct cred *id_resolver_cache; static struct key_type key_type_id_resolver_legacy; -struct idmap { - struct rpc_pipe *idmap_pipe; - struct key_construction *idmap_key_cons; - struct mutex idmap_mutex; -}; - struct idmap_legacy_upcalldata { struct rpc_pipe_msg pipe_msg; struct idmap_msg idmap_msg; + struct key_construction *key_cons; struct idmap *idmap; }; +struct idmap { + struct rpc_pipe *idmap_pipe; + struct idmap_legacy_upcalldata *idmap_upcall_data; + struct mutex idmap_mutex; +}; + /** * nfs_fattr_init_names - initialise the nfs_fattr owner_name/group_name fields * @fattr: fully initialised struct nfs_fattr @@ -158,7 +159,7 @@ static int nfs_map_string_to_numeric(const char *name, size_t namelen, __u32 *re return 0; memcpy(buf, name, namelen); buf[namelen] = '\0'; - if (strict_strtoul(buf, 0, &val) != 0) + if (kstrtoul(buf, 0, &val) != 0) return 0; *res = val; return 1; @@ -330,7 +331,6 @@ static ssize_t nfs_idmap_get_key(const char *name, size_t namelen, ret = nfs_idmap_request_key(&key_type_id_resolver_legacy, name, namelen, type, data, data_size, idmap); - idmap->idmap_key_cons = NULL; mutex_unlock(&idmap->idmap_mutex); } return ret; @@ -364,7 +364,7 @@ static int nfs_idmap_lookup_id(const char *name, size_t namelen, const char *typ if (data_size <= 0) { ret = -EINVAL; } else { - ret = strict_strtol(id_str, 10, &id_long); + ret = kstrtol(id_str, 10, &id_long); *id = (__u32)id_long; } return ret; @@ -465,8 +465,6 @@ nfs_idmap_new(struct nfs_client *clp) struct rpc_pipe *pipe; int error; - BUG_ON(clp->cl_idmap != NULL); - idmap = kzalloc(sizeof(*idmap), GFP_KERNEL); if (idmap == NULL) return -ENOMEM; @@ -510,7 +508,6 @@ static int __rpc_pipefs_event(struct nfs_client *clp, unsigned long event, switch (event) { case RPC_PIPEFS_MOUNT: - BUG_ON(clp->cl_rpcclient->cl_dentry == NULL); err = __nfs_idmap_register(clp->cl_rpcclient->cl_dentry, clp->cl_idmap, clp->cl_idmap->idmap_pipe); @@ -632,9 +629,6 @@ static int nfs_idmap_prepare_message(char *desc, struct idmap *idmap, substring_t substr; int token, ret; - memset(im, 0, sizeof(*im)); - memset(msg, 0, sizeof(*msg)); - im->im_type = IDMAP_TYPE_GROUP; token = match_token(desc, nfs_idmap_tokens, &substr); @@ -665,6 +659,35 @@ out: return ret; } +static bool +nfs_idmap_prepare_pipe_upcall(struct idmap *idmap, + struct idmap_legacy_upcalldata *data) +{ + if (idmap->idmap_upcall_data != NULL) { + WARN_ON_ONCE(1); + return false; + } + idmap->idmap_upcall_data = data; + return true; +} + +static void +nfs_idmap_complete_pipe_upcall_locked(struct idmap *idmap, int ret) +{ + struct key_construction *cons = idmap->idmap_upcall_data->key_cons; + + kfree(idmap->idmap_upcall_data); + idmap->idmap_upcall_data = NULL; + complete_request_key(cons, ret); +} + +static void +nfs_idmap_abort_pipe_upcall(struct idmap *idmap, int ret) +{ + if (idmap->idmap_upcall_data != NULL) + nfs_idmap_complete_pipe_upcall_locked(idmap, ret); +} + static int nfs_idmap_legacy_upcall(struct key_construction *cons, const char *op, void *aux) @@ -677,29 +700,28 @@ static int nfs_idmap_legacy_upcall(struct key_construction *cons, int ret = -ENOMEM; /* msg and im are freed in idmap_pipe_destroy_msg */ - data = kmalloc(sizeof(*data), GFP_KERNEL); + data = kzalloc(sizeof(*data), GFP_KERNEL); if (!data) goto out1; msg = &data->pipe_msg; im = &data->idmap_msg; data->idmap = idmap; + data->key_cons = cons; ret = nfs_idmap_prepare_message(key->description, idmap, im, msg); if (ret < 0) goto out2; - BUG_ON(idmap->idmap_key_cons != NULL); - idmap->idmap_key_cons = cons; + ret = -EAGAIN; + if (!nfs_idmap_prepare_pipe_upcall(idmap, data)) + goto out2; ret = rpc_queue_upcall(idmap->idmap_pipe, msg); if (ret < 0) - goto out3; + nfs_idmap_abort_pipe_upcall(idmap, ret); return ret; - -out3: - idmap->idmap_key_cons = NULL; out2: kfree(data); out1: @@ -714,21 +736,32 @@ static int nfs_idmap_instantiate(struct key *key, struct key *authkey, char *dat authkey); } -static int nfs_idmap_read_message(struct idmap_msg *im, struct key *key, struct key *authkey) +static int nfs_idmap_read_and_verify_message(struct idmap_msg *im, + struct idmap_msg *upcall, + struct key *key, struct key *authkey) { char id_str[NFS_UINT_MAXLEN]; - int ret = -EINVAL; + int ret = -ENOKEY; + /* ret = -ENOKEY */ + if (upcall->im_type != im->im_type || upcall->im_conv != im->im_conv) + goto out; switch (im->im_conv) { case IDMAP_CONV_NAMETOID: + if (strcmp(upcall->im_name, im->im_name) != 0) + break; sprintf(id_str, "%d", im->im_id); ret = nfs_idmap_instantiate(key, authkey, id_str); break; case IDMAP_CONV_IDTONAME: + if (upcall->im_id != im->im_id) + break; ret = nfs_idmap_instantiate(key, authkey, im->im_name); break; + default: + ret = -EINVAL; } - +out: return ret; } @@ -740,14 +773,16 @@ idmap_pipe_downcall(struct file *filp, const char __user *src, size_t mlen) struct key_construction *cons; struct idmap_msg im; size_t namelen_in; - int ret; + int ret = -ENOKEY; /* If instantiation is successful, anyone waiting for key construction * will have been woken up and someone else may now have used * idmap_key_cons - so after this point we may no longer touch it. */ - cons = ACCESS_ONCE(idmap->idmap_key_cons); - idmap->idmap_key_cons = NULL; + if (idmap->idmap_upcall_data == NULL) + goto out_noupcall; + + cons = idmap->idmap_upcall_data->key_cons; if (mlen != sizeof(im)) { ret = -ENOSPC; @@ -768,16 +803,19 @@ idmap_pipe_downcall(struct file *filp, const char __user *src, size_t mlen) if (namelen_in == 0 || namelen_in == IDMAP_NAMESZ) { ret = -EINVAL; goto out; - } +} - ret = nfs_idmap_read_message(&im, cons->key, cons->authkey); + ret = nfs_idmap_read_and_verify_message(&im, + &idmap->idmap_upcall_data->idmap_msg, + cons->key, cons->authkey); if (ret >= 0) { key_set_timeout(cons->key, nfs_idmap_cache_timeout); ret = mlen; } out: - complete_request_key(cons, ret); + nfs_idmap_complete_pipe_upcall_locked(idmap, ret); +out_noupcall: return ret; } @@ -788,14 +826,9 @@ idmap_pipe_destroy_msg(struct rpc_pipe_msg *msg) struct idmap_legacy_upcalldata, pipe_msg); struct idmap *idmap = data->idmap; - struct key_construction *cons; - if (msg->errno) { - cons = ACCESS_ONCE(idmap->idmap_key_cons); - idmap->idmap_key_cons = NULL; - complete_request_key(cons, msg->errno); - } - /* Free memory allocated in nfs_idmap_legacy_upcall() */ - kfree(data); + + if (msg->errno) + nfs_idmap_abort_pipe_upcall(idmap, msg->errno); } static void @@ -803,7 +836,8 @@ idmap_release_pipe(struct inode *inode) { struct rpc_inode *rpci = RPC_I(inode); struct idmap *idmap = (struct idmap *)rpci->private; - idmap->idmap_key_cons = NULL; + + nfs_idmap_abort_pipe_upcall(idmap, -EPIPE); } int nfs_map_name_to_uid(const struct nfs_server *server, const char *name, size_t namelen, __u32 *uid) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index e4c716d374a8..5c7325c5c5e6 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -547,8 +547,8 @@ EXPORT_SYMBOL_GPL(nfs_getattr); static void nfs_init_lock_context(struct nfs_lock_context *l_ctx) { atomic_set(&l_ctx->count, 1); - l_ctx->lockowner = current->files; - l_ctx->pid = current->tgid; + l_ctx->lockowner.l_owner = current->files; + l_ctx->lockowner.l_pid = current->tgid; INIT_LIST_HEAD(&l_ctx->list); } @@ -557,9 +557,9 @@ static struct nfs_lock_context *__nfs_find_lock_context(struct nfs_open_context struct nfs_lock_context *pos; list_for_each_entry(pos, &ctx->lock_context.list, list) { - if (pos->lockowner != current->files) + if (pos->lockowner.l_owner != current->files) continue; - if (pos->pid != current->tgid) + if (pos->lockowner.l_pid != current->tgid) continue; atomic_inc(&pos->count); return pos; @@ -578,7 +578,7 @@ struct nfs_lock_context *nfs_get_lock_context(struct nfs_open_context *ctx) spin_unlock(&inode->i_lock); new = kmalloc(sizeof(*new), GFP_KERNEL); if (new == NULL) - return NULL; + return ERR_PTR(-ENOMEM); nfs_init_lock_context(new); spin_lock(&inode->i_lock); res = __nfs_find_lock_context(ctx); diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 31fdb03225cd..59b133c5d652 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -101,11 +101,11 @@ struct nfs_client_initdata { */ struct nfs_parsed_mount_data { int flags; - int rsize, wsize; - int timeo, retrans; - int acregmin, acregmax, + unsigned int rsize, wsize; + unsigned int timeo, retrans; + unsigned int acregmin, acregmax, acdirmin, acdirmax; - int namlen; + unsigned int namlen; unsigned int options; unsigned int bsize; unsigned int auth_flavor_len; @@ -464,6 +464,7 @@ static inline void nfs_inode_dio_wait(struct inode *inode) { inode_dio_wait(inode); } +extern ssize_t nfs_dreq_bytes_left(struct nfs_direct_req *dreq); /* nfs4proc.c */ extern void __nfs4_read_done_cb(struct nfs_read_data *); @@ -483,6 +484,12 @@ extern int _nfs4_call_sync_session(struct rpc_clnt *clnt, struct nfs4_sequence_args *args, struct nfs4_sequence_res *res, int cache_reply); +extern int nfs40_walk_client_list(struct nfs_client *clp, + struct nfs_client **result, + struct rpc_cred *cred); +extern int nfs41_walk_client_list(struct nfs_client *clp, + struct nfs_client **result, + struct rpc_cred *cred); /* * Determine the device name as a string diff --git a/fs/nfs/netns.h b/fs/nfs/netns.h index 0539de1b8d1f..8ee1fab83268 100644 --- a/fs/nfs/netns.h +++ b/fs/nfs/netns.h @@ -5,6 +5,7 @@ #ifndef __NFS_NETNS_H__ #define __NFS_NETNS_H__ +#include <linux/nfs4.h> #include <net/net_namespace.h> #include <net/netns/generic.h> @@ -22,6 +23,9 @@ struct nfs_net { struct list_head nfs_volume_list; #if IS_ENABLED(CONFIG_NFS_V4) struct idr cb_ident_idr; /* Protected by nfs_client_lock */ + unsigned short nfs_callback_tcpport; + unsigned short nfs_callback_tcpport6; + int cb_users[NFS4_MAX_MINOR_VERSION + 1]; #endif spinlock_t nfs_client_lock; struct timespec boot_time; diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index da0618aeeadb..a525fdefccde 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -132,8 +132,8 @@ struct nfs4_lock_owner { struct nfs4_lock_state { struct list_head ls_locks; /* Other lock stateids */ struct nfs4_state * ls_state; /* Pointer to open state */ -#define NFS_LOCK_INITIALIZED 1 - int ls_flags; +#define NFS_LOCK_INITIALIZED 0 + unsigned long ls_flags; struct nfs_seqid_counter ls_seqid; nfs4_stateid ls_stateid; atomic_t ls_count; @@ -191,6 +191,8 @@ struct nfs4_state_recovery_ops { int (*establish_clid)(struct nfs_client *, struct rpc_cred *); struct rpc_cred * (*get_clid_cred)(struct nfs_client *); int (*reclaim_complete)(struct nfs_client *); + int (*detect_trunking)(struct nfs_client *, struct nfs_client **, + struct rpc_cred *); }; struct nfs4_state_maintenance_ops { @@ -223,7 +225,7 @@ extern int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred); extern int nfs4_destroy_clientid(struct nfs_client *clp); extern int nfs4_init_clientid(struct nfs_client *, struct rpc_cred *); extern int nfs41_init_clientid(struct nfs_client *, struct rpc_cred *); -extern int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait, bool roc); +extern int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait); extern int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle); extern int nfs4_proc_fs_locations(struct rpc_clnt *, struct inode *, const struct qstr *, struct nfs4_fs_locations *, struct page *); @@ -320,9 +322,15 @@ extern void nfs4_renew_state(struct work_struct *); /* nfs4state.c */ struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp); struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp); +int nfs4_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **); +int nfs40_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **, struct rpc_cred *); #if defined(CONFIG_NFS_V4_1) struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp); struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp); +int nfs41_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **, struct rpc_cred *); extern void nfs4_schedule_session_recovery(struct nfs4_session *, int); #else static inline void nfs4_schedule_session_recovery(struct nfs4_session *session, int err) @@ -351,7 +359,7 @@ extern void nfs41_handle_server_scope(struct nfs_client *, extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp); extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl); extern void nfs4_select_rw_stateid(nfs4_stateid *, struct nfs4_state *, - fmode_t, fl_owner_t, pid_t); + fmode_t, const struct nfs_lockowner *); extern struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter, gfp_t gfp_mask); extern int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task); @@ -372,6 +380,9 @@ extern bool nfs4_disable_idmapping; extern unsigned short max_session_slots; extern unsigned short send_implementation_id; +#define NFS4_CLIENT_ID_UNIQ_LEN (64) +extern char nfs4_client_id_uniquifier[NFS4_CLIENT_ID_UNIQ_LEN]; + /* nfs4sysctl.c */ #ifdef CONFIG_SYSCTL int nfs4_register_sysctl(void); diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index 24eb663f8ed5..6bacfde1319a 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -84,7 +84,7 @@ error: static void nfs4_destroy_callback(struct nfs_client *clp) { if (__test_and_clear_bit(NFS_CS_CALLBACK, &clp->cl_res_state)) - nfs_callback_down(clp->cl_mvops->minor_version); + nfs_callback_down(clp->cl_mvops->minor_version, clp->cl_net); } static void nfs4_shutdown_client(struct nfs_client *clp) @@ -185,6 +185,7 @@ struct nfs_client *nfs4_init_client(struct nfs_client *clp, rpc_authflavor_t authflavour) { char buf[INET6_ADDRSTRLEN + 1]; + struct nfs_client *old; int error; if (clp->cl_cons_state == NFS_CS_READY) { @@ -230,6 +231,17 @@ struct nfs_client *nfs4_init_client(struct nfs_client *clp, if (!nfs4_has_session(clp)) nfs_mark_client_ready(clp, NFS_CS_READY); + + error = nfs4_discover_server_trunking(clp, &old); + if (error < 0) + goto error; + if (clp != old) { + clp->cl_preserve_clid = true; + nfs_put_client(clp); + clp = old; + atomic_inc(&clp->cl_count); + } + return clp; error: @@ -239,6 +251,248 @@ error: return ERR_PTR(error); } +/* + * SETCLIENTID just did a callback update with the callback ident in + * "drop," but server trunking discovery claims "drop" and "keep" are + * actually the same server. Swap the callback IDs so that "keep" + * will continue to use the callback ident the server now knows about, + * and so that "keep"'s original callback ident is destroyed when + * "drop" is freed. + */ +static void nfs4_swap_callback_idents(struct nfs_client *keep, + struct nfs_client *drop) +{ + struct nfs_net *nn = net_generic(keep->cl_net, nfs_net_id); + unsigned int save = keep->cl_cb_ident; + + if (keep->cl_cb_ident == drop->cl_cb_ident) + return; + + dprintk("%s: keeping callback ident %u and dropping ident %u\n", + __func__, keep->cl_cb_ident, drop->cl_cb_ident); + + spin_lock(&nn->nfs_client_lock); + + idr_replace(&nn->cb_ident_idr, keep, drop->cl_cb_ident); + keep->cl_cb_ident = drop->cl_cb_ident; + + idr_replace(&nn->cb_ident_idr, drop, save); + drop->cl_cb_ident = save; + + spin_unlock(&nn->nfs_client_lock); +} + +/** + * nfs40_walk_client_list - Find server that recognizes a client ID + * + * @new: nfs_client with client ID to test + * @result: OUT: found nfs_client, or new + * @cred: credential to use for trunking test + * + * Returns zero, a negative errno, or a negative NFS4ERR status. + * If zero is returned, an nfs_client pointer is planted in "result." + * + * NB: nfs40_walk_client_list() relies on the new nfs_client being + * the last nfs_client on the list. + */ +int nfs40_walk_client_list(struct nfs_client *new, + struct nfs_client **result, + struct rpc_cred *cred) +{ + struct nfs_net *nn = net_generic(new->cl_net, nfs_net_id); + struct nfs_client *pos, *n, *prev = NULL; + struct nfs4_setclientid_res clid = { + .clientid = new->cl_clientid, + .confirm = new->cl_confirm, + }; + int status; + + spin_lock(&nn->nfs_client_lock); + list_for_each_entry_safe(pos, n, &nn->nfs_client_list, cl_share_link) { + /* If "pos" isn't marked ready, we can't trust the + * remaining fields in "pos" */ + if (pos->cl_cons_state < NFS_CS_READY) + continue; + + if (pos->rpc_ops != new->rpc_ops) + continue; + + if (pos->cl_proto != new->cl_proto) + continue; + + if (pos->cl_minorversion != new->cl_minorversion) + continue; + + if (pos->cl_clientid != new->cl_clientid) + continue; + + atomic_inc(&pos->cl_count); + spin_unlock(&nn->nfs_client_lock); + + if (prev) + nfs_put_client(prev); + + status = nfs4_proc_setclientid_confirm(pos, &clid, cred); + if (status == 0) { + nfs4_swap_callback_idents(pos, new); + + nfs_put_client(pos); + *result = pos; + dprintk("NFS: <-- %s using nfs_client = %p ({%d})\n", + __func__, pos, atomic_read(&pos->cl_count)); + return 0; + } + if (status != -NFS4ERR_STALE_CLIENTID) { + nfs_put_client(pos); + dprintk("NFS: <-- %s status = %d, no result\n", + __func__, status); + return status; + } + + spin_lock(&nn->nfs_client_lock); + prev = pos; + } + + /* + * No matching nfs_client found. This should be impossible, + * because the new nfs_client has already been added to + * nfs_client_list by nfs_get_client(). + * + * Don't BUG(), since the caller is holding a mutex. + */ + if (prev) + nfs_put_client(prev); + spin_unlock(&nn->nfs_client_lock); + pr_err("NFS: %s Error: no matching nfs_client found\n", __func__); + return -NFS4ERR_STALE_CLIENTID; +} + +#ifdef CONFIG_NFS_V4_1 +/* + * Returns true if the client IDs match + */ +static bool nfs4_match_clientids(struct nfs_client *a, struct nfs_client *b) +{ + if (a->cl_clientid != b->cl_clientid) { + dprintk("NFS: --> %s client ID %llx does not match %llx\n", + __func__, a->cl_clientid, b->cl_clientid); + return false; + } + dprintk("NFS: --> %s client ID %llx matches %llx\n", + __func__, a->cl_clientid, b->cl_clientid); + return true; +} + +/* + * Returns true if the server owners match + */ +static bool +nfs4_match_serverowners(struct nfs_client *a, struct nfs_client *b) +{ + struct nfs41_server_owner *o1 = a->cl_serverowner; + struct nfs41_server_owner *o2 = b->cl_serverowner; + + if (o1->minor_id != o2->minor_id) { + dprintk("NFS: --> %s server owner minor IDs do not match\n", + __func__); + return false; + } + + if (o1->major_id_sz != o2->major_id_sz) + goto out_major_mismatch; + if (memcmp(o1->major_id, o2->major_id, o1->major_id_sz) != 0) + goto out_major_mismatch; + + dprintk("NFS: --> %s server owners match\n", __func__); + return true; + +out_major_mismatch: + dprintk("NFS: --> %s server owner major IDs do not match\n", + __func__); + return false; +} + +/** + * nfs41_walk_client_list - Find nfs_client that matches a client/server owner + * + * @new: nfs_client with client ID to test + * @result: OUT: found nfs_client, or new + * @cred: credential to use for trunking test + * + * Returns zero, a negative errno, or a negative NFS4ERR status. + * If zero is returned, an nfs_client pointer is planted in "result." + * + * NB: nfs41_walk_client_list() relies on the new nfs_client being + * the last nfs_client on the list. + */ +int nfs41_walk_client_list(struct nfs_client *new, + struct nfs_client **result, + struct rpc_cred *cred) +{ + struct nfs_net *nn = net_generic(new->cl_net, nfs_net_id); + struct nfs_client *pos, *n, *prev = NULL; + int error; + + spin_lock(&nn->nfs_client_lock); + list_for_each_entry_safe(pos, n, &nn->nfs_client_list, cl_share_link) { + /* If "pos" isn't marked ready, we can't trust the + * remaining fields in "pos", especially the client + * ID and serverowner fields. Wait for CREATE_SESSION + * to finish. */ + if (pos->cl_cons_state < NFS_CS_READY) { + atomic_inc(&pos->cl_count); + spin_unlock(&nn->nfs_client_lock); + + if (prev) + nfs_put_client(prev); + prev = pos; + + error = nfs_wait_client_init_complete(pos); + if (error < 0) { + nfs_put_client(pos); + spin_lock(&nn->nfs_client_lock); + continue; + } + + spin_lock(&nn->nfs_client_lock); + } + + if (pos->rpc_ops != new->rpc_ops) + continue; + + if (pos->cl_proto != new->cl_proto) + continue; + + if (pos->cl_minorversion != new->cl_minorversion) + continue; + + if (!nfs4_match_clientids(pos, new)) + continue; + + if (!nfs4_match_serverowners(pos, new)) + continue; + + spin_unlock(&nn->nfs_client_lock); + dprintk("NFS: <-- %s using nfs_client = %p ({%d})\n", + __func__, pos, atomic_read(&pos->cl_count)); + + *result = pos; + return 0; + } + + /* + * No matching nfs_client found. This should be impossible, + * because the new nfs_client has already been added to + * nfs_client_list by nfs_get_client(). + * + * Don't BUG(), since the caller is holding a mutex. + */ + spin_unlock(&nn->nfs_client_lock); + pr_err("NFS: %s Error: no matching nfs_client found\n", __func__); + return -NFS4ERR_STALE_CLIENTID; +} +#endif /* CONFIG_NFS_V4_1 */ + static void nfs4_destroy_server(struct nfs_server *server) { nfs_server_return_all_delegations(server); diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c index eb5eb8eef4d3..afddd6639afb 100644 --- a/fs/nfs/nfs4file.c +++ b/fs/nfs/nfs4file.c @@ -95,16 +95,25 @@ nfs4_file_fsync(struct file *file, loff_t start, loff_t end, int datasync) int ret; struct inode *inode = file->f_path.dentry->d_inode; - ret = filemap_write_and_wait_range(inode->i_mapping, start, end); - if (ret != 0) - goto out; - mutex_lock(&inode->i_mutex); - ret = nfs_file_fsync_commit(file, start, end, datasync); - if (!ret && !datasync) - /* application has asked for meta-data sync */ - ret = pnfs_layoutcommit_inode(inode, true); - mutex_unlock(&inode->i_mutex); -out: + do { + ret = filemap_write_and_wait_range(inode->i_mapping, start, end); + if (ret != 0) + break; + mutex_lock(&inode->i_mutex); + ret = nfs_file_fsync_commit(file, start, end, datasync); + if (!ret && !datasync) + /* application has asked for meta-data sync */ + ret = pnfs_layoutcommit_inode(inode, true); + mutex_unlock(&inode->i_mutex); + /* + * If nfs_file_fsync_commit detected a server reboot, then + * resend all dirty pages that might have been covered by + * the NFS_CONTEXT_RESEND_WRITES flag + */ + start = 0; + end = LLONG_MAX; + } while (ret == -EAGAIN); + return ret; } diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c index 53f94d915bd1..2e45fd9c02a3 100644 --- a/fs/nfs/nfs4filelayout.c +++ b/fs/nfs/nfs4filelayout.c @@ -122,12 +122,21 @@ static void filelayout_reset_read(struct nfs_read_data *data) } } +static void filelayout_fenceme(struct inode *inode, struct pnfs_layout_hdr *lo) +{ + if (!test_and_clear_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) + return; + clear_bit(NFS_INO_LAYOUTCOMMIT, &NFS_I(inode)->flags); + pnfs_return_layout(inode); +} + static int filelayout_async_handle_error(struct rpc_task *task, struct nfs4_state *state, struct nfs_client *clp, struct pnfs_layout_segment *lseg) { - struct inode *inode = lseg->pls_layout->plh_inode; + struct pnfs_layout_hdr *lo = lseg->pls_layout; + struct inode *inode = lo->plh_inode; struct nfs_server *mds_server = NFS_SERVER(inode); struct nfs4_deviceid_node *devid = FILELAYOUT_DEVID_NODE(lseg); struct nfs_client *mds_client = mds_server->nfs_client; @@ -190,8 +199,6 @@ static int filelayout_async_handle_error(struct rpc_task *task, * i/o and all i/o waiting on the slot table to the MDS until * layout is destroyed and a new valid layout is obtained. */ - set_bit(NFS_LAYOUT_INVALID, - &NFS_I(inode)->layout->plh_flags); pnfs_destroy_layout(NFS_I(inode)); rpc_wake_up(&tbl->slot_tbl_waitq); goto reset; @@ -205,11 +212,9 @@ static int filelayout_async_handle_error(struct rpc_task *task, case -EPIPE: dprintk("%s DS connection error %d\n", __func__, task->tk_status); - filelayout_mark_devid_invalid(devid); - clear_bit(NFS_INO_LAYOUTCOMMIT, &NFS_I(inode)->flags); - _pnfs_return_layout(inode); + nfs4_mark_deviceid_unavailable(devid); + set_bit(NFS_LAYOUT_RETURN, &lo->plh_flags); rpc_wake_up(&tbl->slot_tbl_waitq); - nfs4_ds_disconnect(clp); /* fall through */ default: reset: @@ -269,6 +274,21 @@ filelayout_set_layoutcommit(struct nfs_write_data *wdata) (unsigned long) NFS_I(hdr->inode)->layout->plh_lwb); } +bool +filelayout_test_devid_unavailable(struct nfs4_deviceid_node *node) +{ + return filelayout_test_devid_invalid(node) || + nfs4_test_deviceid_unavailable(node); +} + +static bool +filelayout_reset_to_mds(struct pnfs_layout_segment *lseg) +{ + struct nfs4_deviceid_node *node = FILELAYOUT_DEVID_NODE(lseg); + + return filelayout_test_devid_unavailable(node); +} + /* * Call ops for the async read/write cases * In the case of dense layouts, the offset needs to be reset to its @@ -318,7 +338,9 @@ static void filelayout_read_count_stats(struct rpc_task *task, void *data) static void filelayout_read_release(void *data) { struct nfs_read_data *rdata = data; + struct pnfs_layout_hdr *lo = rdata->header->lseg->pls_layout; + filelayout_fenceme(lo->plh_inode, lo); nfs_put_client(rdata->ds_clp); rdata->header->mds_ops->rpc_release(data); } @@ -416,7 +438,9 @@ static void filelayout_write_count_stats(struct rpc_task *task, void *data) static void filelayout_write_release(void *data) { struct nfs_write_data *wdata = data; + struct pnfs_layout_hdr *lo = wdata->header->lseg->pls_layout; + filelayout_fenceme(lo->plh_inode, lo); nfs_put_client(wdata->ds_clp); wdata->header->mds_ops->rpc_release(data); } @@ -453,7 +477,7 @@ static void filelayout_commit_release(void *calldata) struct nfs_commit_data *data = calldata; data->completion_ops->completion(data); - put_lseg(data->lseg); + pnfs_put_lseg(data->lseg); nfs_put_client(data->ds_clp); nfs_commitdata_release(data); } @@ -608,13 +632,13 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo, d = nfs4_find_get_deviceid(NFS_SERVER(lo->plh_inode)->pnfs_curr_ld, NFS_SERVER(lo->plh_inode)->nfs_client, id); if (d == NULL) { - dsaddr = get_device_info(lo->plh_inode, id, gfp_flags); + dsaddr = filelayout_get_device_info(lo->plh_inode, id, gfp_flags); if (dsaddr == NULL) goto out; } else dsaddr = container_of(d, struct nfs4_file_layout_dsaddr, id_node); - /* Found deviceid is being reaped */ - if (test_bit(NFS_DEVICEID_INVALID, &dsaddr->id_node.flags)) + /* Found deviceid is unavailable */ + if (filelayout_test_devid_unavailable(&dsaddr->id_node)) goto out_put; fl->dsaddr = dsaddr; @@ -726,7 +750,7 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo, goto out_err; if (fl->num_fh > 0) { - fl->fh_array = kzalloc(fl->num_fh * sizeof(struct nfs_fh *), + fl->fh_array = kcalloc(fl->num_fh, sizeof(fl->fh_array[0]), gfp_flags); if (!fl->fh_array) goto out_err; @@ -931,7 +955,7 @@ filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio, nfs_init_cinfo(&cinfo, pgio->pg_inode, pgio->pg_dreq); status = filelayout_alloc_commit_info(pgio->pg_lseg, &cinfo, GFP_NOFS); if (status < 0) { - put_lseg(pgio->pg_lseg); + pnfs_put_lseg(pgio->pg_lseg); pgio->pg_lseg = NULL; goto out_mds; } @@ -985,7 +1009,7 @@ filelayout_clear_request_commit(struct nfs_page *req, out: nfs_request_remove_commit_list(req, cinfo); spin_unlock(cinfo->lock); - put_lseg(freeme); + pnfs_put_lseg(freeme); } static struct list_head * @@ -1018,7 +1042,7 @@ filelayout_choose_commit_list(struct nfs_page *req, * off due to a rewrite, in which case it will be done in * filelayout_clear_request_commit */ - buckets[i].wlseg = get_lseg(lseg); + buckets[i].wlseg = pnfs_get_lseg(lseg); } set_bit(PG_COMMIT_TO_DS, &req->wb_flags); cinfo->ds->nwritten++; @@ -1128,7 +1152,7 @@ filelayout_scan_ds_commit_list(struct pnfs_commit_bucket *bucket, if (list_empty(src)) bucket->wlseg = NULL; else - get_lseg(bucket->clseg); + pnfs_get_lseg(bucket->clseg); } return ret; } @@ -1159,12 +1183,12 @@ static void filelayout_recover_commit_reqs(struct list_head *dst, /* NOTE cinfo->lock is NOT held, relying on fact that this is * only called on single thread per dreq. - * Can't take the lock because need to do put_lseg + * Can't take the lock because need to do pnfs_put_lseg */ for (i = 0, b = cinfo->ds->buckets; i < cinfo->ds->nbuckets; i++, b++) { if (transfer_commit_list(&b->written, dst, cinfo, 0)) { BUG_ON(!list_empty(&b->written)); - put_lseg(b->wlseg); + pnfs_put_lseg(b->wlseg); b->wlseg = NULL; } } @@ -1200,7 +1224,7 @@ alloc_ds_commits(struct nfs_commit_info *cinfo, struct list_head *list) if (list_empty(&bucket->committing)) continue; nfs_retry_commit(&bucket->committing, bucket->clseg, cinfo); - put_lseg(bucket->clseg); + pnfs_put_lseg(bucket->clseg); bucket->clseg = NULL; } /* Caller will clean up entries put on list */ diff --git a/fs/nfs/nfs4filelayout.h b/fs/nfs/nfs4filelayout.h index 43fe802dd678..8c07241fe52b 100644 --- a/fs/nfs/nfs4filelayout.h +++ b/fs/nfs/nfs4filelayout.h @@ -129,23 +129,13 @@ filelayout_mark_devid_invalid(struct nfs4_deviceid_node *node) } static inline bool -filelayout_test_layout_invalid(struct pnfs_layout_hdr *lo) -{ - return test_bit(NFS_LAYOUT_INVALID, &lo->plh_flags); -} - -static inline bool filelayout_test_devid_invalid(struct nfs4_deviceid_node *node) { return test_bit(NFS_DEVICEID_INVALID, &node->flags); } -static inline bool -filelayout_reset_to_mds(struct pnfs_layout_segment *lseg) -{ - return filelayout_test_devid_invalid(FILELAYOUT_DEVID_NODE(lseg)) || - filelayout_test_layout_invalid(lseg->pls_layout); -} +extern bool +filelayout_test_devid_unavailable(struct nfs4_deviceid_node *node); extern struct nfs_fh * nfs4_fl_select_ds_fh(struct pnfs_layout_segment *lseg, u32 j); @@ -158,7 +148,6 @@ struct nfs4_pnfs_ds *nfs4_fl_prepare_ds(struct pnfs_layout_segment *lseg, extern void nfs4_fl_put_deviceid(struct nfs4_file_layout_dsaddr *dsaddr); extern void nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr); struct nfs4_file_layout_dsaddr * -get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id, gfp_t gfp_flags); -void nfs4_ds_disconnect(struct nfs_client *clp); +filelayout_get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id, gfp_t gfp_flags); #endif /* FS_NFS_NFS4FILELAYOUT_H */ diff --git a/fs/nfs/nfs4filelayoutdev.c b/fs/nfs/nfs4filelayoutdev.c index f81231f30d94..a8eaa9b7bb0f 100644 --- a/fs/nfs/nfs4filelayoutdev.c +++ b/fs/nfs/nfs4filelayoutdev.c @@ -149,28 +149,6 @@ _data_server_lookup_locked(const struct list_head *dsaddrs) } /* - * Lookup DS by nfs_client pointer. Zero data server client pointer - */ -void nfs4_ds_disconnect(struct nfs_client *clp) -{ - struct nfs4_pnfs_ds *ds; - struct nfs_client *found = NULL; - - dprintk("%s clp %p\n", __func__, clp); - spin_lock(&nfs4_ds_cache_lock); - list_for_each_entry(ds, &nfs4_data_server_cache, ds_node) - if (ds->ds_clp && ds->ds_clp == clp) { - found = ds->ds_clp; - ds->ds_clp = NULL; - } - spin_unlock(&nfs4_ds_cache_lock); - if (found) { - set_bit(NFS_CS_STOP_RENEW, &clp->cl_res_state); - nfs_put_client(clp); - } -} - -/* * Create an rpc connection to the nfs4_pnfs_ds data server * Currently only supports IPv4 and IPv6 addresses */ @@ -690,7 +668,7 @@ decode_and_add_device(struct inode *inode, struct pnfs_device *dev, gfp_t gfp_fl * of available devices, and return it. */ struct nfs4_file_layout_dsaddr * -get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id, gfp_t gfp_flags) +filelayout_get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id, gfp_t gfp_flags) { struct pnfs_device *pdev = NULL; u32 max_resp_sz; @@ -804,13 +782,14 @@ nfs4_fl_prepare_ds(struct pnfs_layout_segment *lseg, u32 ds_idx) struct nfs4_pnfs_ds *ds = dsaddr->ds_list[ds_idx]; struct nfs4_deviceid_node *devid = FILELAYOUT_DEVID_NODE(lseg); - if (filelayout_test_devid_invalid(devid)) + if (filelayout_test_devid_unavailable(devid)) return NULL; if (ds == NULL) { printk(KERN_ERR "NFS: %s: No data server for offset index %d\n", __func__, ds_idx); - goto mark_dev_invalid; + filelayout_mark_devid_invalid(devid); + return NULL; } if (!ds->ds_clp) { @@ -818,14 +797,12 @@ nfs4_fl_prepare_ds(struct pnfs_layout_segment *lseg, u32 ds_idx) int err; err = nfs4_ds_connect(s, ds); - if (err) - goto mark_dev_invalid; + if (err) { + nfs4_mark_deviceid_unavailable(devid); + return NULL; + } } return ds; - -mark_dev_invalid: - filelayout_mark_devid_invalid(devid); - return NULL; } module_param(dataserver_retrans, uint, 0644); diff --git a/fs/nfs/nfs4getroot.c b/fs/nfs/nfs4getroot.c index 6a83780e0ce6..549462e5b9b0 100644 --- a/fs/nfs/nfs4getroot.c +++ b/fs/nfs/nfs4getroot.c @@ -5,6 +5,7 @@ #include <linux/nfs_fs.h> #include "nfs4_fs.h" +#include "internal.h" #define NFSDBG_FACILITY NFSDBG_CLIENT diff --git a/fs/nfs/nfs4namespace.c b/fs/nfs/nfs4namespace.c index 017b4b01a69c..79fbb61ce202 100644 --- a/fs/nfs/nfs4namespace.c +++ b/fs/nfs/nfs4namespace.c @@ -192,25 +192,13 @@ out: struct rpc_clnt *nfs4_create_sec_client(struct rpc_clnt *clnt, struct inode *inode, struct qstr *name) { - struct rpc_clnt *clone; - struct rpc_auth *auth; rpc_authflavor_t flavor; flavor = nfs4_negotiate_security(inode, name); if ((int)flavor < 0) - return ERR_PTR(flavor); + return ERR_PTR((int)flavor); - clone = rpc_clone_client(clnt); - if (IS_ERR(clone)) - return clone; - - auth = rpcauth_create(flavor, clone); - if (!auth) { - rpc_shutdown_client(clone); - clone = ERR_PTR(-EIO); - } - - return clone; + return rpc_clone_client_set_auth(clnt, flavor); } static struct vfsmount *try_location(struct nfs_clone_mount *mountdata, diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 1e50326d00dd..68b21d81b7ac 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -104,6 +104,8 @@ static int nfs4_map_errors(int err) return -EACCES; case -NFS4ERR_MINOR_VERS_MISMATCH: return -EPROTONOSUPPORT; + case -NFS4ERR_ACCESS: + return -EACCES; default: dprintk("%s could not handle NFSv4 error %d\n", __func__, -err); @@ -150,6 +152,12 @@ static const u32 nfs4_pnfs_open_bitmap[3] = { FATTR4_WORD2_MDSTHRESHOLD }; +static const u32 nfs4_open_noattr_bitmap[3] = { + FATTR4_WORD0_TYPE + | FATTR4_WORD0_CHANGE + | FATTR4_WORD0_FILEID, +}; + const u32 nfs4_statfs_bitmap[2] = { FATTR4_WORD0_FILES_AVAIL | FATTR4_WORD0_FILES_FREE @@ -832,6 +840,7 @@ static void nfs4_init_opendata_res(struct nfs4_opendata *p) p->o_res.seqid = p->o_arg.seqid; p->c_res.seqid = p->c_arg.seqid; p->o_res.server = p->o_arg.server; + p->o_res.access_request = p->o_arg.access; nfs_fattr_init(&p->f_attr); nfs_fattr_init_names(&p->f_attr, &p->owner_name, &p->group_name); } @@ -860,6 +869,14 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry, p->o_arg.fh = NFS_FH(dir); p->o_arg.open_flags = flags; p->o_arg.fmode = fmode & (FMODE_READ|FMODE_WRITE); + /* don't put an ACCESS op in OPEN compound if O_EXCL, because ACCESS + * will return permission denied for all bits until close */ + if (!(flags & O_EXCL)) { + /* ask server to check for all possible rights as results + * are cached */ + p->o_arg.access = NFS4_ACCESS_READ | NFS4_ACCESS_MODIFY | + NFS4_ACCESS_EXTEND | NFS4_ACCESS_EXECUTE; + } p->o_arg.clientid = server->nfs_client->cl_clientid; p->o_arg.id.create_time = ktime_to_ns(sp->so_seqid.create_time); p->o_arg.id.uniquifier = sp->so_seqid.owner_id; @@ -1115,11 +1132,80 @@ out_return_state: return state; } -static struct nfs4_state *nfs4_opendata_to_nfs4_state(struct nfs4_opendata *data) +static void +nfs4_opendata_check_deleg(struct nfs4_opendata *data, struct nfs4_state *state) +{ + struct nfs_client *clp = NFS_SERVER(state->inode)->nfs_client; + struct nfs_delegation *delegation; + int delegation_flags = 0; + + rcu_read_lock(); + delegation = rcu_dereference(NFS_I(state->inode)->delegation); + if (delegation) + delegation_flags = delegation->flags; + rcu_read_unlock(); + if (data->o_arg.claim == NFS4_OPEN_CLAIM_DELEGATE_CUR) { + pr_err_ratelimited("NFS: Broken NFSv4 server %s is " + "returning a delegation for " + "OPEN(CLAIM_DELEGATE_CUR)\n", + clp->cl_hostname); + } else if ((delegation_flags & 1UL<<NFS_DELEGATION_NEED_RECLAIM) == 0) + nfs_inode_set_delegation(state->inode, + data->owner->so_cred, + &data->o_res); + else + nfs_inode_reclaim_delegation(state->inode, + data->owner->so_cred, + &data->o_res); +} + +/* + * Check the inode attributes against the CLAIM_PREVIOUS returned attributes + * and update the nfs4_state. + */ +static struct nfs4_state * +_nfs4_opendata_reclaim_to_nfs4_state(struct nfs4_opendata *data) +{ + struct inode *inode = data->state->inode; + struct nfs4_state *state = data->state; + int ret; + + if (!data->rpc_done) { + ret = data->rpc_status; + goto err; + } + + ret = -ESTALE; + if (!(data->f_attr.valid & NFS_ATTR_FATTR_TYPE) || + !(data->f_attr.valid & NFS_ATTR_FATTR_FILEID) || + !(data->f_attr.valid & NFS_ATTR_FATTR_CHANGE)) + goto err; + + ret = -ENOMEM; + state = nfs4_get_open_state(inode, data->owner); + if (state == NULL) + goto err; + + ret = nfs_refresh_inode(inode, &data->f_attr); + if (ret) + goto err; + + if (data->o_res.delegation_type != 0) + nfs4_opendata_check_deleg(data, state); + update_open_stateid(state, &data->o_res.stateid, NULL, + data->o_arg.fmode); + + return state; +err: + return ERR_PTR(ret); + +} + +static struct nfs4_state * +_nfs4_opendata_to_nfs4_state(struct nfs4_opendata *data) { struct inode *inode; struct nfs4_state *state = NULL; - struct nfs_delegation *delegation; int ret; if (!data->rpc_done) { @@ -1138,30 +1224,8 @@ static struct nfs4_state *nfs4_opendata_to_nfs4_state(struct nfs4_opendata *data state = nfs4_get_open_state(inode, data->owner); if (state == NULL) goto err_put_inode; - if (data->o_res.delegation_type != 0) { - struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; - int delegation_flags = 0; - - rcu_read_lock(); - delegation = rcu_dereference(NFS_I(inode)->delegation); - if (delegation) - delegation_flags = delegation->flags; - rcu_read_unlock(); - if (data->o_arg.claim == NFS4_OPEN_CLAIM_DELEGATE_CUR) { - pr_err_ratelimited("NFS: Broken NFSv4 server %s is " - "returning a delegation for " - "OPEN(CLAIM_DELEGATE_CUR)\n", - clp->cl_hostname); - } else if ((delegation_flags & 1UL<<NFS_DELEGATION_NEED_RECLAIM) == 0) - nfs_inode_set_delegation(state->inode, - data->owner->so_cred, - &data->o_res); - else - nfs_inode_reclaim_delegation(state->inode, - data->owner->so_cred, - &data->o_res); - } - + if (data->o_res.delegation_type != 0) + nfs4_opendata_check_deleg(data, state); update_open_stateid(state, &data->o_res.stateid, NULL, data->o_arg.fmode); iput(inode); @@ -1173,6 +1237,14 @@ err: return ERR_PTR(ret); } +static struct nfs4_state * +nfs4_opendata_to_nfs4_state(struct nfs4_opendata *data) +{ + if (data->o_arg.claim == NFS4_OPEN_CLAIM_PREVIOUS) + return _nfs4_opendata_reclaim_to_nfs4_state(data); + return _nfs4_opendata_to_nfs4_state(data); +} + static struct nfs_open_context *nfs4_state_find_open_context(struct nfs4_state *state) { struct nfs_inode *nfsi = NFS_I(state->inode); @@ -1494,6 +1566,7 @@ static void nfs4_open_prepare(struct rpc_task *task, void *calldata) data->o_arg.clientid = sp->so_server->nfs_client->cl_clientid; if (data->o_arg.claim == NFS4_OPEN_CLAIM_PREVIOUS) { task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_NOATTR]; + data->o_arg.open_bitmap = &nfs4_open_noattr_bitmap[0]; nfs_copy_fh(&data->o_res.fh, data->o_arg.fh); } data->timestamp = jiffies; @@ -1526,7 +1599,8 @@ static void nfs4_open_done(struct rpc_task *task, void *calldata) return; if (task->tk_status == 0) { - switch (data->o_res.f_attr->mode & S_IFMT) { + if (data->o_res.f_attr->valid & NFS_ATTR_FATTR_TYPE) { + switch (data->o_res.f_attr->mode & S_IFMT) { case S_IFREG: break; case S_IFLNK: @@ -1537,6 +1611,7 @@ static void nfs4_open_done(struct rpc_task *task, void *calldata) break; default: data->rpc_status = -ENOTDIR; + } } renew_lease(data->o_res.server, data->timestamp); if (!(data->o_res.rflags & NFS4_OPEN_RESULT_CONFIRM)) @@ -1643,6 +1718,39 @@ static int _nfs4_recover_proc_open(struct nfs4_opendata *data) return status; } +static int nfs4_opendata_access(struct rpc_cred *cred, + struct nfs4_opendata *opendata, + struct nfs4_state *state, fmode_t fmode) +{ + struct nfs_access_entry cache; + u32 mask; + + /* access call failed or for some reason the server doesn't + * support any access modes -- defer access call until later */ + if (opendata->o_res.access_supported == 0) + return 0; + + mask = 0; + /* don't check MAY_WRITE - a newly created file may not have + * write mode bits, but POSIX allows the creating process to write */ + if (fmode & FMODE_READ) + mask |= MAY_READ; + if (fmode & FMODE_EXEC) + mask |= MAY_EXEC; + + cache.cred = cred; + cache.jiffies = jiffies; + nfs_access_set_mask(&cache, opendata->o_res.access_result); + nfs_access_add_cache(state->inode, &cache); + + if ((mask & ~cache.mask & (MAY_READ | MAY_EXEC)) == 0) + return 0; + + /* even though OPEN succeeded, access is denied. Close the file */ + nfs4_close_state(state, fmode); + return -NFS4ERR_ACCESS; +} + /* * Note: On error, nfs4_proc_open will free the struct nfs4_opendata */ @@ -1774,7 +1882,11 @@ static void nfs41_clear_delegation_stateid(struct nfs4_state *state) * informs us the stateid is unrecognized. */ if (status != -NFS4ERR_BAD_STATEID) nfs41_free_stateid(server, stateid); + nfs_remove_bad_delegation(state->inode); + write_seqlock(&state->seqlock); + nfs4_stateid_copy(&state->stateid, &state->open_stateid); + write_sequnlock(&state->seqlock); clear_bit(NFS_DELEGATED_STATE, &state->flags); } } @@ -1790,7 +1902,7 @@ static void nfs41_clear_delegation_stateid(struct nfs4_state *state) static int nfs41_check_open_stateid(struct nfs4_state *state) { struct nfs_server *server = NFS_SERVER(state->inode); - nfs4_stateid *stateid = &state->stateid; + nfs4_stateid *stateid = &state->open_stateid; int status; /* If a state reset has been done, test_stateid is unneeded */ @@ -1896,6 +2008,10 @@ static int _nfs4_do_open(struct inode *dir, if (server->caps & NFS_CAP_POSIX_LOCK) set_bit(NFS_STATE_POSIX_LOCKS, &state->flags); + status = nfs4_opendata_access(cred, opendata, state, fmode); + if (status != 0) + goto err_opendata_put; + if (opendata->o_arg.open_flags & O_EXCL) { nfs4_exclusive_attrset(opendata, sattr); @@ -1941,7 +2057,7 @@ static struct nfs4_state *nfs4_do_open(struct inode *dir, struct nfs4_state *res; int status; - fmode &= FMODE_READ|FMODE_WRITE; + fmode &= FMODE_READ|FMODE_WRITE|FMODE_EXEC; do { status = _nfs4_do_open(dir, dentry, fmode, flags, sattr, cred, &res, ctx_th); @@ -2013,8 +2129,12 @@ static int _nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred, nfs_fattr_init(fattr); if (state != NULL) { + struct nfs_lockowner lockowner = { + .l_owner = current->files, + .l_pid = current->tgid, + }; nfs4_select_rw_stateid(&arg.stateid, state, FMODE_WRITE, - current->files, current->tgid); + &lockowner); } else if (nfs4_copy_delegation_stateid(&arg.stateid, inode, FMODE_WRITE)) { /* Use that stateid */ @@ -2133,6 +2253,7 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data) { struct nfs4_closedata *calldata = data; struct nfs4_state *state = calldata->state; + struct inode *inode = calldata->inode; int call_close = 0; dprintk("%s: begin!\n", __func__); @@ -2166,16 +2287,13 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data) if (calldata->arg.fmode == 0) { task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE]; if (calldata->roc && - pnfs_roc_drain(calldata->inode, &calldata->roc_barrier)) { - rpc_sleep_on(&NFS_SERVER(calldata->inode)->roc_rpcwaitq, - task, NULL); + pnfs_roc_drain(inode, &calldata->roc_barrier, task)) goto out; - } } nfs_fattr_init(calldata->res.fattr); calldata->timestamp = jiffies; - if (nfs4_setup_sequence(NFS_SERVER(calldata->inode), + if (nfs4_setup_sequence(NFS_SERVER(inode), &calldata->arg.seq_args, &calldata->res.seq_res, task)) @@ -2202,7 +2320,7 @@ static const struct rpc_call_ops nfs4_close_ops = { * * NOTE: Caller must be holding the sp->so_owner semaphore! */ -int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait, bool roc) +int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait) { struct nfs_server *server = NFS_SERVER(state->inode); struct nfs4_closedata *calldata; @@ -2238,7 +2356,7 @@ int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait, bool roc) calldata->res.fattr = &calldata->fattr; calldata->res.seqid = calldata->arg.seqid; calldata->res.server = server; - calldata->roc = roc; + calldata->roc = pnfs_roc(state->inode); nfs_sb_active(calldata->inode->i_sb); msg.rpc_argp = &calldata->arg; @@ -2255,8 +2373,6 @@ int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait, bool roc) out_free_calldata: kfree(calldata); out: - if (roc) - pnfs_roc_release(state->inode); nfs4_put_open_state(state); nfs4_put_state_owner(sp); return status; @@ -2399,7 +2515,7 @@ static int nfs4_lookup_root_sec(struct nfs_server *server, struct nfs_fh *fhandl int ret; auth = rpcauth_create(flavor, server->client); - if (!auth) { + if (IS_ERR(auth)) { ret = -EIO; goto out; } @@ -2767,13 +2883,7 @@ static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); if (!status) { - entry->mask = 0; - if (res.access & NFS4_ACCESS_READ) - entry->mask |= MAY_READ; - if (res.access & (NFS4_ACCESS_MODIFY | NFS4_ACCESS_EXTEND | NFS4_ACCESS_DELETE)) - entry->mask |= MAY_WRITE; - if (res.access & (NFS4_ACCESS_LOOKUP|NFS4_ACCESS_EXECUTE)) - entry->mask |= MAY_EXEC; + nfs_access_set_mask(entry, res.access); nfs_refresh_inode(inode, res.fattr); } nfs_free_fattr(res.fattr); @@ -3362,8 +3472,11 @@ static int nfs4_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, s nfs_fattr_init(fsinfo->fattr); error = nfs4_do_fsinfo(server, fhandle, fsinfo); - if (error == 0) + if (error == 0) { + /* block layout checks this! */ + server->pnfs_blksize = fsinfo->blksize; set_pnfs_layoutdriver(server, fhandle, fsinfo->layouttype); + } return error; } @@ -4007,6 +4120,36 @@ static void nfs4_init_boot_verifier(const struct nfs_client *clp, memcpy(bootverf->data, verf, sizeof(bootverf->data)); } +static unsigned int +nfs4_init_nonuniform_client_string(const struct nfs_client *clp, + char *buf, size_t len) +{ + unsigned int result; + + rcu_read_lock(); + result = scnprintf(buf, len, "Linux NFSv4.0 %s/%s %s", + clp->cl_ipaddr, + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_ADDR), + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_PROTO)); + rcu_read_unlock(); + return result; +} + +static unsigned int +nfs4_init_uniform_client_string(const struct nfs_client *clp, + char *buf, size_t len) +{ + char *nodename = clp->cl_rpcclient->cl_nodename; + + if (nfs4_client_id_uniquifier[0] != '\0') + nodename = nfs4_client_id_uniquifier; + return scnprintf(buf, len, "Linux NFSv%u.%u %s", + clp->rpc_ops->version, clp->cl_minorversion, + nodename); +} + /** * nfs4_proc_setclientid - Negotiate client ID * @clp: state data structure @@ -4037,15 +4180,18 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, /* nfs_client_id4 */ nfs4_init_boot_verifier(clp, &sc_verifier); - rcu_read_lock(); - setclientid.sc_name_len = scnprintf(setclientid.sc_name, - sizeof(setclientid.sc_name), "%s/%s %s", - clp->cl_ipaddr, - rpc_peeraddr2str(clp->cl_rpcclient, - RPC_DISPLAY_ADDR), - rpc_peeraddr2str(clp->cl_rpcclient, - RPC_DISPLAY_PROTO)); + if (test_bit(NFS_CS_MIGRATION, &clp->cl_flags)) + setclientid.sc_name_len = + nfs4_init_uniform_client_string(clp, + setclientid.sc_name, + sizeof(setclientid.sc_name)); + else + setclientid.sc_name_len = + nfs4_init_nonuniform_client_string(clp, + setclientid.sc_name, + sizeof(setclientid.sc_name)); /* cb_client4 */ + rcu_read_lock(); setclientid.sc_netid_len = scnprintf(setclientid.sc_netid, sizeof(setclientid.sc_netid), rpc_peeraddr2str(clp->cl_rpcclient, @@ -4391,7 +4537,7 @@ static void nfs4_locku_prepare(struct rpc_task *task, void *data) if (nfs_wait_on_sequence(calldata->arg.seqid, task) != 0) return; - if ((calldata->lsp->ls_flags & NFS_LOCK_INITIALIZED) == 0) { + if (test_bit(NFS_LOCK_INITIALIZED, &calldata->lsp->ls_flags) == 0) { /* Note: exit _without_ running nfs4_locku_done */ task->tk_action = NULL; return; @@ -4585,7 +4731,7 @@ static void nfs4_lock_done(struct rpc_task *task, void *calldata) } if (data->rpc_status == 0) { nfs4_stateid_copy(&data->lsp->ls_stateid, &data->res.stateid); - data->lsp->ls_flags |= NFS_LOCK_INITIALIZED; + set_bit(NFS_LOCK_INITIALIZED, &data->lsp->ls_flags); renew_lease(NFS_SERVER(data->ctx->dentry->d_inode), data->timestamp); } out: @@ -4632,7 +4778,7 @@ static void nfs4_handle_setlk_error(struct nfs_server *server, struct nfs4_lock_ case -NFS4ERR_BAD_STATEID: lsp->ls_seqid.flags &= ~NFS_SEQID_CONFIRMED; if (new_lock_owner != 0 || - (lsp->ls_flags & NFS_LOCK_INITIALIZED) != 0) + test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) != 0) nfs4_schedule_stateid_recovery(server, lsp->ls_state); break; case -NFS4ERR_STALE_STATEID: @@ -4756,7 +4902,7 @@ static int nfs41_check_expired_locks(struct nfs4_state *state) struct nfs_server *server = NFS_SERVER(state->inode); list_for_each_entry(lsp, &state->lock_states, ls_locks) { - if (lsp->ls_flags & NFS_LOCK_INITIALIZED) { + if (test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags)) { status = nfs41_test_stateid(server, &lsp->ls_stateid); if (status != NFS_OK) { /* Free the stateid unless the server @@ -4764,7 +4910,7 @@ static int nfs41_check_expired_locks(struct nfs4_state *state) if (status != -NFS4ERR_BAD_STATEID) nfs41_free_stateid(server, &lsp->ls_stateid); - lsp->ls_flags &= ~NFS_LOCK_INITIALIZED; + clear_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags); ret = status; } } @@ -5267,10 +5413,8 @@ int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred) }; nfs4_init_boot_verifier(clp, &verifier); - args.id_len = scnprintf(args.id, sizeof(args.id), - "%s/%s", - clp->cl_ipaddr, - clp->cl_rpcclient->cl_nodename); + args.id_len = nfs4_init_uniform_client_string(clp, args.id, + sizeof(args.id)); dprintk("NFS call exchange_id auth=%s, '%.*s'\n", clp->cl_rpcclient->cl_auth->au_ops->au_name, args.id_len, args.id); @@ -5391,6 +5535,8 @@ int nfs4_destroy_clientid(struct nfs_client *clp) goto out; if (clp->cl_exchange_flags == 0) goto out; + if (clp->cl_preserve_clid) + goto out; cred = nfs4_get_exchange_id_cred(clp); ret = nfs4_proc_destroy_clientid(clp, cred); if (cred) @@ -6196,26 +6342,44 @@ nfs4_layoutget_prepare(struct rpc_task *task, void *calldata) static void nfs4_layoutget_done(struct rpc_task *task, void *calldata) { struct nfs4_layoutget *lgp = calldata; - struct nfs_server *server = NFS_SERVER(lgp->args.inode); + struct inode *inode = lgp->args.inode; + struct nfs_server *server = NFS_SERVER(inode); + struct pnfs_layout_hdr *lo; + struct nfs4_state *state = NULL; dprintk("--> %s\n", __func__); if (!nfs4_sequence_done(task, &lgp->res.seq_res)) - return; + goto out; switch (task->tk_status) { case 0: - break; + goto out; case -NFS4ERR_LAYOUTTRYLATER: case -NFS4ERR_RECALLCONFLICT: task->tk_status = -NFS4ERR_DELAY; - /* Fall through */ - default: - if (nfs4_async_handle_error(task, server, NULL) == -EAGAIN) { - rpc_restart_call_prepare(task); - return; + break; + case -NFS4ERR_EXPIRED: + case -NFS4ERR_BAD_STATEID: + spin_lock(&inode->i_lock); + lo = NFS_I(inode)->layout; + if (!lo || list_empty(&lo->plh_segs)) { + spin_unlock(&inode->i_lock); + /* If the open stateid was bad, then recover it. */ + state = lgp->args.ctx->state; + } else { + LIST_HEAD(head); + + pnfs_mark_matching_lsegs_invalid(lo, &head, NULL); + spin_unlock(&inode->i_lock); + /* Mark the bad layout state as invalid, then + * retry using the open stateid. */ + pnfs_free_lseg_list(&head); } } + if (nfs4_async_handle_error(task, server, state) == -EAGAIN) + rpc_restart_call_prepare(task); +out: dprintk("<-- %s\n", __func__); } @@ -6282,7 +6446,8 @@ static const struct rpc_call_ops nfs4_layoutget_call_ops = { .rpc_release = nfs4_layoutget_release, }; -void nfs4_proc_layoutget(struct nfs4_layoutget *lgp, gfp_t gfp_flags) +struct pnfs_layout_segment * +nfs4_proc_layoutget(struct nfs4_layoutget *lgp, gfp_t gfp_flags) { struct nfs_server *server = NFS_SERVER(lgp->args.inode); size_t max_pages = max_response_pages(server); @@ -6299,6 +6464,7 @@ void nfs4_proc_layoutget(struct nfs4_layoutget *lgp, gfp_t gfp_flags) .callback_data = lgp, .flags = RPC_TASK_ASYNC, }; + struct pnfs_layout_segment *lseg = NULL; int status = 0; dprintk("--> %s\n", __func__); @@ -6306,7 +6472,7 @@ void nfs4_proc_layoutget(struct nfs4_layoutget *lgp, gfp_t gfp_flags) lgp->args.layout.pages = nfs4_alloc_pages(max_pages, gfp_flags); if (!lgp->args.layout.pages) { nfs4_layoutget_release(lgp); - return; + return ERR_PTR(-ENOMEM); } lgp->args.layout.pglen = max_pages * PAGE_SIZE; @@ -6315,15 +6481,17 @@ void nfs4_proc_layoutget(struct nfs4_layoutget *lgp, gfp_t gfp_flags) nfs41_init_sequence(&lgp->args.seq_args, &lgp->res.seq_res, 0); task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) - return; + return ERR_CAST(task); status = nfs4_wait_for_completion_rpc_task(task); if (status == 0) status = task->tk_status; if (status == 0) - status = pnfs_layout_process(lgp); + lseg = pnfs_layout_process(lgp); rpc_put_task(task); dprintk("<-- %s status=%d\n", __func__, status); - return; + if (status) + return ERR_PTR(status); + return lseg; } static void @@ -6342,7 +6510,6 @@ static void nfs4_layoutreturn_done(struct rpc_task *task, void *calldata) { struct nfs4_layoutreturn *lrp = calldata; struct nfs_server *server; - struct pnfs_layout_hdr *lo = lrp->args.layout; dprintk("--> %s\n", __func__); @@ -6354,20 +6521,21 @@ static void nfs4_layoutreturn_done(struct rpc_task *task, void *calldata) rpc_restart_call_prepare(task); return; } - spin_lock(&lo->plh_inode->i_lock); - if (task->tk_status == 0 && lrp->res.lrs_present) - pnfs_set_layout_stateid(lo, &lrp->res.stateid, true); - lo->plh_block_lgets--; - spin_unlock(&lo->plh_inode->i_lock); dprintk("<-- %s\n", __func__); } static void nfs4_layoutreturn_release(void *calldata) { struct nfs4_layoutreturn *lrp = calldata; + struct pnfs_layout_hdr *lo = lrp->args.layout; dprintk("--> %s\n", __func__); - put_layout_hdr(lrp->args.layout); + spin_lock(&lo->plh_inode->i_lock); + if (lrp->res.lrs_present) + pnfs_set_layout_stateid(lo, &lrp->res.stateid, true); + lo->plh_block_lgets--; + spin_unlock(&lo->plh_inode->i_lock); + pnfs_put_layout_hdr(lrp->args.layout); kfree(calldata); dprintk("<-- %s\n", __func__); } @@ -6541,7 +6709,7 @@ static void nfs4_layoutcommit_release(void *calldata) list_del_init(&lseg->pls_lc_list); if (test_and_clear_bit(NFS_LSEG_LAYOUTCOMMIT, &lseg->pls_flags)) - put_lseg(lseg); + pnfs_put_lseg(lseg); } clear_bit_unlock(NFS_INO_LAYOUTCOMMITTING, bitlock); @@ -6800,6 +6968,7 @@ static const struct nfs4_state_recovery_ops nfs40_reboot_recovery_ops = { .recover_lock = nfs4_lock_reclaim, .establish_clid = nfs4_init_clientid, .get_clid_cred = nfs4_get_setclientid_cred, + .detect_trunking = nfs40_discover_server_trunking, }; #if defined(CONFIG_NFS_V4_1) @@ -6811,6 +6980,7 @@ static const struct nfs4_state_recovery_ops nfs41_reboot_recovery_ops = { .establish_clid = nfs41_init_clientid, .get_clid_cred = nfs4_get_exchange_id_cred, .reclaim_complete = nfs41_proc_reclaim_complete, + .detect_trunking = nfs41_discover_server_trunking, }; #endif /* CONFIG_NFS_V4_1 */ diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 55148def5540..c351e6b39838 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -51,18 +51,21 @@ #include <linux/bitops.h> #include <linux/jiffies.h> +#include <linux/sunrpc/clnt.h> + #include "nfs4_fs.h" #include "callback.h" #include "delegation.h" #include "internal.h" #include "pnfs.h" +#include "netns.h" #define NFSDBG_FACILITY NFSDBG_STATE #define OPENOWNER_POOL_SIZE 8 const nfs4_stateid zero_stateid; - +static DEFINE_MUTEX(nfs_clid_init_mutex); static LIST_HEAD(nfs4_clientid_list); int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred) @@ -73,12 +76,13 @@ int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred) }; unsigned short port; int status; + struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); if (test_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state)) goto do_confirm; - port = nfs_callback_tcpport; + port = nn->nfs_callback_tcpport; if (clp->cl_addr.ss_family == AF_INET6) - port = nfs_callback_tcpport6; + port = nn->nfs_callback_tcpport6; status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, port, cred, &clid); if (status != 0) @@ -96,6 +100,56 @@ out: return status; } +/** + * nfs40_discover_server_trunking - Detect server IP address trunking (mv0) + * + * @clp: nfs_client under test + * @result: OUT: found nfs_client, or clp + * @cred: credential to use for trunking test + * + * Returns zero, a negative errno, or a negative NFS4ERR status. + * If zero is returned, an nfs_client pointer is planted in + * "result". + * + * Note: The returned client may not yet be marked ready. + */ +int nfs40_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **result, + struct rpc_cred *cred) +{ + struct nfs4_setclientid_res clid = { + .clientid = clp->cl_clientid, + .confirm = clp->cl_confirm, + }; + struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); + unsigned short port; + int status; + + port = nn->nfs_callback_tcpport; + if (clp->cl_addr.ss_family == AF_INET6) + port = nn->nfs_callback_tcpport6; + + status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, port, cred, &clid); + if (status != 0) + goto out; + clp->cl_clientid = clid.clientid; + clp->cl_confirm = clid.confirm; + + status = nfs40_walk_client_list(clp, result, cred); + switch (status) { + case -NFS4ERR_STALE_CLIENTID: + set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); + case 0: + /* Sustain the lease, even if it's empty. If the clientid4 + * goes stale it's of no use for trunking discovery. */ + nfs4_schedule_state_renewal(*result); + break; + } + +out: + return status; +} + struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp) { struct rpc_cred *cred = NULL; @@ -275,6 +329,33 @@ out: return status; } +/** + * nfs41_discover_server_trunking - Detect server IP address trunking (mv1) + * + * @clp: nfs_client under test + * @result: OUT: found nfs_client, or clp + * @cred: credential to use for trunking test + * + * Returns NFS4_OK, a negative errno, or a negative NFS4ERR status. + * If NFS4_OK is returned, an nfs_client pointer is planted in + * "result". + * + * Note: The returned client may not yet be marked ready. + */ +int nfs41_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **result, + struct rpc_cred *cred) +{ + int status; + + status = nfs4_proc_exchange_id(clp, cred); + if (status != NFS4_OK) + return status; + set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); + + return nfs41_walk_client_list(clp, result, cred); +} + struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp) { struct rpc_cred *cred; @@ -729,11 +810,8 @@ static void __nfs4_close(struct nfs4_state *state, if (!call_close) { nfs4_put_open_state(state); nfs4_put_state_owner(owner); - } else { - bool roc = pnfs_roc(state->inode); - - nfs4_do_close(state, gfp_mask, wait, roc); - } + } else + nfs4_do_close(state, gfp_mask, wait); } void nfs4_close_state(struct nfs4_state *state, fmode_t fmode) @@ -865,7 +943,7 @@ void nfs4_put_lock_state(struct nfs4_lock_state *lsp) if (list_empty(&state->lock_states)) clear_bit(LK_STATE_IN_USE, &state->flags); spin_unlock(&state->state_lock); - if (lsp->ls_flags & NFS_LOCK_INITIALIZED) { + if (test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags)) { if (nfs4_release_lockowner(lsp) == 0) return; } @@ -911,17 +989,25 @@ int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl) } static bool nfs4_copy_lock_stateid(nfs4_stateid *dst, struct nfs4_state *state, - fl_owner_t fl_owner, pid_t fl_pid) + const struct nfs_lockowner *lockowner) { struct nfs4_lock_state *lsp; + fl_owner_t fl_owner; + pid_t fl_pid; bool ret = false; + + if (lockowner == NULL) + goto out; + if (test_bit(LK_STATE_IN_USE, &state->flags) == 0) goto out; + fl_owner = lockowner->l_owner; + fl_pid = lockowner->l_pid; spin_lock(&state->state_lock); lsp = __nfs4_find_lock_state(state, fl_owner, fl_pid, NFS4_ANY_LOCK_TYPE); - if (lsp != NULL && (lsp->ls_flags & NFS_LOCK_INITIALIZED) != 0) { + if (lsp != NULL && test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) != 0) { nfs4_stateid_copy(dst, &lsp->ls_stateid); ret = true; } @@ -946,11 +1032,11 @@ static void nfs4_copy_open_stateid(nfs4_stateid *dst, struct nfs4_state *state) * requests. */ void nfs4_select_rw_stateid(nfs4_stateid *dst, struct nfs4_state *state, - fmode_t fmode, fl_owner_t fl_owner, pid_t fl_pid) + fmode_t fmode, const struct nfs_lockowner *lockowner) { if (nfs4_copy_delegation_stateid(dst, state->inode, fmode)) return; - if (nfs4_copy_lock_stateid(dst, state, fl_owner, fl_pid)) + if (nfs4_copy_lock_stateid(dst, state, lockowner)) return; nfs4_copy_open_stateid(dst, state); } @@ -1289,7 +1375,7 @@ restart: if (status >= 0) { spin_lock(&state->state_lock); list_for_each_entry(lock, &state->lock_states, ls_locks) { - if (!(lock->ls_flags & NFS_LOCK_INITIALIZED)) + if (!test_bit(NFS_LOCK_INITIALIZED, &lock->ls_flags)) pr_warn_ratelimited("NFS: " "%s: Lock reclaim " "failed!\n", __func__); @@ -1361,7 +1447,7 @@ static void nfs4_clear_open_state(struct nfs4_state *state) spin_lock(&state->state_lock); list_for_each_entry(lock, &state->lock_states, ls_locks) { lock->ls_seqid.flags = 0; - lock->ls_flags &= ~NFS_LOCK_INITIALIZED; + clear_bit(NFS_LOCK_INITIALIZED, &lock->ls_flags); } spin_unlock(&state->state_lock); } @@ -1595,8 +1681,8 @@ out: return nfs4_recovery_handle_error(clp, status); } -/* Set NFS4CLNT_LEASE_EXPIRED for all v4.0 errors and for recoverable errors - * on EXCHANGE_ID for v4.1 +/* Set NFS4CLNT_LEASE_EXPIRED and reclaim reboot state for all v4.0 errors + * and for recoverable errors on EXCHANGE_ID for v4.1 */ static int nfs4_handle_reclaim_lease_error(struct nfs_client *clp, int status) { @@ -1606,8 +1692,12 @@ static int nfs4_handle_reclaim_lease_error(struct nfs_client *clp, int status) return -ESERVERFAULT; /* Lease confirmation error: retry after purging the lease */ ssleep(1); + clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); + break; case -NFS4ERR_STALE_CLIENTID: clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); + nfs4_state_clear_reclaim_reboot(clp); + nfs4_state_start_reclaim_reboot(clp); break; case -NFS4ERR_CLID_INUSE: pr_err("NFS: Server %s reports our clientid is in use\n", @@ -1698,6 +1788,109 @@ static int nfs4_purge_lease(struct nfs_client *clp) return 0; } +/** + * nfs4_discover_server_trunking - Detect server IP address trunking + * + * @clp: nfs_client under test + * @result: OUT: found nfs_client, or clp + * + * Returns zero or a negative errno. If zero is returned, + * an nfs_client pointer is planted in "result". + * + * Note: since we are invoked in process context, and + * not from inside the state manager, we cannot use + * nfs4_handle_reclaim_lease_error(). + */ +int nfs4_discover_server_trunking(struct nfs_client *clp, + struct nfs_client **result) +{ + const struct nfs4_state_recovery_ops *ops = + clp->cl_mvops->reboot_recovery_ops; + rpc_authflavor_t *flavors, flav, save; + struct rpc_clnt *clnt; + struct rpc_cred *cred; + int i, len, status; + + dprintk("NFS: %s: testing '%s'\n", __func__, clp->cl_hostname); + + len = NFS_MAX_SECFLAVORS; + flavors = kcalloc(len, sizeof(*flavors), GFP_KERNEL); + if (flavors == NULL) { + status = -ENOMEM; + goto out; + } + len = rpcauth_list_flavors(flavors, len); + if (len < 0) { + status = len; + goto out_free; + } + clnt = clp->cl_rpcclient; + save = clnt->cl_auth->au_flavor; + i = 0; + + mutex_lock(&nfs_clid_init_mutex); + status = -ENOENT; +again: + cred = ops->get_clid_cred(clp); + if (cred == NULL) + goto out_unlock; + + status = ops->detect_trunking(clp, result, cred); + put_rpccred(cred); + switch (status) { + case 0: + break; + + case -EACCES: + if (clp->cl_machine_cred == NULL) + break; + /* Handle case where the user hasn't set up machine creds */ + nfs4_clear_machine_cred(clp); + case -NFS4ERR_DELAY: + case -ETIMEDOUT: + case -EAGAIN: + ssleep(1); + dprintk("NFS: %s after status %d, retrying\n", + __func__, status); + goto again; + + case -NFS4ERR_CLID_INUSE: + case -NFS4ERR_WRONGSEC: + status = -EPERM; + if (i >= len) + break; + + flav = flavors[i++]; + if (flav == save) + flav = flavors[i++]; + clnt = rpc_clone_client_set_auth(clnt, flav); + if (IS_ERR(clnt)) { + status = PTR_ERR(clnt); + break; + } + clp->cl_rpcclient = clnt; + goto again; + + case -NFS4ERR_MINOR_VERS_MISMATCH: + status = -EPROTONOSUPPORT; + break; + + case -EKEYEXPIRED: + nfs4_warn_keyexpired(clp->cl_hostname); + case -NFS4ERR_NOT_SAME: /* FixMe: implement recovery + * in nfs4_exchange_id */ + status = -EKEYEXPIRED; + } + +out_unlock: + mutex_unlock(&nfs_clid_init_mutex); +out_free: + kfree(flavors); +out: + dprintk("NFS: %s: status = %d\n", __func__, status); + return status; +} + #ifdef CONFIG_NFS_V4_1 void nfs4_schedule_session_recovery(struct nfs4_session *session, int err) { @@ -2008,6 +2201,7 @@ out_error: pr_warn_ratelimited("NFS: state manager%s%s failed on NFSv4 server %s" " with error %d\n", section_sep, section, clp->cl_hostname, -status); + ssleep(1); nfs4_end_drain_session(clp); nfs4_clear_state_manager_bit(clp); } diff --git a/fs/nfs/nfs4sysctl.c b/fs/nfs/nfs4sysctl.c index 5729bc8aa75d..2628d921b7e3 100644 --- a/fs/nfs/nfs4sysctl.c +++ b/fs/nfs/nfs4sysctl.c @@ -9,6 +9,7 @@ #include <linux/nfs_idmap.h> #include <linux/nfs_fs.h> +#include "nfs4_fs.h" #include "callback.h" static const int nfs_set_port_min = 0; diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 8dba6bd48557..40836ee5dc3a 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -447,12 +447,14 @@ static int nfs4_stat_to_errno(int); encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_open_maxsz + \ + encode_access_maxsz + \ encode_getfh_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_open_sz (compound_decode_hdr_maxsz + \ decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_open_maxsz + \ + decode_access_maxsz + \ decode_getfh_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_open_confirm_sz \ @@ -467,11 +469,13 @@ static int nfs4_stat_to_errno(int); encode_sequence_maxsz + \ encode_putfh_maxsz + \ encode_open_maxsz + \ + encode_access_maxsz + \ encode_getattr_maxsz) #define NFS4_dec_open_noattr_sz (compound_decode_hdr_maxsz + \ decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_open_maxsz + \ + decode_access_maxsz + \ decode_getattr_maxsz) #define NFS4_enc_open_downgrade_sz \ (compound_encode_hdr_maxsz + \ @@ -1509,8 +1513,12 @@ static void encode_open_stateid(struct xdr_stream *xdr, nfs4_stateid stateid; if (ctx->state != NULL) { + const struct nfs_lockowner *lockowner = NULL; + + if (l_ctx != NULL) + lockowner = &l_ctx->lockowner; nfs4_select_rw_stateid(&stateid, ctx->state, - fmode, l_ctx->lockowner, l_ctx->pid); + fmode, lockowner); if (zero_seqid) stateid.seqid = 0; encode_nfs4_stateid(xdr, &stateid); @@ -2216,6 +2224,8 @@ static void nfs4_xdr_enc_open(struct rpc_rqst *req, struct xdr_stream *xdr, encode_putfh(xdr, args->fh, &hdr); encode_open(xdr, args, &hdr); encode_getfh(xdr, &hdr); + if (args->access) + encode_access(xdr, args->access, &hdr); encode_getfattr_open(xdr, args->bitmask, args->open_bitmap, &hdr); encode_nops(&hdr); } @@ -2252,7 +2262,9 @@ static void nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, encode_sequence(xdr, &args->seq_args, &hdr); encode_putfh(xdr, args->fh, &hdr); encode_open(xdr, args, &hdr); - encode_getfattr(xdr, args->bitmask, &hdr); + if (args->access) + encode_access(xdr, args->access, &hdr); + encode_getfattr_open(xdr, args->bitmask, args->open_bitmap, &hdr); encode_nops(&hdr); } @@ -4095,7 +4107,7 @@ out_overflow: return -EIO; } -static int decode_access(struct xdr_stream *xdr, struct nfs4_accessres *access) +static int decode_access(struct xdr_stream *xdr, u32 *supported, u32 *access) { __be32 *p; uint32_t supp, acc; @@ -4109,8 +4121,8 @@ static int decode_access(struct xdr_stream *xdr, struct nfs4_accessres *access) goto out_overflow; supp = be32_to_cpup(p++); acc = be32_to_cpup(p); - access->supported = supp; - access->access = acc; + *supported = supp; + *access = acc; return 0; out_overflow: print_overflow_msg(__func__, xdr); @@ -5642,7 +5654,8 @@ static int decode_getdeviceinfo(struct xdr_stream *xdr, * and places the remaining xdr data in xdr_buf->tail */ pdev->mincount = be32_to_cpup(p); - xdr_read_pages(xdr, pdev->mincount); /* include space for the length */ + if (xdr_read_pages(xdr, pdev->mincount) != pdev->mincount) + goto out_overflow; /* Parse notification bitmap, verifying that it is zero. */ p = xdr_inline_decode(xdr, 4); @@ -5887,7 +5900,7 @@ static int nfs4_xdr_dec_access(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_putfh(xdr); if (status != 0) goto out; - status = decode_access(xdr, res); + status = decode_access(xdr, &res->supported, &res->access); if (status != 0) goto out; decode_getfattr(xdr, res->fattr, res->server); @@ -6228,6 +6241,8 @@ static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_getfh(xdr, &res->fh); if (status) goto out; + if (res->access_request) + decode_access(xdr, &res->access_supported, &res->access_result); decode_getfattr(xdr, res->f_attr, res->server); out: return status; @@ -6276,6 +6291,8 @@ static int nfs4_xdr_dec_open_noattr(struct rpc_rqst *rqstp, status = decode_open(xdr, res); if (status) goto out; + if (res->access_request) + decode_access(xdr, &res->access_supported, &res->access_result); decode_getfattr(xdr, res->f_attr, res->server); out: return status; diff --git a/fs/nfs/objlayout/objio_osd.c b/fs/nfs/objlayout/objio_osd.c index ea6d111b03e9..c6f990656f89 100644 --- a/fs/nfs/objlayout/objio_osd.c +++ b/fs/nfs/objlayout/objio_osd.c @@ -41,6 +41,7 @@ #include <scsi/osd_ore.h> #include "objlayout.h" +#include "../internal.h" #define NFSDBG_FACILITY NFSDBG_PNFS_LD @@ -368,7 +369,7 @@ void objio_free_result(struct objlayout_io_res *oir) kfree(objios); } -enum pnfs_osd_errno osd_pri_2_pnfs_err(enum osd_err_priority oep) +static enum pnfs_osd_errno osd_pri_2_pnfs_err(enum osd_err_priority oep) { switch (oep) { case OSD_ERR_PRI_NO_ERROR: @@ -573,7 +574,7 @@ static bool objio_pg_test(struct nfs_pageio_descriptor *pgio, (unsigned long)pgio->pg_layout_private; } -void objio_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) +static void objio_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { pnfs_generic_pg_init_read(pgio, req); if (unlikely(pgio->pg_lseg == NULL)) @@ -603,11 +604,17 @@ static bool aligned_on_raid_stripe(u64 offset, struct ore_layout *layout, return false; } -void objio_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) +static void objio_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { unsigned long stripe_end = 0; + u64 wb_size; - pnfs_generic_pg_init_write(pgio, req); + if (pgio->pg_dreq == NULL) + wb_size = i_size_read(pgio->pg_inode) - req_offset(req); + else + wb_size = nfs_dreq_bytes_left(pgio->pg_dreq); + + pnfs_generic_pg_init_write(pgio, req, wb_size); if (unlikely(pgio->pg_lseg == NULL)) return; /* Not pNFS */ diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 311a79681e2b..e56e846e9d2d 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -102,6 +102,7 @@ nfs_create_request(struct nfs_open_context *ctx, struct inode *inode, unsigned int offset, unsigned int count) { struct nfs_page *req; + struct nfs_lock_context *l_ctx; /* try to allocate the request struct */ req = nfs_page_alloc(); @@ -109,11 +110,12 @@ nfs_create_request(struct nfs_open_context *ctx, struct inode *inode, return ERR_PTR(-ENOMEM); /* get lock context early so we can deal with alloc failures */ - req->wb_lock_context = nfs_get_lock_context(ctx); - if (req->wb_lock_context == NULL) { + l_ctx = nfs_get_lock_context(ctx); + if (IS_ERR(l_ctx)) { nfs_page_free(req); - return ERR_PTR(-ENOMEM); + return ERR_CAST(l_ctx); } + req->wb_lock_context = l_ctx; /* Initialize the request struct. Initially, we assume a * long write-back delay. This will be adjusted in @@ -290,7 +292,9 @@ static bool nfs_can_coalesce_requests(struct nfs_page *prev, { if (req->wb_context->cred != prev->wb_context->cred) return false; - if (req->wb_lock_context->lockowner != prev->wb_lock_context->lockowner) + if (req->wb_lock_context->lockowner.l_owner != prev->wb_lock_context->lockowner.l_owner) + return false; + if (req->wb_lock_context->lockowner.l_pid != prev->wb_lock_context->lockowner.l_pid) return false; if (req->wb_context->state != prev->wb_context->state) return false; diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 2e00feacd4be..fe624c91bd00 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -35,6 +35,7 @@ #include "iostat.h" #define NFSDBG_FACILITY NFSDBG_PNFS +#define PNFS_LAYOUTGET_RETRY_TIMEOUT (120*HZ) /* Locking: * @@ -190,7 +191,7 @@ EXPORT_SYMBOL_GPL(pnfs_unregister_layoutdriver); /* Need to hold i_lock if caller does not already hold reference */ void -get_layout_hdr(struct pnfs_layout_hdr *lo) +pnfs_get_layout_hdr(struct pnfs_layout_hdr *lo) { atomic_inc(&lo->plh_refcount); } @@ -199,43 +200,107 @@ static struct pnfs_layout_hdr * pnfs_alloc_layout_hdr(struct inode *ino, gfp_t gfp_flags) { struct pnfs_layoutdriver_type *ld = NFS_SERVER(ino)->pnfs_curr_ld; - return ld->alloc_layout_hdr ? ld->alloc_layout_hdr(ino, gfp_flags) : - kzalloc(sizeof(struct pnfs_layout_hdr), gfp_flags); + return ld->alloc_layout_hdr(ino, gfp_flags); } static void pnfs_free_layout_hdr(struct pnfs_layout_hdr *lo) { - struct pnfs_layoutdriver_type *ld = NFS_SERVER(lo->plh_inode)->pnfs_curr_ld; + struct nfs_server *server = NFS_SERVER(lo->plh_inode); + struct pnfs_layoutdriver_type *ld = server->pnfs_curr_ld; + + if (!list_empty(&lo->plh_layouts)) { + struct nfs_client *clp = server->nfs_client; + + spin_lock(&clp->cl_lock); + list_del_init(&lo->plh_layouts); + spin_unlock(&clp->cl_lock); + } put_rpccred(lo->plh_lc_cred); - return ld->alloc_layout_hdr ? ld->free_layout_hdr(lo) : kfree(lo); + return ld->free_layout_hdr(lo); } static void -destroy_layout_hdr(struct pnfs_layout_hdr *lo) +pnfs_detach_layout_hdr(struct pnfs_layout_hdr *lo) { + struct nfs_inode *nfsi = NFS_I(lo->plh_inode); dprintk("%s: freeing layout cache %p\n", __func__, lo); - BUG_ON(!list_empty(&lo->plh_layouts)); - NFS_I(lo->plh_inode)->layout = NULL; - pnfs_free_layout_hdr(lo); + nfsi->layout = NULL; + /* Reset MDS Threshold I/O counters */ + nfsi->write_io = 0; + nfsi->read_io = 0; +} + +void +pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo) +{ + struct inode *inode = lo->plh_inode; + + if (atomic_dec_and_lock(&lo->plh_refcount, &inode->i_lock)) { + pnfs_detach_layout_hdr(lo); + spin_unlock(&inode->i_lock); + pnfs_free_layout_hdr(lo); + } +} + +static int +pnfs_iomode_to_fail_bit(u32 iomode) +{ + return iomode == IOMODE_RW ? + NFS_LAYOUT_RW_FAILED : NFS_LAYOUT_RO_FAILED; } static void -put_layout_hdr_locked(struct pnfs_layout_hdr *lo) +pnfs_layout_set_fail_bit(struct pnfs_layout_hdr *lo, int fail_bit) { - if (atomic_dec_and_test(&lo->plh_refcount)) - destroy_layout_hdr(lo); + lo->plh_retry_timestamp = jiffies; + if (test_and_set_bit(fail_bit, &lo->plh_flags)) + atomic_inc(&lo->plh_refcount); } -void -put_layout_hdr(struct pnfs_layout_hdr *lo) +static void +pnfs_layout_clear_fail_bit(struct pnfs_layout_hdr *lo, int fail_bit) +{ + if (test_and_clear_bit(fail_bit, &lo->plh_flags)) + atomic_dec(&lo->plh_refcount); +} + +static void +pnfs_layout_io_set_failed(struct pnfs_layout_hdr *lo, u32 iomode) { struct inode *inode = lo->plh_inode; + struct pnfs_layout_range range = { + .iomode = iomode, + .offset = 0, + .length = NFS4_MAX_UINT64, + }; + LIST_HEAD(head); - if (atomic_dec_and_lock(&lo->plh_refcount, &inode->i_lock)) { - destroy_layout_hdr(lo); - spin_unlock(&inode->i_lock); + spin_lock(&inode->i_lock); + pnfs_layout_set_fail_bit(lo, pnfs_iomode_to_fail_bit(iomode)); + pnfs_mark_matching_lsegs_invalid(lo, &head, &range); + spin_unlock(&inode->i_lock); + pnfs_free_lseg_list(&head); + dprintk("%s Setting layout IOMODE_%s fail bit\n", __func__, + iomode == IOMODE_RW ? "RW" : "READ"); +} + +static bool +pnfs_layout_io_test_failed(struct pnfs_layout_hdr *lo, u32 iomode) +{ + unsigned long start, end; + int fail_bit = pnfs_iomode_to_fail_bit(iomode); + + if (test_bit(fail_bit, &lo->plh_flags) == 0) + return false; + end = jiffies; + start = end - PNFS_LAYOUTGET_RETRY_TIMEOUT; + if (!time_in_range(lo->plh_retry_timestamp, start, end)) { + /* It is time to retry the failed layoutgets */ + pnfs_layout_clear_fail_bit(lo, fail_bit); + return false; } + return true; } static void @@ -249,33 +314,32 @@ init_lseg(struct pnfs_layout_hdr *lo, struct pnfs_layout_segment *lseg) lseg->pls_layout = lo; } -static void free_lseg(struct pnfs_layout_segment *lseg) +static void pnfs_free_lseg(struct pnfs_layout_segment *lseg) { struct inode *ino = lseg->pls_layout->plh_inode; NFS_SERVER(ino)->pnfs_curr_ld->free_lseg(lseg); - /* Matched by get_layout_hdr in pnfs_insert_layout */ - put_layout_hdr(NFS_I(ino)->layout); } static void -put_lseg_common(struct pnfs_layout_segment *lseg) +pnfs_layout_remove_lseg(struct pnfs_layout_hdr *lo, + struct pnfs_layout_segment *lseg) { - struct inode *inode = lseg->pls_layout->plh_inode; + struct inode *inode = lo->plh_inode; WARN_ON(test_bit(NFS_LSEG_VALID, &lseg->pls_flags)); list_del_init(&lseg->pls_list); - if (list_empty(&lseg->pls_layout->plh_segs)) { - set_bit(NFS_LAYOUT_DESTROYED, &lseg->pls_layout->plh_flags); - /* Matched by initial refcount set in alloc_init_layout_hdr */ - put_layout_hdr_locked(lseg->pls_layout); - } + /* Matched by pnfs_get_layout_hdr in pnfs_layout_insert_lseg */ + atomic_dec(&lo->plh_refcount); + if (list_empty(&lo->plh_segs)) + clear_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags); rpc_wake_up(&NFS_SERVER(inode)->roc_rpcwaitq); } void -put_lseg(struct pnfs_layout_segment *lseg) +pnfs_put_lseg(struct pnfs_layout_segment *lseg) { + struct pnfs_layout_hdr *lo; struct inode *inode; if (!lseg) @@ -284,17 +348,17 @@ put_lseg(struct pnfs_layout_segment *lseg) dprintk("%s: lseg %p ref %d valid %d\n", __func__, lseg, atomic_read(&lseg->pls_refcount), test_bit(NFS_LSEG_VALID, &lseg->pls_flags)); - inode = lseg->pls_layout->plh_inode; + lo = lseg->pls_layout; + inode = lo->plh_inode; if (atomic_dec_and_lock(&lseg->pls_refcount, &inode->i_lock)) { - LIST_HEAD(free_me); - - put_lseg_common(lseg); - list_add(&lseg->pls_list, &free_me); + pnfs_get_layout_hdr(lo); + pnfs_layout_remove_lseg(lo, lseg); spin_unlock(&inode->i_lock); - pnfs_free_lseg_list(&free_me); + pnfs_free_lseg(lseg); + pnfs_put_layout_hdr(lo); } } -EXPORT_SYMBOL_GPL(put_lseg); +EXPORT_SYMBOL_GPL(pnfs_put_lseg); static inline u64 end_offset(u64 start, u64 len) @@ -378,7 +442,7 @@ static int mark_lseg_invalid(struct pnfs_layout_segment *lseg, dprintk("%s: lseg %p ref %d\n", __func__, lseg, atomic_read(&lseg->pls_refcount)); if (atomic_dec_and_test(&lseg->pls_refcount)) { - put_lseg_common(lseg); + pnfs_layout_remove_lseg(lseg->pls_layout, lseg); list_add(&lseg->pls_list, tmp_list); rv = 1; } @@ -390,7 +454,7 @@ static int mark_lseg_invalid(struct pnfs_layout_segment *lseg, * after call. */ int -mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo, +pnfs_mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo, struct list_head *tmp_list, struct pnfs_layout_range *recall_range) { @@ -399,14 +463,8 @@ mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo, dprintk("%s:Begin lo %p\n", __func__, lo); - if (list_empty(&lo->plh_segs)) { - /* Reset MDS Threshold I/O counters */ - NFS_I(lo->plh_inode)->write_io = 0; - NFS_I(lo->plh_inode)->read_io = 0; - if (!test_and_set_bit(NFS_LAYOUT_DESTROYED, &lo->plh_flags)) - put_layout_hdr_locked(lo); + if (list_empty(&lo->plh_segs)) return 0; - } list_for_each_entry_safe(lseg, next, &lo->plh_segs, pls_list) if (!recall_range || should_free_lseg(&lseg->pls_range, recall_range)) { @@ -426,25 +484,13 @@ void pnfs_free_lseg_list(struct list_head *free_me) { struct pnfs_layout_segment *lseg, *tmp; - struct pnfs_layout_hdr *lo; if (list_empty(free_me)) return; - lo = list_first_entry(free_me, struct pnfs_layout_segment, - pls_list)->pls_layout; - - if (test_bit(NFS_LAYOUT_DESTROYED, &lo->plh_flags)) { - struct nfs_client *clp; - - clp = NFS_SERVER(lo->plh_inode)->nfs_client; - spin_lock(&clp->cl_lock); - list_del_init(&lo->plh_layouts); - spin_unlock(&clp->cl_lock); - } list_for_each_entry_safe(lseg, tmp, free_me, pls_list) { list_del(&lseg->pls_list); - free_lseg(lseg); + pnfs_free_lseg(lseg); } } @@ -458,10 +504,15 @@ pnfs_destroy_layout(struct nfs_inode *nfsi) lo = nfsi->layout; if (lo) { lo->plh_block_lgets++; /* permanently block new LAYOUTGETs */ - mark_matching_lsegs_invalid(lo, &tmp_list, NULL); - } - spin_unlock(&nfsi->vfs_inode.i_lock); - pnfs_free_lseg_list(&tmp_list); + pnfs_mark_matching_lsegs_invalid(lo, &tmp_list, NULL); + pnfs_get_layout_hdr(lo); + pnfs_layout_clear_fail_bit(lo, NFS_LAYOUT_RO_FAILED); + pnfs_layout_clear_fail_bit(lo, NFS_LAYOUT_RW_FAILED); + spin_unlock(&nfsi->vfs_inode.i_lock); + pnfs_free_lseg_list(&tmp_list); + pnfs_put_layout_hdr(lo); + } else + spin_unlock(&nfsi->vfs_inode.i_lock); } EXPORT_SYMBOL_GPL(pnfs_destroy_layout); @@ -498,46 +549,54 @@ pnfs_destroy_all_layouts(struct nfs_client *clp) } } +/* + * Compare 2 layout stateid sequence ids, to see which is newer, + * taking into account wraparound issues. + */ +static bool pnfs_seqid_is_newer(u32 s1, u32 s2) +{ + return (s32)s1 - (s32)s2 > 0; +} + /* update lo->plh_stateid with new if is more recent */ void pnfs_set_layout_stateid(struct pnfs_layout_hdr *lo, const nfs4_stateid *new, bool update_barrier) { - u32 oldseq, newseq; + u32 oldseq, newseq, new_barrier; + int empty = list_empty(&lo->plh_segs); oldseq = be32_to_cpu(lo->plh_stateid.seqid); newseq = be32_to_cpu(new->seqid); - if ((int)(newseq - oldseq) > 0) { + if (empty || pnfs_seqid_is_newer(newseq, oldseq)) { nfs4_stateid_copy(&lo->plh_stateid, new); if (update_barrier) { - u32 new_barrier = be32_to_cpu(new->seqid); - - if ((int)(new_barrier - lo->plh_barrier)) - lo->plh_barrier = new_barrier; + new_barrier = be32_to_cpu(new->seqid); } else { /* Because of wraparound, we want to keep the barrier - * "close" to the current seqids. It needs to be - * within 2**31 to count as "behind", so if it - * gets too near that limit, give us a litle leeway - * and bring it to within 2**30. - * NOTE - and yes, this is all unsigned arithmetic. + * "close" to the current seqids. */ - if (unlikely((newseq - lo->plh_barrier) > (3 << 29))) - lo->plh_barrier = newseq - (1 << 30); + new_barrier = newseq - atomic_read(&lo->plh_outstanding); } + if (empty || pnfs_seqid_is_newer(new_barrier, lo->plh_barrier)) + lo->plh_barrier = new_barrier; } } +static bool +pnfs_layout_stateid_blocked(const struct pnfs_layout_hdr *lo, + const nfs4_stateid *stateid) +{ + u32 seqid = be32_to_cpu(stateid->seqid); + + return !pnfs_seqid_is_newer(seqid, lo->plh_barrier); +} + /* lget is set to 1 if called from inside send_layoutget call chain */ static bool -pnfs_layoutgets_blocked(struct pnfs_layout_hdr *lo, nfs4_stateid *stateid, - int lget) +pnfs_layoutgets_blocked(const struct pnfs_layout_hdr *lo, int lget) { - if ((stateid) && - (int)(lo->plh_barrier - be32_to_cpu(stateid->seqid)) >= 0) - return true; return lo->plh_block_lgets || - test_bit(NFS_LAYOUT_DESTROYED, &lo->plh_flags) || test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags) || (list_empty(&lo->plh_segs) && (atomic_read(&lo->plh_outstanding) > lget)); @@ -551,7 +610,7 @@ pnfs_choose_layoutget_stateid(nfs4_stateid *dst, struct pnfs_layout_hdr *lo, dprintk("--> %s\n", __func__); spin_lock(&lo->plh_inode->i_lock); - if (pnfs_layoutgets_blocked(lo, NULL, 1)) { + if (pnfs_layoutgets_blocked(lo, 1)) { status = -EAGAIN; } else if (list_empty(&lo->plh_segs)) { int seq; @@ -582,7 +641,7 @@ send_layoutget(struct pnfs_layout_hdr *lo, struct inode *ino = lo->plh_inode; struct nfs_server *server = NFS_SERVER(ino); struct nfs4_layoutget *lgp; - struct pnfs_layout_segment *lseg = NULL; + struct pnfs_layout_segment *lseg; dprintk("--> %s\n", __func__); @@ -599,16 +658,22 @@ send_layoutget(struct pnfs_layout_hdr *lo, lgp->args.type = server->pnfs_curr_ld->id; lgp->args.inode = ino; lgp->args.ctx = get_nfs_open_context(ctx); - lgp->lsegpp = &lseg; lgp->gfp_flags = gfp_flags; /* Synchronously retrieve layout information from server and * store in lseg. */ - nfs4_proc_layoutget(lgp, gfp_flags); - if (!lseg) { - /* remember that LAYOUTGET failed and suspend trying */ - set_bit(lo_fail_bit(range->iomode), &lo->plh_flags); + lseg = nfs4_proc_layoutget(lgp, gfp_flags); + if (IS_ERR(lseg)) { + switch (PTR_ERR(lseg)) { + case -ENOMEM: + case -ERESTARTSYS: + break; + default: + /* remember that LAYOUTGET failed and suspend trying */ + pnfs_layout_io_set_failed(lo, range->iomode); + } + return NULL; } return lseg; @@ -636,25 +701,24 @@ _pnfs_return_layout(struct inode *ino) spin_lock(&ino->i_lock); lo = nfsi->layout; - if (!lo || pnfs_test_layout_returned(lo)) { + if (!lo) { spin_unlock(&ino->i_lock); dprintk("NFS: %s no layout to return\n", __func__); goto out; } stateid = nfsi->layout->plh_stateid; /* Reference matched in nfs4_layoutreturn_release */ - get_layout_hdr(lo); + pnfs_get_layout_hdr(lo); empty = list_empty(&lo->plh_segs); - mark_matching_lsegs_invalid(lo, &tmp_list, NULL); + pnfs_mark_matching_lsegs_invalid(lo, &tmp_list, NULL); /* Don't send a LAYOUTRETURN if list was initially empty */ if (empty) { spin_unlock(&ino->i_lock); - put_layout_hdr(lo); + pnfs_put_layout_hdr(lo); dprintk("NFS: %s no layout segments to return\n", __func__); goto out; } lo->plh_block_lgets++; - pnfs_mark_layout_returned(lo); spin_unlock(&ino->i_lock); pnfs_free_lseg_list(&tmp_list); @@ -663,10 +727,10 @@ _pnfs_return_layout(struct inode *ino) lrp = kzalloc(sizeof(*lrp), GFP_KERNEL); if (unlikely(lrp == NULL)) { status = -ENOMEM; - set_bit(NFS_LAYOUT_RW_FAILED, &lo->plh_flags); - set_bit(NFS_LAYOUT_RO_FAILED, &lo->plh_flags); - pnfs_clear_layout_returned(lo); - put_layout_hdr(lo); + spin_lock(&ino->i_lock); + lo->plh_block_lgets--; + spin_unlock(&ino->i_lock); + pnfs_put_layout_hdr(lo); goto out; } @@ -703,7 +767,7 @@ bool pnfs_roc(struct inode *ino) if (!found) goto out_nolayout; lo->plh_block_lgets++; - get_layout_hdr(lo); /* matched in pnfs_roc_release */ + pnfs_get_layout_hdr(lo); /* matched in pnfs_roc_release */ spin_unlock(&ino->i_lock); pnfs_free_lseg_list(&tmp_list); return true; @@ -720,8 +784,12 @@ void pnfs_roc_release(struct inode *ino) spin_lock(&ino->i_lock); lo = NFS_I(ino)->layout; lo->plh_block_lgets--; - put_layout_hdr_locked(lo); - spin_unlock(&ino->i_lock); + if (atomic_dec_and_test(&lo->plh_refcount)) { + pnfs_detach_layout_hdr(lo); + spin_unlock(&ino->i_lock); + pnfs_free_layout_hdr(lo); + } else + spin_unlock(&ino->i_lock); } void pnfs_roc_set_barrier(struct inode *ino, u32 barrier) @@ -730,32 +798,34 @@ void pnfs_roc_set_barrier(struct inode *ino, u32 barrier) spin_lock(&ino->i_lock); lo = NFS_I(ino)->layout; - if ((int)(barrier - lo->plh_barrier) > 0) + if (pnfs_seqid_is_newer(barrier, lo->plh_barrier)) lo->plh_barrier = barrier; spin_unlock(&ino->i_lock); } -bool pnfs_roc_drain(struct inode *ino, u32 *barrier) +bool pnfs_roc_drain(struct inode *ino, u32 *barrier, struct rpc_task *task) { struct nfs_inode *nfsi = NFS_I(ino); + struct pnfs_layout_hdr *lo; struct pnfs_layout_segment *lseg; + u32 current_seqid; bool found = false; spin_lock(&ino->i_lock); list_for_each_entry(lseg, &nfsi->layout->plh_segs, pls_list) if (test_bit(NFS_LSEG_ROC, &lseg->pls_flags)) { + rpc_sleep_on(&NFS_SERVER(ino)->roc_rpcwaitq, task, NULL); found = true; - break; + goto out; } - if (!found) { - struct pnfs_layout_hdr *lo = nfsi->layout; - u32 current_seqid = be32_to_cpu(lo->plh_stateid.seqid); + lo = nfsi->layout; + current_seqid = be32_to_cpu(lo->plh_stateid.seqid); - /* Since close does not return a layout stateid for use as - * a barrier, we choose the worst-case barrier. - */ - *barrier = current_seqid + atomic_read(&lo->plh_outstanding); - } + /* Since close does not return a layout stateid for use as + * a barrier, we choose the worst-case barrier. + */ + *barrier = current_seqid + atomic_read(&lo->plh_outstanding); +out: spin_unlock(&ino->i_lock); return found; } @@ -786,14 +856,13 @@ cmp_layout(struct pnfs_layout_range *l1, } static void -pnfs_insert_layout(struct pnfs_layout_hdr *lo, +pnfs_layout_insert_lseg(struct pnfs_layout_hdr *lo, struct pnfs_layout_segment *lseg) { struct pnfs_layout_segment *lp; dprintk("%s:Begin\n", __func__); - assert_spin_locked(&lo->plh_inode->i_lock); list_for_each_entry(lp, &lo->plh_segs, pls_list) { if (cmp_layout(&lseg->pls_range, &lp->pls_range) > 0) continue; @@ -813,7 +882,7 @@ pnfs_insert_layout(struct pnfs_layout_hdr *lo, __func__, lseg, lseg->pls_range.iomode, lseg->pls_range.offset, lseg->pls_range.length); out: - get_layout_hdr(lo); + pnfs_get_layout_hdr(lo); dprintk("%s:Return\n", __func__); } @@ -847,21 +916,19 @@ pnfs_find_alloc_layout(struct inode *ino, dprintk("%s Begin ino=%p layout=%p\n", __func__, ino, nfsi->layout); - assert_spin_locked(&ino->i_lock); - if (nfsi->layout) { - if (test_bit(NFS_LAYOUT_DESTROYED, &nfsi->layout->plh_flags)) - return NULL; - else - return nfsi->layout; - } + if (nfsi->layout != NULL) + goto out_existing; spin_unlock(&ino->i_lock); new = alloc_init_layout_hdr(ino, ctx, gfp_flags); spin_lock(&ino->i_lock); - if (likely(nfsi->layout == NULL)) /* Won the race? */ + if (likely(nfsi->layout == NULL)) { /* Won the race? */ nfsi->layout = new; - else - pnfs_free_layout_hdr(new); + return new; + } + pnfs_free_layout_hdr(new); +out_existing: + pnfs_get_layout_hdr(nfsi->layout); return nfsi->layout; } @@ -904,11 +971,10 @@ pnfs_find_lseg(struct pnfs_layout_hdr *lo, dprintk("%s:Begin\n", __func__); - assert_spin_locked(&lo->plh_inode->i_lock); list_for_each_entry(lseg, &lo->plh_segs, pls_list) { if (test_bit(NFS_LSEG_VALID, &lseg->pls_flags) && is_matching_lseg(&lseg->pls_range, range)) { - ret = get_lseg(lseg); + ret = pnfs_get_lseg(lseg); break; } if (lseg->pls_range.offset > range->offset) @@ -1013,7 +1079,6 @@ pnfs_update_layout(struct inode *ino, .length = count, }; unsigned pg_offset; - struct nfs_inode *nfsi = NFS_I(ino); struct nfs_server *server = NFS_SERVER(ino); struct nfs_client *clp = server->nfs_client; struct pnfs_layout_hdr *lo; @@ -1021,16 +1086,16 @@ pnfs_update_layout(struct inode *ino, bool first = false; if (!pnfs_enabled_sb(NFS_SERVER(ino))) - return NULL; + goto out; if (pnfs_within_mdsthreshold(ctx, ino, iomode)) - return NULL; + goto out; spin_lock(&ino->i_lock); lo = pnfs_find_alloc_layout(ino, ctx, gfp_flags); if (lo == NULL) { - dprintk("%s ERROR: can't get pnfs_layout_hdr\n", __func__); - goto out_unlock; + spin_unlock(&ino->i_lock); + goto out; } /* Do we even need to bother with this? */ @@ -1040,7 +1105,7 @@ pnfs_update_layout(struct inode *ino, } /* if LAYOUTGET already failed once we don't try again */ - if (test_bit(lo_fail_bit(iomode), &nfsi->layout->plh_flags)) + if (pnfs_layout_io_test_failed(lo, iomode)) goto out_unlock; /* Check to see if the layout for the given range already exists */ @@ -1048,17 +1113,13 @@ pnfs_update_layout(struct inode *ino, if (lseg) goto out_unlock; - if (pnfs_layoutgets_blocked(lo, NULL, 0)) + if (pnfs_layoutgets_blocked(lo, 0)) goto out_unlock; atomic_inc(&lo->plh_outstanding); - get_layout_hdr(lo); if (list_empty(&lo->plh_segs)) first = true; - /* Enable LAYOUTRETURNs */ - pnfs_clear_layout_returned(lo); - spin_unlock(&ino->i_lock); if (first) { /* The lo must be on the clp list if there is any @@ -1079,24 +1140,26 @@ pnfs_update_layout(struct inode *ino, arg.length = PAGE_CACHE_ALIGN(arg.length); lseg = send_layoutget(lo, ctx, &arg, gfp_flags); - if (!lseg && first) { - spin_lock(&clp->cl_lock); - list_del_init(&lo->plh_layouts); - spin_unlock(&clp->cl_lock); - } atomic_dec(&lo->plh_outstanding); - put_layout_hdr(lo); +out_put_layout_hdr: + pnfs_put_layout_hdr(lo); out: - dprintk("%s end, state 0x%lx lseg %p\n", __func__, - nfsi->layout ? nfsi->layout->plh_flags : -1, lseg); + dprintk("%s: inode %s/%llu pNFS layout segment %s for " + "(%s, offset: %llu, length: %llu)\n", + __func__, ino->i_sb->s_id, + (unsigned long long)NFS_FILEID(ino), + lseg == NULL ? "not found" : "found", + iomode==IOMODE_RW ? "read/write" : "read-only", + (unsigned long long)pos, + (unsigned long long)count); return lseg; out_unlock: spin_unlock(&ino->i_lock); - goto out; + goto out_put_layout_hdr; } EXPORT_SYMBOL_GPL(pnfs_update_layout); -int +struct pnfs_layout_segment * pnfs_layout_process(struct nfs4_layoutget *lgp) { struct pnfs_layout_hdr *lo = NFS_I(lgp->args.inode)->layout; @@ -1123,25 +1186,29 @@ pnfs_layout_process(struct nfs4_layoutget *lgp) goto out_forget_reply; } - if (pnfs_layoutgets_blocked(lo, &res->stateid, 1)) { + if (pnfs_layoutgets_blocked(lo, 1) || + pnfs_layout_stateid_blocked(lo, &res->stateid)) { dprintk("%s forget reply due to state\n", __func__); goto out_forget_reply; } + + /* Done processing layoutget. Set the layout stateid */ + pnfs_set_layout_stateid(lo, &res->stateid, false); + init_lseg(lo, lseg); lseg->pls_range = res->range; - *lgp->lsegpp = get_lseg(lseg); - pnfs_insert_layout(lo, lseg); + pnfs_get_lseg(lseg); + pnfs_layout_insert_lseg(lo, lseg); if (res->return_on_close) { set_bit(NFS_LSEG_ROC, &lseg->pls_flags); set_bit(NFS_LAYOUT_ROC, &lo->plh_flags); } - /* Done processing layoutget. Set the layout stateid */ - pnfs_set_layout_stateid(lo, &res->stateid, false); spin_unlock(&ino->i_lock); + return lseg; out: - return status; + return ERR_PTR(status); out_forget_reply: spin_unlock(&ino->i_lock); @@ -1153,16 +1220,24 @@ out_forget_reply: void pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { + u64 rd_size = req->wb_bytes; + BUG_ON(pgio->pg_lseg != NULL); if (req->wb_offset != req->wb_pgbase) { nfs_pageio_reset_read_mds(pgio); return; } + + if (pgio->pg_dreq == NULL) + rd_size = i_size_read(pgio->pg_inode) - req_offset(req); + else + rd_size = nfs_dreq_bytes_left(pgio->pg_dreq); + pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, req->wb_context, req_offset(req), - req->wb_bytes, + rd_size, IOMODE_READ, GFP_KERNEL); /* If no lseg, fall back to read through mds */ @@ -1173,7 +1248,8 @@ pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *r EXPORT_SYMBOL_GPL(pnfs_generic_pg_init_read); void -pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) +pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req, u64 wb_size) { BUG_ON(pgio->pg_lseg != NULL); @@ -1181,10 +1257,11 @@ pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page * nfs_pageio_reset_write_mds(pgio); return; } + pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, req->wb_context, req_offset(req), - req->wb_bytes, + wb_size, IOMODE_RW, GFP_NOFS); /* If no lseg, fall back to write through mds */ @@ -1362,12 +1439,12 @@ pnfs_do_multiple_writes(struct nfs_pageio_descriptor *desc, struct list_head *he if (trypnfs == PNFS_NOT_ATTEMPTED) pnfs_write_through_mds(desc, data); } - put_lseg(lseg); + pnfs_put_lseg(lseg); } static void pnfs_writehdr_free(struct nfs_pgio_header *hdr) { - put_lseg(hdr->lseg); + pnfs_put_lseg(hdr->lseg); nfs_writehdr_free(hdr); } EXPORT_SYMBOL_GPL(pnfs_writehdr_free); @@ -1382,17 +1459,17 @@ pnfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc) whdr = nfs_writehdr_alloc(); if (!whdr) { desc->pg_completion_ops->error_cleanup(&desc->pg_list); - put_lseg(desc->pg_lseg); + pnfs_put_lseg(desc->pg_lseg); desc->pg_lseg = NULL; return -ENOMEM; } hdr = &whdr->header; nfs_pgheader_init(desc, hdr, pnfs_writehdr_free); - hdr->lseg = get_lseg(desc->pg_lseg); + hdr->lseg = pnfs_get_lseg(desc->pg_lseg); atomic_inc(&hdr->refcnt); ret = nfs_generic_flush(desc, hdr); if (ret != 0) { - put_lseg(desc->pg_lseg); + pnfs_put_lseg(desc->pg_lseg); desc->pg_lseg = NULL; } else pnfs_do_multiple_writes(desc, &hdr->rpc_list, desc->pg_ioflags); @@ -1517,12 +1594,12 @@ pnfs_do_multiple_reads(struct nfs_pageio_descriptor *desc, struct list_head *hea if (trypnfs == PNFS_NOT_ATTEMPTED) pnfs_read_through_mds(desc, data); } - put_lseg(lseg); + pnfs_put_lseg(lseg); } static void pnfs_readhdr_free(struct nfs_pgio_header *hdr) { - put_lseg(hdr->lseg); + pnfs_put_lseg(hdr->lseg); nfs_readhdr_free(hdr); } EXPORT_SYMBOL_GPL(pnfs_readhdr_free); @@ -1538,17 +1615,17 @@ pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc) if (!rhdr) { desc->pg_completion_ops->error_cleanup(&desc->pg_list); ret = -ENOMEM; - put_lseg(desc->pg_lseg); + pnfs_put_lseg(desc->pg_lseg); desc->pg_lseg = NULL; return ret; } hdr = &rhdr->header; nfs_pgheader_init(desc, hdr, pnfs_readhdr_free); - hdr->lseg = get_lseg(desc->pg_lseg); + hdr->lseg = pnfs_get_lseg(desc->pg_lseg); atomic_inc(&hdr->refcnt); ret = nfs_generic_pagein(desc, hdr); if (ret != 0) { - put_lseg(desc->pg_lseg); + pnfs_put_lseg(desc->pg_lseg); desc->pg_lseg = NULL; } else pnfs_do_multiple_reads(desc, &hdr->rpc_list); @@ -1574,13 +1651,7 @@ static void pnfs_list_write_lseg(struct inode *inode, struct list_head *listp) void pnfs_set_lo_fail(struct pnfs_layout_segment *lseg) { - if (lseg->pls_range.iomode == IOMODE_RW) { - dprintk("%s Setting layout IOMODE_RW fail bit\n", __func__); - set_bit(lo_fail_bit(IOMODE_RW), &lseg->pls_layout->plh_flags); - } else { - dprintk("%s Setting layout IOMODE_READ fail bit\n", __func__); - set_bit(lo_fail_bit(IOMODE_READ), &lseg->pls_layout->plh_flags); - } + pnfs_layout_io_set_failed(lseg->pls_layout, lseg->pls_range.iomode); } EXPORT_SYMBOL_GPL(pnfs_set_lo_fail); @@ -1601,7 +1672,7 @@ pnfs_set_layoutcommit(struct nfs_write_data *wdata) } if (!test_and_set_bit(NFS_LSEG_LAYOUTCOMMIT, &hdr->lseg->pls_flags)) { /* references matched in nfs4_layoutcommit_release */ - get_lseg(hdr->lseg); + pnfs_get_lseg(hdr->lseg); } if (end_pos > nfsi->layout->plh_lwb) nfsi->layout->plh_lwb = end_pos; diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index 745aa1b39e7c..dbf7bba52da0 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -62,9 +62,7 @@ enum { NFS_LAYOUT_RW_FAILED, /* get rw layout failed stop trying */ NFS_LAYOUT_BULK_RECALL, /* bulk recall affecting layout */ NFS_LAYOUT_ROC, /* some lseg had roc bit set */ - NFS_LAYOUT_DESTROYED, /* no new use of layout allowed */ - NFS_LAYOUT_INVALID, /* layout is being destroyed */ - NFS_LAYOUT_RETURNED, /* layout has already been returned */ + NFS_LAYOUT_RETURN, /* Return this layout ASAP */ }; enum layoutdriver_policy_flags { @@ -140,6 +138,7 @@ struct pnfs_layout_hdr { atomic_t plh_outstanding; /* number of RPCs out */ unsigned long plh_block_lgets; /* block LAYOUTGET if >0 */ u32 plh_barrier; /* ignore lower seqids */ + unsigned long plh_retry_timestamp; unsigned long plh_flags; loff_t plh_lwb; /* last write byte for layoutcommit */ struct rpc_cred *plh_lc_cred; /* layoutcommit cred */ @@ -172,12 +171,12 @@ extern int nfs4_proc_getdevicelist(struct nfs_server *server, struct pnfs_devicelist *devlist); extern int nfs4_proc_getdeviceinfo(struct nfs_server *server, struct pnfs_device *dev); -extern void nfs4_proc_layoutget(struct nfs4_layoutget *lgp, gfp_t gfp_flags); +extern struct pnfs_layout_segment* nfs4_proc_layoutget(struct nfs4_layoutget *lgp, gfp_t gfp_flags); extern int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp); /* pnfs.c */ -void get_layout_hdr(struct pnfs_layout_hdr *lo); -void put_lseg(struct pnfs_layout_segment *lseg); +void pnfs_get_layout_hdr(struct pnfs_layout_hdr *lo); +void pnfs_put_lseg(struct pnfs_layout_segment *lseg); void pnfs_pageio_init_read(struct nfs_pageio_descriptor *, struct inode *, const struct nfs_pgio_completion_ops *); @@ -188,28 +187,29 @@ void set_pnfs_layoutdriver(struct nfs_server *, const struct nfs_fh *, u32); void unset_pnfs_layoutdriver(struct nfs_server *); void pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *, struct nfs_page *); int pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc); -void pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *, struct nfs_page *); +void pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, + struct nfs_page *req, u64 wb_size); int pnfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc); bool pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, struct nfs_page *req); void pnfs_set_lo_fail(struct pnfs_layout_segment *lseg); -int pnfs_layout_process(struct nfs4_layoutget *lgp); +struct pnfs_layout_segment *pnfs_layout_process(struct nfs4_layoutget *lgp); void pnfs_free_lseg_list(struct list_head *tmp_list); void pnfs_destroy_layout(struct nfs_inode *); void pnfs_destroy_all_layouts(struct nfs_client *); -void put_layout_hdr(struct pnfs_layout_hdr *lo); +void pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo); void pnfs_set_layout_stateid(struct pnfs_layout_hdr *lo, const nfs4_stateid *new, bool update_barrier); int pnfs_choose_layoutget_stateid(nfs4_stateid *dst, struct pnfs_layout_hdr *lo, struct nfs4_state *open_state); -int mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo, +int pnfs_mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo, struct list_head *tmp_list, struct pnfs_layout_range *recall_range); bool pnfs_roc(struct inode *ino); void pnfs_roc_release(struct inode *ino); void pnfs_roc_set_barrier(struct inode *ino, u32 barrier); -bool pnfs_roc_drain(struct inode *ino, u32 *barrier); +bool pnfs_roc_drain(struct inode *ino, u32 *barrier, struct rpc_task *task); void pnfs_set_layoutcommit(struct nfs_write_data *wdata); void pnfs_cleanup_layoutcommit(struct nfs4_layoutcommit_data *data); int pnfs_layoutcommit_inode(struct inode *inode, bool sync); @@ -233,6 +233,7 @@ struct nfs4_threshold *pnfs_mdsthreshold_alloc(void); /* nfs4_deviceid_flags */ enum { NFS_DEVICEID_INVALID = 0, /* set when MDS clientid recalled */ + NFS_DEVICEID_UNAVAILABLE, /* device temporarily unavailable */ }; /* pnfs_dev.c */ @@ -242,6 +243,7 @@ struct nfs4_deviceid_node { const struct pnfs_layoutdriver_type *ld; const struct nfs_client *nfs_client; unsigned long flags; + unsigned long timestamp_unavailable; struct nfs4_deviceid deviceid; atomic_t ref; }; @@ -254,34 +256,12 @@ void nfs4_init_deviceid_node(struct nfs4_deviceid_node *, const struct nfs4_deviceid *); struct nfs4_deviceid_node *nfs4_insert_deviceid_node(struct nfs4_deviceid_node *); bool nfs4_put_deviceid_node(struct nfs4_deviceid_node *); +void nfs4_mark_deviceid_unavailable(struct nfs4_deviceid_node *node); +bool nfs4_test_deviceid_unavailable(struct nfs4_deviceid_node *node); void nfs4_deviceid_purge_client(const struct nfs_client *); -static inline void -pnfs_mark_layout_returned(struct pnfs_layout_hdr *lo) -{ - set_bit(NFS_LAYOUT_RETURNED, &lo->plh_flags); -} - -static inline void -pnfs_clear_layout_returned(struct pnfs_layout_hdr *lo) -{ - clear_bit(NFS_LAYOUT_RETURNED, &lo->plh_flags); -} - -static inline bool -pnfs_test_layout_returned(struct pnfs_layout_hdr *lo) -{ - return test_bit(NFS_LAYOUT_RETURNED, &lo->plh_flags); -} - -static inline int lo_fail_bit(u32 iomode) -{ - return iomode == IOMODE_RW ? - NFS_LAYOUT_RW_FAILED : NFS_LAYOUT_RO_FAILED; -} - static inline struct pnfs_layout_segment * -get_lseg(struct pnfs_layout_segment *lseg) +pnfs_get_lseg(struct pnfs_layout_segment *lseg) { if (lseg) { atomic_inc(&lseg->pls_refcount); @@ -406,12 +386,12 @@ static inline void pnfs_destroy_layout(struct nfs_inode *nfsi) } static inline struct pnfs_layout_segment * -get_lseg(struct pnfs_layout_segment *lseg) +pnfs_get_lseg(struct pnfs_layout_segment *lseg) { return NULL; } -static inline void put_lseg(struct pnfs_layout_segment *lseg) +static inline void pnfs_put_lseg(struct pnfs_layout_segment *lseg) { } @@ -443,7 +423,7 @@ pnfs_roc_set_barrier(struct inode *ino, u32 barrier) } static inline bool -pnfs_roc_drain(struct inode *ino, u32 *barrier) +pnfs_roc_drain(struct inode *ino, u32 *barrier, struct rpc_task *task) { return false; } diff --git a/fs/nfs/pnfs_dev.c b/fs/nfs/pnfs_dev.c index 73f701f1f4d3..d35b62e83ea6 100644 --- a/fs/nfs/pnfs_dev.c +++ b/fs/nfs/pnfs_dev.c @@ -40,6 +40,8 @@ #define NFS4_DEVICE_ID_HASH_SIZE (1 << NFS4_DEVICE_ID_HASH_BITS) #define NFS4_DEVICE_ID_HASH_MASK (NFS4_DEVICE_ID_HASH_SIZE - 1) +#define PNFS_DEVICE_RETRY_TIMEOUT (120*HZ) + static struct hlist_head nfs4_deviceid_cache[NFS4_DEVICE_ID_HASH_SIZE]; static DEFINE_SPINLOCK(nfs4_deviceid_lock); @@ -218,6 +220,30 @@ nfs4_put_deviceid_node(struct nfs4_deviceid_node *d) } EXPORT_SYMBOL_GPL(nfs4_put_deviceid_node); +void +nfs4_mark_deviceid_unavailable(struct nfs4_deviceid_node *node) +{ + node->timestamp_unavailable = jiffies; + set_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags); +} +EXPORT_SYMBOL_GPL(nfs4_mark_deviceid_unavailable); + +bool +nfs4_test_deviceid_unavailable(struct nfs4_deviceid_node *node) +{ + if (test_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags)) { + unsigned long start, end; + + end = jiffies; + start = end - PNFS_DEVICE_RETRY_TIMEOUT; + if (time_in_range(node->timestamp_unavailable, start, end)) + return true; + clear_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags); + } + return false; +} +EXPORT_SYMBOL_GPL(nfs4_test_deviceid_unavailable); + static void _deviceid_purge_client(const struct nfs_client *clp, long hash) { @@ -276,3 +302,4 @@ nfs4_deviceid_mark_client_invalid(struct nfs_client *clp) } rcu_read_unlock(); } + diff --git a/fs/nfs/super.c b/fs/nfs/super.c index d2c7f5db0847..e831bce49766 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -88,6 +88,7 @@ enum { Opt_sharecache, Opt_nosharecache, Opt_resvport, Opt_noresvport, Opt_fscache, Opt_nofscache, + Opt_migration, Opt_nomigration, /* Mount options that take integer arguments */ Opt_port, @@ -147,6 +148,8 @@ static const match_table_t nfs_mount_option_tokens = { { Opt_noresvport, "noresvport" }, { Opt_fscache, "fsc" }, { Opt_nofscache, "nofsc" }, + { Opt_migration, "migration" }, + { Opt_nomigration, "nomigration" }, { Opt_port, "port=%s" }, { Opt_rsize, "rsize=%s" }, @@ -676,6 +679,9 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, if (nfss->options & NFS_OPTION_FSCACHE) seq_printf(m, ",fsc"); + if (nfss->options & NFS_OPTION_MIGRATION) + seq_printf(m, ",migration"); + if (nfss->flags & NFS_MOUNT_LOOKUP_CACHE_NONEG) { if (nfss->flags & NFS_MOUNT_LOOKUP_CACHE_NONE) seq_printf(m, ",lookupcache=none"); @@ -1106,7 +1112,7 @@ static int nfs_get_option_ul(substring_t args[], unsigned long *option) string = match_strdup(args); if (string == NULL) return -ENOMEM; - rc = strict_strtoul(string, 10, option); + rc = kstrtoul(string, 10, option); kfree(string); return rc; @@ -1243,6 +1249,12 @@ static int nfs_parse_mount_options(char *raw, kfree(mnt->fscache_uniq); mnt->fscache_uniq = NULL; break; + case Opt_migration: + mnt->options |= NFS_OPTION_MIGRATION; + break; + case Opt_nomigration: + mnt->options &= NFS_OPTION_MIGRATION; + break; /* * options that take numeric values @@ -1535,6 +1547,10 @@ static int nfs_parse_mount_options(char *raw, if (mnt->minorversion && mnt->version != 4) goto out_minorversion_mismatch; + if (mnt->options & NFS_OPTION_MIGRATION && + mnt->version != 4 && mnt->minorversion != 0) + goto out_migration_misuse; + /* * verify that any proto=/mountproto= options match the address * families in the addr=/mountaddr= options. @@ -1572,6 +1588,10 @@ out_minorversion_mismatch: printk(KERN_INFO "NFS: mount option vers=%u does not support " "minorversion=%u\n", mnt->version, mnt->minorversion); return 0; +out_migration_misuse: + printk(KERN_INFO + "NFS: 'migration' not supported for this NFS version\n"); + return 0; out_nomem: printk(KERN_INFO "NFS: not enough memory to parse option\n"); return 0; @@ -2494,7 +2514,7 @@ EXPORT_SYMBOL_GPL(nfs_kill_super); /* * Clone an NFS2/3/4 server record on xdev traversal (FSID-change) */ -struct dentry * +static struct dentry * nfs_xdev_mount(struct file_system_type *fs_type, int flags, const char *dev_name, void *raw_data) { @@ -2642,6 +2662,7 @@ unsigned int nfs_idmap_cache_timeout = 600; bool nfs4_disable_idmapping = true; unsigned short max_session_slots = NFS4_DEF_SLOT_TABLE_SIZE; unsigned short send_implementation_id = 1; +char nfs4_client_id_uniquifier[NFS4_CLIENT_ID_UNIQ_LEN] = ""; EXPORT_SYMBOL_GPL(nfs_callback_set_tcpport); EXPORT_SYMBOL_GPL(nfs_callback_tcpport); @@ -2649,6 +2670,7 @@ EXPORT_SYMBOL_GPL(nfs_idmap_cache_timeout); EXPORT_SYMBOL_GPL(nfs4_disable_idmapping); EXPORT_SYMBOL_GPL(max_session_slots); EXPORT_SYMBOL_GPL(send_implementation_id); +EXPORT_SYMBOL_GPL(nfs4_client_id_uniquifier); #define NFS_CALLBACK_MAXPORTNR (65535U) @@ -2659,7 +2681,7 @@ static int param_set_portnr(const char *val, const struct kernel_param *kp) if (!val) return -EINVAL; - ret = strict_strtoul(val, 0, &num); + ret = kstrtoul(val, 0, &num); if (ret == -EINVAL || num > NFS_CALLBACK_MAXPORTNR) return -EINVAL; *((unsigned int *)kp->arg) = num; @@ -2674,6 +2696,8 @@ static struct kernel_param_ops param_ops_portnr = { module_param_named(callback_tcpport, nfs_callback_set_tcpport, portnr, 0644); module_param(nfs_idmap_cache_timeout, int, 0644); module_param(nfs4_disable_idmapping, bool, 0644); +module_param_string(nfs4_unique_id, nfs4_client_id_uniquifier, + NFS4_CLIENT_ID_UNIQ_LEN, 0600); MODULE_PARM_DESC(nfs4_disable_idmapping, "Turn off NFSv4 idmapping when using 'sec=sys'"); module_param(max_session_slots, ushort, 0644); @@ -2682,6 +2706,7 @@ MODULE_PARM_DESC(max_session_slots, "Maximum number of outstanding NFSv4.1 " module_param(send_implementation_id, ushort, 0644); MODULE_PARM_DESC(send_implementation_id, "Send implementation ID with NFSv4.1 exchange_id"); +MODULE_PARM_DESC(nfs4_unique_id, "nfs_client_id4 uniquifier string"); MODULE_ALIAS("nfs4"); #endif /* CONFIG_NFS_V4 */ diff --git a/fs/nfs/write.c b/fs/nfs/write.c index e3b55372726c..9347ab7c9574 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -846,6 +846,7 @@ static int nfs_writepage_setup(struct nfs_open_context *ctx, struct page *page, int nfs_flush_incompatible(struct file *file, struct page *page) { struct nfs_open_context *ctx = nfs_file_open_context(file); + struct nfs_lock_context *l_ctx; struct nfs_page *req; int do_flush, status; /* @@ -860,9 +861,12 @@ int nfs_flush_incompatible(struct file *file, struct page *page) req = nfs_page_find_request(page); if (req == NULL) return 0; - do_flush = req->wb_page != page || req->wb_context != ctx || - req->wb_lock_context->lockowner != current->files || - req->wb_lock_context->pid != current->tgid; + l_ctx = req->wb_lock_context; + do_flush = req->wb_page != page || req->wb_context != ctx; + if (l_ctx) { + do_flush |= l_ctx->lockowner.l_owner != current->files + || l_ctx->lockowner.l_pid != current->tgid; + } nfs_release_request(req); if (!do_flush) return 0; @@ -1576,6 +1580,7 @@ static void nfs_commit_release_pages(struct nfs_commit_data *data) /* We have a mismatch. Write the page again */ dprintk(" mismatch\n"); nfs_mark_request_dirty(req); + set_bit(NFS_CONTEXT_RESEND_WRITES, &req->wb_context->flags); next: nfs_unlock_and_release_request(req); } diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c index 6aa5590c3679..b314888825d5 100644 --- a/fs/nfsd/nfs2acl.c +++ b/fs/nfsd/nfs2acl.c @@ -218,8 +218,7 @@ static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, __be32 *p, * There must be an encoding function for void results so svc_process * will work properly. */ -int -nfsaclsvc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy) +static int nfsaclsvc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index 9095f3c21df9..97d90d1c8608 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -247,7 +247,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, /* Now create the file and set attributes */ nfserr = do_nfsd_create(rqstp, dirfhp, argp->name, argp->len, attr, newfhp, - argp->createmode, argp->verf, NULL, NULL); + argp->createmode, (u32 *)argp->verf, NULL, NULL); RETURN_STATUS(nfserr); } diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 4c7bd35b1876..bdf29c96e4cd 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -1028,7 +1028,6 @@ void nfsd4_cb_recall(struct nfs4_delegation *dp) cb->cb_msg.rpc_cred = callback_cred; cb->cb_ops = &nfsd4_cb_recall_ops; - dp->dl_retries = 1; INIT_LIST_HEAD(&cb->cb_per_client); cb->cb_done = true; diff --git a/fs/nfsd/nfs4idmap.c b/fs/nfsd/nfs4idmap.c index fdc91a6fc9c4..a1f10c0a6255 100644 --- a/fs/nfsd/nfs4idmap.c +++ b/fs/nfsd/nfs4idmap.c @@ -478,7 +478,7 @@ nfsd_idmap_init(struct net *net) goto destroy_idtoname_cache; nn->nametoid_cache = cache_create_net(&nametoid_cache_template, net); if (IS_ERR(nn->nametoid_cache)) { - rv = PTR_ERR(nn->idtoname_cache); + rv = PTR_ERR(nn->nametoid_cache); goto unregister_idtoname_cache; } rv = cache_register_net(nn->nametoid_cache, net); @@ -598,7 +598,7 @@ numeric_name_to_id(struct svc_rqst *rqstp, int type, const char *name, u32 namel /* Just to make sure it's null-terminated: */ memcpy(buf, name, namelen); buf[namelen] = '\0'; - ret = kstrtouint(name, 10, id); + ret = kstrtouint(buf, 10, id); return ret == 0; } diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index c9c1c0a25417..6c9a4b291dba 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -370,7 +370,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, break; case NFS4_OPEN_CLAIM_PREVIOUS: open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED; - status = nfs4_check_open_reclaim(&open->op_clientid); + status = nfs4_check_open_reclaim(&open->op_clientid, cstate->minorversion); if (status) goto out; case NFS4_OPEN_CLAIM_FH: @@ -1054,8 +1054,8 @@ struct nfsd4_operation { char *op_name; /* Try to get response size before operation */ nfsd4op_rsize op_rsize_bop; - stateid_setter op_get_currentstateid; - stateid_getter op_set_currentstateid; + stateid_getter op_get_currentstateid; + stateid_setter op_set_currentstateid; }; static struct nfsd4_operation nfsd4_ops[]; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 48a1bad37334..d0237f872cc4 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -758,7 +758,7 @@ static void nfsd4_put_drc_mem(int slotsize, int num) spin_unlock(&nfsd_drc_lock); } -static struct nfsd4_session *alloc_session(int slotsize, int numslots) +static struct nfsd4_session *__alloc_session(int slotsize, int numslots) { struct nfsd4_session *new; int mem, i; @@ -852,35 +852,28 @@ static int nfsd4_register_conn(struct nfsd4_conn *conn) return register_xpt_user(conn->cn_xprt, &conn->cn_xpt_user); } -static __be32 nfsd4_new_conn(struct svc_rqst *rqstp, struct nfsd4_session *ses, u32 dir) +static void nfsd4_init_conn(struct svc_rqst *rqstp, struct nfsd4_conn *conn, struct nfsd4_session *ses) { - struct nfsd4_conn *conn; int ret; - conn = alloc_conn(rqstp, dir); - if (!conn) - return nfserr_jukebox; nfsd4_hash_conn(conn, ses); ret = nfsd4_register_conn(conn); if (ret) /* oops; xprt is already down: */ nfsd4_conn_lost(&conn->cn_xpt_user); - if (ses->se_client->cl_cb_state == NFSD4_CB_DOWN && - dir & NFS4_CDFC4_BACK) { + if (conn->cn_flags & NFS4_CDFC4_BACK) { /* callback channel may be back up */ nfsd4_probe_callback(ses->se_client); } - return nfs_ok; } -static __be32 nfsd4_new_conn_from_crses(struct svc_rqst *rqstp, struct nfsd4_session *ses) +static struct nfsd4_conn *alloc_conn_from_crses(struct svc_rqst *rqstp, struct nfsd4_create_session *cses) { u32 dir = NFS4_CDFC4_FORE; - if (ses->se_flags & SESSION4_BACK_CHAN) + if (cses->flags & SESSION4_BACK_CHAN) dir |= NFS4_CDFC4_BACK; - - return nfsd4_new_conn(rqstp, ses, dir); + return alloc_conn(rqstp, dir); } /* must be called under client_lock */ @@ -903,20 +896,21 @@ static void nfsd4_del_conns(struct nfsd4_session *s) spin_unlock(&clp->cl_lock); } +static void __free_session(struct nfsd4_session *ses) +{ + nfsd4_put_drc_mem(slot_bytes(&ses->se_fchannel), ses->se_fchannel.maxreqs); + free_session_slots(ses); + kfree(ses); +} + static void free_session(struct kref *kref) { struct nfsd4_session *ses; - int mem; lockdep_assert_held(&client_lock); ses = container_of(kref, struct nfsd4_session, se_ref); nfsd4_del_conns(ses); - spin_lock(&nfsd_drc_lock); - mem = ses->se_fchannel.maxreqs * slot_bytes(&ses->se_fchannel); - nfsd_drc_mem_used -= mem; - spin_unlock(&nfsd_drc_lock); - free_session_slots(ses); - kfree(ses); + __free_session(ses); } void nfsd4_put_session(struct nfsd4_session *ses) @@ -926,14 +920,10 @@ void nfsd4_put_session(struct nfsd4_session *ses) spin_unlock(&client_lock); } -static struct nfsd4_session *alloc_init_session(struct svc_rqst *rqstp, struct nfs4_client *clp, struct nfsd4_create_session *cses) +static struct nfsd4_session *alloc_session(struct nfsd4_channel_attrs *fchan) { struct nfsd4_session *new; - struct nfsd4_channel_attrs *fchan = &cses->fore_channel; int numslots, slotsize; - __be32 status; - int idx; - /* * Note decreasing slot size below client's request may * make it difficult for client to function correctly, whereas @@ -946,12 +936,18 @@ static struct nfsd4_session *alloc_init_session(struct svc_rqst *rqstp, struct n if (numslots < 1) return NULL; - new = alloc_session(slotsize, numslots); + new = __alloc_session(slotsize, numslots); if (!new) { nfsd4_put_drc_mem(slotsize, fchan->maxreqs); return NULL; } init_forechannel_attrs(&new->se_fchannel, fchan, numslots, slotsize); + return new; +} + +static struct nfsd4_session *init_session(struct svc_rqst *rqstp, struct nfsd4_session *new, struct nfs4_client *clp, struct nfsd4_create_session *cses) +{ + int idx; new->se_client = clp; gen_sessionid(new); @@ -970,14 +966,6 @@ static struct nfsd4_session *alloc_init_session(struct svc_rqst *rqstp, struct n spin_unlock(&clp->cl_lock); spin_unlock(&client_lock); - status = nfsd4_new_conn_from_crses(rqstp, new); - /* whoops: benny points out, status is ignored! (err, or bogus) */ - if (status) { - spin_lock(&client_lock); - free_session(&new->se_ref); - spin_unlock(&client_lock); - return NULL; - } if (cses->flags & SESSION4_BACK_CHAN) { struct sockaddr *sa = svc_addr(rqstp); /* @@ -990,7 +978,6 @@ static struct nfsd4_session *alloc_init_session(struct svc_rqst *rqstp, struct n rpc_copy_addr((struct sockaddr *)&clp->cl_cb_conn.cb_addr, sa); clp->cl_cb_conn.cb_addrlen = svc_addr_len(sa); } - nfsd4_probe_callback(clp); return new; } @@ -1131,7 +1118,7 @@ unhash_client_locked(struct nfs4_client *clp) } static void -expire_client(struct nfs4_client *clp) +destroy_client(struct nfs4_client *clp) { struct nfs4_openowner *oo; struct nfs4_delegation *dp; @@ -1165,6 +1152,12 @@ expire_client(struct nfs4_client *clp) spin_unlock(&client_lock); } +static void expire_client(struct nfs4_client *clp) +{ + nfsd4_client_record_remove(clp); + destroy_client(clp); +} + static void copy_verf(struct nfs4_client *target, nfs4_verifier *source) { memcpy(target->cl_verifier.data, source->data, @@ -1223,10 +1216,26 @@ static bool groups_equal(struct group_info *g1, struct group_info *g2) return true; } +/* + * RFC 3530 language requires clid_inuse be returned when the + * "principal" associated with a requests differs from that previously + * used. We use uid, gid's, and gss principal string as our best + * approximation. We also don't want to allow non-gss use of a client + * established using gss: in theory cr_principal should catch that + * change, but in practice cr_principal can be null even in the gss case + * since gssd doesn't always pass down a principal string. + */ +static bool is_gss_cred(struct svc_cred *cr) +{ + /* Is cr_flavor one of the gss "pseudoflavors"?: */ + return (cr->cr_flavor > RPC_AUTH_MAXFLAVOR); +} + + static bool same_creds(struct svc_cred *cr1, struct svc_cred *cr2) { - if ((cr1->cr_flavor != cr2->cr_flavor) + if ((is_gss_cred(cr1) != is_gss_cred(cr2)) || (cr1->cr_uid != cr2->cr_uid) || (cr1->cr_gid != cr2->cr_gid) || !groups_equal(cr1->cr_group_info, cr2->cr_group_info)) @@ -1340,13 +1349,15 @@ move_to_confirmed(struct nfs4_client *clp) } static struct nfs4_client * -find_confirmed_client(clientid_t *clid) +find_confirmed_client(clientid_t *clid, bool sessions) { struct nfs4_client *clp; unsigned int idhashval = clientid_hashval(clid->cl_id); list_for_each_entry(clp, &conf_id_hashtbl[idhashval], cl_idhash) { if (same_clid(&clp->cl_clientid, clid)) { + if ((bool)clp->cl_minorversion != sessions) + return NULL; renew_client(clp); return clp; } @@ -1355,14 +1366,17 @@ find_confirmed_client(clientid_t *clid) } static struct nfs4_client * -find_unconfirmed_client(clientid_t *clid) +find_unconfirmed_client(clientid_t *clid, bool sessions) { struct nfs4_client *clp; unsigned int idhashval = clientid_hashval(clid->cl_id); list_for_each_entry(clp, &unconf_id_hashtbl[idhashval], cl_idhash) { - if (same_clid(&clp->cl_clientid, clid)) + if (same_clid(&clp->cl_clientid, clid)) { + if ((bool)clp->cl_minorversion != sessions) + return NULL; return clp; + } } return NULL; } @@ -1651,6 +1665,7 @@ out_new: status = nfserr_jukebox; goto out; } + new->cl_minorversion = 1; gen_clid(new); add_to_unconfirmed(new, strhashval); @@ -1743,67 +1758,71 @@ nfsd4_create_session(struct svc_rqst *rqstp, struct sockaddr *sa = svc_addr(rqstp); struct nfs4_client *conf, *unconf; struct nfsd4_session *new; + struct nfsd4_conn *conn; struct nfsd4_clid_slot *cs_slot = NULL; - bool confirm_me = false; __be32 status = 0; if (cr_ses->flags & ~SESSION4_FLAG_MASK_A) return nfserr_inval; + if (check_forechannel_attrs(cr_ses->fore_channel)) + return nfserr_toosmall; + new = alloc_session(&cr_ses->fore_channel); + if (!new) + return nfserr_jukebox; + status = nfserr_jukebox; + conn = alloc_conn_from_crses(rqstp, cr_ses); + if (!conn) + goto out_free_session; nfs4_lock_state(); - unconf = find_unconfirmed_client(&cr_ses->clientid); - conf = find_confirmed_client(&cr_ses->clientid); + unconf = find_unconfirmed_client(&cr_ses->clientid, true); + conf = find_confirmed_client(&cr_ses->clientid, true); if (conf) { cs_slot = &conf->cl_cs_slot; status = check_slot_seqid(cr_ses->seqid, cs_slot->sl_seqid, 0); if (status == nfserr_replay_cache) { status = nfsd4_replay_create_session(cr_ses, cs_slot); - goto out; + goto out_free_conn; } else if (cr_ses->seqid != cs_slot->sl_seqid + 1) { status = nfserr_seq_misordered; - goto out; + goto out_free_conn; } } else if (unconf) { + unsigned int hash; + struct nfs4_client *old; if (!same_creds(&unconf->cl_cred, &rqstp->rq_cred) || !rpc_cmp_addr(sa, (struct sockaddr *) &unconf->cl_addr)) { status = nfserr_clid_inuse; - goto out; + goto out_free_conn; } cs_slot = &unconf->cl_cs_slot; status = check_slot_seqid(cr_ses->seqid, cs_slot->sl_seqid, 0); if (status) { /* an unconfirmed replay returns misordered */ status = nfserr_seq_misordered; - goto out; + goto out_free_conn; } - confirm_me = true; + hash = clientstr_hashval(unconf->cl_recdir); + old = find_confirmed_client_by_str(unconf->cl_recdir, hash); + if (old) + expire_client(old); + move_to_confirmed(unconf); conf = unconf; } else { status = nfserr_stale_clientid; - goto out; + goto out_free_conn; } - - /* - * XXX: we should probably set this at creation time, and check - * for consistent minorversion use throughout: - */ - conf->cl_minorversion = 1; + status = nfs_ok; /* * We do not support RDMA or persistent sessions */ cr_ses->flags &= ~SESSION4_PERSIST; cr_ses->flags &= ~SESSION4_RDMA; - status = nfserr_toosmall; - if (check_forechannel_attrs(cr_ses->fore_channel)) - goto out; + init_session(rqstp, new, conf, cr_ses); + nfsd4_init_conn(rqstp, conn, new); - status = nfserr_jukebox; - new = alloc_init_session(rqstp, conf, cr_ses); - if (!new) - goto out; - status = nfs_ok; memcpy(cr_ses->sessionid.data, new->se_sessionid.data, NFS4_MAX_SESSIONID_LEN); memcpy(&cr_ses->fore_channel, &new->se_fchannel, @@ -1813,18 +1832,15 @@ nfsd4_create_session(struct svc_rqst *rqstp, /* cache solo and embedded create sessions under the state lock */ nfsd4_cache_create_session(cr_ses, cs_slot, status); - if (confirm_me) { - unsigned int hash = clientstr_hashval(unconf->cl_recdir); - struct nfs4_client *old = - find_confirmed_client_by_str(conf->cl_recdir, hash); - if (old) - expire_client(old); - move_to_confirmed(conf); - } out: nfs4_unlock_state(); dprintk("%s returns %d\n", __func__, ntohl(status)); return status; +out_free_conn: + free_conn(conn); +out_free_session: + __free_session(new); + goto out; } static bool nfsd4_last_compound_op(struct svc_rqst *rqstp) @@ -1854,6 +1870,7 @@ __be32 nfsd4_bind_conn_to_session(struct svc_rqst *rqstp, struct nfsd4_bind_conn_to_session *bcts) { __be32 status; + struct nfsd4_conn *conn; if (!nfsd4_last_compound_op(rqstp)) return nfserr_not_only_op; @@ -1870,9 +1887,13 @@ __be32 nfsd4_bind_conn_to_session(struct svc_rqst *rqstp, return nfserr_badsession; status = nfsd4_map_bcts_dir(&bcts->dir); - if (!status) - nfsd4_new_conn(rqstp, cstate->session, bcts->dir); - return status; + if (status) + return status; + conn = alloc_conn(rqstp, bcts->dir); + if (!conn) + return nfserr_jukebox; + nfsd4_init_conn(rqstp, conn, cstate->session); + return nfs_ok; } static bool nfsd4_compound_in_session(struct nfsd4_session *session, struct nfs4_sessionid *sid) @@ -2085,8 +2106,8 @@ nfsd4_destroy_clientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *csta __be32 status = 0; nfs4_lock_state(); - unconf = find_unconfirmed_client(&dc->clientid); - conf = find_confirmed_client(&dc->clientid); + unconf = find_unconfirmed_client(&dc->clientid, true); + conf = find_confirmed_client(&dc->clientid, true); if (conf) { clp = conf; @@ -2200,10 +2221,6 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, copy_clid(new, conf); else /* case 4 (new client) or cases 2, 3 (client reboot): */ gen_clid(new); - /* - * XXX: we should probably set this at creation time, and check - * for consistent minorversion use throughout: - */ new->cl_minorversion = 0; gen_callback(new, setclid, rqstp); add_to_unconfirmed(new, strhashval); @@ -2232,8 +2249,8 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, return nfserr_stale_clientid; nfs4_lock_state(); - conf = find_confirmed_client(clid); - unconf = find_unconfirmed_client(clid); + conf = find_confirmed_client(clid, false); + unconf = find_unconfirmed_client(clid, false); /* * We try hard to give out unique clientid's, so if we get an * attempt to confirm the same clientid with a different cred, @@ -2262,10 +2279,8 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp, unsigned int hash = clientstr_hashval(unconf->cl_recdir); conf = find_confirmed_client_by_str(unconf->cl_recdir, hash); - if (conf) { - nfsd4_client_record_remove(conf); + if (conf) expire_client(conf); - } move_to_confirmed(unconf); nfsd4_probe_callback(unconf); } @@ -2447,16 +2462,20 @@ same_owner_str(struct nfs4_stateowner *sop, struct xdr_netobj *owner, } static struct nfs4_openowner * -find_openstateowner_str(unsigned int hashval, struct nfsd4_open *open) +find_openstateowner_str(unsigned int hashval, struct nfsd4_open *open, bool sessions) { struct nfs4_stateowner *so; struct nfs4_openowner *oo; + struct nfs4_client *clp; list_for_each_entry(so, &ownerstr_hashtbl[hashval], so_strhash) { if (!so->so_is_open_owner) continue; if (same_owner_str(so, &open->op_owner, &open->op_clientid)) { oo = openowner(so); + clp = oo->oo_owner.so_client; + if ((bool)clp->cl_minorversion != sessions) + return NULL; renew_client(oo->oo_owner.so_client); return oo; } @@ -2600,10 +2619,10 @@ nfsd4_process_open1(struct nfsd4_compound_state *cstate, return nfserr_jukebox; strhashval = ownerstr_hashval(clientid->cl_id, &open->op_owner); - oo = find_openstateowner_str(strhashval, open); + oo = find_openstateowner_str(strhashval, open, cstate->minorversion); open->op_openowner = oo; if (!oo) { - clp = find_confirmed_client(clientid); + clp = find_confirmed_client(clientid, cstate->minorversion); if (clp == NULL) return nfserr_expired; goto new_owner; @@ -2705,11 +2724,6 @@ nfs4_check_open(struct nfs4_file *fp, struct nfsd4_open *open, struct nfs4_ol_st return nfs_ok; } -static void nfs4_free_stateid(struct nfs4_ol_stateid *s) -{ - kmem_cache_free(stateid_slab, s); -} - static inline int nfs4_access_to_access(u32 nfs4_access) { int flags = 0; @@ -3087,7 +3101,7 @@ void nfsd4_cleanup_open_state(struct nfsd4_open *open, __be32 status) if (open->op_file) nfsd4_free_file(open->op_file); if (open->op_stp) - nfs4_free_stateid(open->op_stp); + free_generic_stateid(open->op_stp); } __be32 @@ -3104,7 +3118,7 @@ nfsd4_renew(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, status = nfserr_stale_clientid; if (STALE_CLIENTID(clid, nn)) goto out; - clp = find_confirmed_client(clid); + clp = find_confirmed_client(clid, cstate->minorversion); status = nfserr_expired; if (clp == NULL) { /* We assume the client took too long to RENEW. */ @@ -3180,7 +3194,6 @@ nfs4_laundromat(void) clp = list_entry(pos, struct nfs4_client, cl_lru); dprintk("NFSD: purging unused client (clientid %08x)\n", clp->cl_clientid.cl_id); - nfsd4_client_record_remove(clp); expire_client(clp); } spin_lock(&recall_lock); @@ -3372,7 +3385,7 @@ static __be32 nfsd4_validate_stateid(struct nfs4_client *cl, stateid_t *stateid) return nfs_ok; } -static __be32 nfsd4_lookup_stateid(stateid_t *stateid, unsigned char typemask, struct nfs4_stid **s) +static __be32 nfsd4_lookup_stateid(stateid_t *stateid, unsigned char typemask, struct nfs4_stid **s, bool sessions) { struct nfs4_client *cl; struct nfsd_net *nn = net_generic(&init_net, nfsd_net_id); @@ -3381,7 +3394,7 @@ static __be32 nfsd4_lookup_stateid(stateid_t *stateid, unsigned char typemask, s return nfserr_bad_stateid; if (STALE_STATEID(stateid, nn)) return nfserr_stale_stateid; - cl = find_confirmed_client(&stateid->si_opaque.so_clid); + cl = find_confirmed_client(&stateid->si_opaque.so_clid, sessions); if (!cl) return nfserr_expired; *s = find_stateid_by_type(cl, stateid, typemask); @@ -3414,7 +3427,7 @@ nfs4_preprocess_stateid_op(struct net *net, struct nfsd4_compound_state *cstate, if (ZERO_STATEID(stateid) || ONE_STATEID(stateid)) return check_special_stateids(net, current_fh, stateid, flags); - status = nfsd4_lookup_stateid(stateid, NFS4_DELEG_STID|NFS4_OPEN_STID|NFS4_LOCK_STID, &s); + status = nfsd4_lookup_stateid(stateid, NFS4_DELEG_STID|NFS4_OPEN_STID|NFS4_LOCK_STID, &s, cstate->minorversion); if (status) return status; status = check_stateid_generation(stateid, &s->sc_stateid, nfsd4_has_session(cstate)); @@ -3564,7 +3577,7 @@ nfs4_preprocess_seqid_op(struct nfsd4_compound_state *cstate, u32 seqid, seqid, STATEID_VAL(stateid)); *stpp = NULL; - status = nfsd4_lookup_stateid(stateid, typemask, &s); + status = nfsd4_lookup_stateid(stateid, typemask, &s, cstate->minorversion); if (status) return status; *stpp = openlockstateid(s); @@ -3765,6 +3778,7 @@ nfsd4_close(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, memcpy(&close->cl_stateid, &stp->st_stid.sc_stateid, sizeof(stateid_t)); nfsd4_close_open_stateid(stp); + release_last_closed_stateid(oo); oo->oo_last_closed_stid = stp; if (list_empty(&oo->oo_owner.so_stateids)) { @@ -3801,7 +3815,7 @@ nfsd4_delegreturn(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, inode = cstate->current_fh.fh_dentry->d_inode; nfs4_lock_state(); - status = nfsd4_lookup_stateid(stateid, NFS4_DELEG_STID, &s); + status = nfsd4_lookup_stateid(stateid, NFS4_DELEG_STID, &s, cstate->minorversion); if (status) goto out; dp = delegstateid(s); @@ -4045,8 +4059,8 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfs4_lockowner *lock_sop = NULL; struct nfs4_ol_stateid *lock_stp; struct file *filp = NULL; - struct file_lock file_lock; - struct file_lock conflock; + struct file_lock *file_lock = NULL; + struct file_lock *conflock = NULL; __be32 status = 0; bool new_state = false; int lkflg; @@ -4116,21 +4130,28 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, if (!locks_in_grace(SVC_NET(rqstp)) && lock->lk_reclaim) goto out; - locks_init_lock(&file_lock); + file_lock = locks_alloc_lock(); + if (!file_lock) { + dprintk("NFSD: %s: unable to allocate lock!\n", __func__); + status = nfserr_jukebox; + goto out; + } + + locks_init_lock(file_lock); switch (lock->lk_type) { case NFS4_READ_LT: case NFS4_READW_LT: filp = find_readable_file(lock_stp->st_file); if (filp) get_lock_access(lock_stp, NFS4_SHARE_ACCESS_READ); - file_lock.fl_type = F_RDLCK; + file_lock->fl_type = F_RDLCK; break; case NFS4_WRITE_LT: case NFS4_WRITEW_LT: filp = find_writeable_file(lock_stp->st_file); if (filp) get_lock_access(lock_stp, NFS4_SHARE_ACCESS_WRITE); - file_lock.fl_type = F_WRLCK; + file_lock->fl_type = F_WRLCK; break; default: status = nfserr_inval; @@ -4140,22 +4161,23 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, status = nfserr_openmode; goto out; } - file_lock.fl_owner = (fl_owner_t)lock_sop; - file_lock.fl_pid = current->tgid; - file_lock.fl_file = filp; - file_lock.fl_flags = FL_POSIX; - file_lock.fl_lmops = &nfsd_posix_mng_ops; - - file_lock.fl_start = lock->lk_offset; - file_lock.fl_end = last_byte_offset(lock->lk_offset, lock->lk_length); - nfs4_transform_lock_offset(&file_lock); - - /* - * Try to lock the file in the VFS. - * Note: locks.c uses the BKL to protect the inode's lock list. - */ + file_lock->fl_owner = (fl_owner_t)lock_sop; + file_lock->fl_pid = current->tgid; + file_lock->fl_file = filp; + file_lock->fl_flags = FL_POSIX; + file_lock->fl_lmops = &nfsd_posix_mng_ops; + file_lock->fl_start = lock->lk_offset; + file_lock->fl_end = last_byte_offset(lock->lk_offset, lock->lk_length); + nfs4_transform_lock_offset(file_lock); + + conflock = locks_alloc_lock(); + if (!conflock) { + dprintk("NFSD: %s: unable to allocate lock!\n", __func__); + status = nfserr_jukebox; + goto out; + } - err = vfs_lock_file(filp, F_SETLK, &file_lock, &conflock); + err = vfs_lock_file(filp, F_SETLK, file_lock, conflock); switch (-err) { case 0: /* success! */ update_stateid(&lock_stp->st_stid.sc_stateid); @@ -4166,7 +4188,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, case (EAGAIN): /* conflock holds conflicting lock */ status = nfserr_denied; dprintk("NFSD: nfsd4_lock: conflicting lock found!\n"); - nfs4_set_lock_denied(&conflock, &lock->lk_denied); + nfs4_set_lock_denied(conflock, &lock->lk_denied); break; case (EDEADLK): status = nfserr_deadlock; @@ -4181,6 +4203,10 @@ out: release_lockowner(lock_sop); if (!cstate->replay_owner) nfs4_unlock_state(); + if (file_lock) + locks_free_lock(file_lock); + if (conflock) + locks_free_lock(conflock); return status; } @@ -4209,7 +4235,7 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_lockt *lockt) { struct inode *inode; - struct file_lock file_lock; + struct file_lock *file_lock = NULL; struct nfs4_lockowner *lo; __be32 status; struct nfsd_net *nn = net_generic(&init_net, nfsd_net_id); @@ -4230,15 +4256,21 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, goto out; inode = cstate->current_fh.fh_dentry->d_inode; - locks_init_lock(&file_lock); + file_lock = locks_alloc_lock(); + if (!file_lock) { + dprintk("NFSD: %s: unable to allocate lock!\n", __func__); + status = nfserr_jukebox; + goto out; + } + locks_init_lock(file_lock); switch (lockt->lt_type) { case NFS4_READ_LT: case NFS4_READW_LT: - file_lock.fl_type = F_RDLCK; + file_lock->fl_type = F_RDLCK; break; case NFS4_WRITE_LT: case NFS4_WRITEW_LT: - file_lock.fl_type = F_WRLCK; + file_lock->fl_type = F_WRLCK; break; default: dprintk("NFSD: nfs4_lockt: bad lock type!\n"); @@ -4248,25 +4280,27 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, lo = find_lockowner_str(inode, &lockt->lt_clientid, &lockt->lt_owner); if (lo) - file_lock.fl_owner = (fl_owner_t)lo; - file_lock.fl_pid = current->tgid; - file_lock.fl_flags = FL_POSIX; + file_lock->fl_owner = (fl_owner_t)lo; + file_lock->fl_pid = current->tgid; + file_lock->fl_flags = FL_POSIX; - file_lock.fl_start = lockt->lt_offset; - file_lock.fl_end = last_byte_offset(lockt->lt_offset, lockt->lt_length); + file_lock->fl_start = lockt->lt_offset; + file_lock->fl_end = last_byte_offset(lockt->lt_offset, lockt->lt_length); - nfs4_transform_lock_offset(&file_lock); + nfs4_transform_lock_offset(file_lock); - status = nfsd_test_lock(rqstp, &cstate->current_fh, &file_lock); + status = nfsd_test_lock(rqstp, &cstate->current_fh, file_lock); if (status) goto out; - if (file_lock.fl_type != F_UNLCK) { + if (file_lock->fl_type != F_UNLCK) { status = nfserr_denied; - nfs4_set_lock_denied(&file_lock, &lockt->lt_denied); + nfs4_set_lock_denied(file_lock, &lockt->lt_denied); } out: nfs4_unlock_state(); + if (file_lock) + locks_free_lock(file_lock); return status; } @@ -4276,7 +4310,7 @@ nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, { struct nfs4_ol_stateid *stp; struct file *filp = NULL; - struct file_lock file_lock; + struct file_lock *file_lock = NULL; __be32 status; int err; @@ -4298,23 +4332,29 @@ nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, status = nfserr_lock_range; goto out; } - BUG_ON(!filp); - locks_init_lock(&file_lock); - file_lock.fl_type = F_UNLCK; - file_lock.fl_owner = (fl_owner_t)lockowner(stp->st_stateowner); - file_lock.fl_pid = current->tgid; - file_lock.fl_file = filp; - file_lock.fl_flags = FL_POSIX; - file_lock.fl_lmops = &nfsd_posix_mng_ops; - file_lock.fl_start = locku->lu_offset; - - file_lock.fl_end = last_byte_offset(locku->lu_offset, locku->lu_length); - nfs4_transform_lock_offset(&file_lock); + file_lock = locks_alloc_lock(); + if (!file_lock) { + dprintk("NFSD: %s: unable to allocate lock!\n", __func__); + status = nfserr_jukebox; + goto out; + } + locks_init_lock(file_lock); + file_lock->fl_type = F_UNLCK; + file_lock->fl_owner = (fl_owner_t)lockowner(stp->st_stateowner); + file_lock->fl_pid = current->tgid; + file_lock->fl_file = filp; + file_lock->fl_flags = FL_POSIX; + file_lock->fl_lmops = &nfsd_posix_mng_ops; + file_lock->fl_start = locku->lu_offset; + + file_lock->fl_end = last_byte_offset(locku->lu_offset, + locku->lu_length); + nfs4_transform_lock_offset(file_lock); /* * Try to unlock the file in the VFS. */ - err = vfs_lock_file(filp, F_SETLK, &file_lock, NULL); + err = vfs_lock_file(filp, F_SETLK, file_lock, NULL); if (err) { dprintk("NFSD: nfs4_locku: vfs_lock_file failed!\n"); goto out_nfserr; @@ -4328,6 +4368,8 @@ nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, out: if (!cstate->replay_owner) nfs4_unlock_state(); + if (file_lock) + locks_free_lock(file_lock); return status; out_nfserr: @@ -4501,12 +4543,12 @@ nfsd4_find_reclaim_client(struct nfs4_client *clp) * Called from OPEN. Look for clientid in reclaim list. */ __be32 -nfs4_check_open_reclaim(clientid_t *clid) +nfs4_check_open_reclaim(clientid_t *clid, bool sessions) { struct nfs4_client *clp; /* find clientid in conf_id_hashtbl */ - clp = find_confirmed_client(clid); + clp = find_confirmed_client(clid, sessions); if (clp == NULL) return nfserr_reclaim_bad; @@ -4522,7 +4564,6 @@ void nfsd_forget_clients(u64 num) nfs4_lock_state(); list_for_each_entry_safe(clp, next, &client_lru, cl_lru) { - nfsd4_client_record_remove(clp); expire_client(clp); if (++count == num) break; @@ -4582,7 +4623,7 @@ void nfsd_forget_openowners(u64 num) printk(KERN_INFO "NFSD: Forgot %d open owners", count); } -int nfsd_process_n_delegations(u64 num, struct list_head *list) +static int nfsd_process_n_delegations(u64 num, struct list_head *list) { int i, count = 0; struct nfs4_file *fp, *fnext; @@ -4747,11 +4788,11 @@ __nfs4_state_shutdown(void) for (i = 0; i < CLIENT_HASH_SIZE; i++) { while (!list_empty(&conf_id_hashtbl[i])) { clp = list_entry(conf_id_hashtbl[i].next, struct nfs4_client, cl_idhash); - expire_client(clp); + destroy_client(clp); } while (!list_empty(&unconf_str_hashtbl[i])) { clp = list_entry(unconf_str_hashtbl[i].next, struct nfs4_client, cl_strhash); - expire_client(clp); + destroy_client(clp); } } INIT_LIST_HEAD(&reaplist); diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 6322df36031f..fd548d155088 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -2659,7 +2659,7 @@ static __be32 nfsd4_encode_bind_conn_to_session(struct nfsd4_compoundres *resp, RESERVE_SPACE(NFS4_MAX_SESSIONID_LEN + 8); WRITEMEM(bcts->sessionid.data, NFS4_MAX_SESSIONID_LEN); WRITE32(bcts->dir); - /* XXX: ? */ + /* Sorry, we do not yet support RDMA over 4.1: */ WRITE32(0); ADJUST_ARGS(); } diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index fa49cff5ee65..dab350dfc376 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -406,7 +406,7 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size) return rv; if (newthreads < 0) return -EINVAL; - rv = nfsd_svc(NFS_PORT, newthreads); + rv = nfsd_svc(newthreads); if (rv < 0) return rv; } else @@ -683,25 +683,6 @@ static ssize_t __write_ports_addfd(char *buf) } /* - * A '-' followed by the 'name' of a socket means we close the socket. - */ -static ssize_t __write_ports_delfd(char *buf) -{ - char *toclose; - int len = 0; - - toclose = kstrdup(buf + 1, GFP_KERNEL); - if (toclose == NULL) - return -ENOMEM; - - if (nfsd_serv != NULL) - len = svc_sock_names(nfsd_serv, buf, - SIMPLE_TRANSACTION_LIMIT, toclose); - kfree(toclose); - return len; -} - -/* * A transport listener is added by writing it's transport name and * a port number. */ @@ -712,7 +693,7 @@ static ssize_t __write_ports_addxprt(char *buf) int port, err; struct net *net = &init_net; - if (sscanf(buf, "%15s %4u", transport, &port) != 2) + if (sscanf(buf, "%15s %5u", transport, &port) != 2) return -EINVAL; if (port < 1 || port > USHRT_MAX) @@ -746,31 +727,6 @@ out_err: return err; } -/* - * A transport listener is removed by writing a "-", it's transport - * name, and it's port number. - */ -static ssize_t __write_ports_delxprt(char *buf) -{ - struct svc_xprt *xprt; - char transport[16]; - int port; - - if (sscanf(&buf[1], "%15s %4u", transport, &port) != 2) - return -EINVAL; - - if (port < 1 || port > USHRT_MAX || nfsd_serv == NULL) - return -EINVAL; - - xprt = svc_find_xprt(nfsd_serv, transport, &init_net, AF_UNSPEC, port); - if (xprt == NULL) - return -ENOTCONN; - - svc_close_xprt(xprt); - svc_xprt_put(xprt); - return 0; -} - static ssize_t __write_ports(struct file *file, char *buf, size_t size) { if (size == 0) @@ -779,15 +735,9 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size) if (isdigit(buf[0])) return __write_ports_addfd(buf); - if (buf[0] == '-' && isdigit(buf[1])) - return __write_ports_delfd(buf); - if (isalpha(buf[0])) return __write_ports_addxprt(buf); - if (buf[0] == '-' && isalpha(buf[1])) - return __write_ports_delxprt(buf); - return -EINVAL; } @@ -825,21 +775,6 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size) * OR * * Input: - * buf: C string containing a "-" followed - * by an integer value representing a - * previously passed in socket file - * descriptor - * size: non-zero length of C string in @buf - * Output: - * On success: NFS service no longer listens on that socket; - * passed-in buffer filled with a '\n'-terminated C - * string containing a unique name of the listener; - * return code is the size in bytes of the string - * On error: return code is a negative errno value - * - * OR - * - * Input: * buf: C string containing a transport * name and an unsigned integer value * representing the port to listen on, @@ -848,19 +783,6 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size) * Output: * On success: returns zero; NFS service is started * On error: return code is a negative errno value - * - * OR - * - * Input: - * buf: C string containing a "-" followed - * by a transport name and an unsigned - * integer value representing the port - * to listen on, separated by whitespace - * size: non-zero length of C string in @buf - * Output: - * On success: returns zero; NFS service no longer listens - * on that transport - * On error: return code is a negative errno value */ static ssize_t write_ports(struct file *file, char *buf, size_t size) { @@ -1008,8 +930,6 @@ static ssize_t write_gracetime(struct file *file, char *buf, size_t size) return nfsd4_write_time(file, buf, size, &nfsd4_grace); } -extern char *nfs4_recoverydir(void); - static ssize_t __write_recoverydir(struct file *file, char *buf, size_t size) { char *mesg = buf; diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h index 2244222368ab..80d5ce40aadb 100644 --- a/fs/nfsd/nfsd.h +++ b/fs/nfsd/nfsd.h @@ -65,7 +65,7 @@ extern const struct seq_operations nfs_exports_op; /* * Function prototypes. */ -int nfsd_svc(unsigned short port, int nrservs); +int nfsd_svc(int nrservs); int nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp); int nfsd_nrthreads(void); @@ -124,6 +124,7 @@ int nfs4_state_start(void); void nfs4_state_shutdown(void); void nfs4_reset_lease(time_t leasetime); int nfs4_reset_recoverydir(char *recdir); +char * nfs4_recoverydir(void); #else static inline void nfs4_state_init(void) { } static inline int nfsd4_init_slabs(void) { return 0; } @@ -132,6 +133,7 @@ static inline int nfs4_state_start(void) { return 0; } static inline void nfs4_state_shutdown(void) { } static inline void nfs4_reset_lease(time_t leasetime) { } static inline int nfs4_reset_recoverydir(char *recdir) { return 0; } +static inline char * nfs4_recoverydir(void) {return NULL; } #endif /* diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 240473cb708f..2013aa001dab 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -183,18 +183,18 @@ int nfsd_nrthreads(void) return rv; } -static int nfsd_init_socks(int port) +static int nfsd_init_socks(void) { int error; if (!list_empty(&nfsd_serv->sv_permsocks)) return 0; - error = svc_create_xprt(nfsd_serv, "udp", &init_net, PF_INET, port, + error = svc_create_xprt(nfsd_serv, "udp", &init_net, PF_INET, NFS_PORT, SVC_SOCK_DEFAULTS); if (error < 0) return error; - error = svc_create_xprt(nfsd_serv, "tcp", &init_net, PF_INET, port, + error = svc_create_xprt(nfsd_serv, "tcp", &init_net, PF_INET, NFS_PORT, SVC_SOCK_DEFAULTS); if (error < 0) return error; @@ -204,7 +204,7 @@ static int nfsd_init_socks(int port) static bool nfsd_up = false; -static int nfsd_startup(unsigned short port, int nrservs) +static int nfsd_startup(int nrservs) { int ret; @@ -218,7 +218,7 @@ static int nfsd_startup(unsigned short port, int nrservs) ret = nfsd_racache_init(2*nrservs); if (ret) return ret; - ret = nfsd_init_socks(port); + ret = nfsd_init_socks(); if (ret) goto out_racache; ret = lockd_up(&init_net); @@ -436,7 +436,7 @@ int nfsd_set_nrthreads(int n, int *nthreads) * this is the first time nrservs is nonzero. */ int -nfsd_svc(unsigned short port, int nrservs) +nfsd_svc(int nrservs) { int error; bool nfsd_up_before; @@ -458,7 +458,7 @@ nfsd_svc(unsigned short port, int nrservs) nfsd_up_before = nfsd_up; - error = nfsd_startup(port, nrservs); + error = nfsd_startup(nrservs); if (error) goto out_destroy; error = svc_set_num_threads(nfsd_serv, NULL, nrservs); @@ -487,7 +487,7 @@ static int nfsd(void *vrqstp) { struct svc_rqst *rqstp = (struct svc_rqst *) vrqstp; - int err, preverr = 0; + int err; /* Lock module and set up kernel thread */ mutex_lock(&nfsd_mutex); @@ -534,16 +534,6 @@ nfsd(void *vrqstp) ; if (err == -EINTR) break; - else if (err < 0) { - if (err != preverr) { - printk(KERN_WARNING "%s: unexpected error " - "from svc_recv (%d)\n", __func__, -err); - preverr = err; - } - schedule_timeout_uninterruptible(HZ); - continue; - } - validate_process_creds(); svc_process(rqstp); validate_process_creds(); diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index 22bd0a66c356..e036894bce57 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -373,11 +373,7 @@ static inline struct nfs4_lockowner * lockowner(struct nfs4_stateowner *so) return container_of(so, struct nfs4_lockowner, lo_owner); } -/* -* nfs4_file: a file opened by some number of (open) nfs4_stateowners. -* o fi_perfile list is used to search for conflicting -* share_acces, share_deny on the file. -*/ +/* nfs4_file: a file opened by some number of (open) nfs4_stateowners. */ struct nfs4_file { atomic_t fi_ref; struct list_head fi_hash; /* hash by "struct inode *" */ @@ -459,7 +455,7 @@ extern void nfs4_unlock_state(void); extern int nfs4_in_grace(void); extern void nfs4_release_reclaim(void); extern struct nfs4_client_reclaim *nfsd4_find_reclaim_client(struct nfs4_client *crp); -extern __be32 nfs4_check_open_reclaim(clientid_t *clid); +extern __be32 nfs4_check_open_reclaim(clientid_t *clid, bool sessions); extern void nfs4_free_openowner(struct nfs4_openowner *); extern void nfs4_free_lockowner(struct nfs4_lockowner *); extern int set_callback_cred(void); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 3f67b8e12251..c120b48ec305 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1581,7 +1581,7 @@ nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp) */ oldfs = get_fs(); set_fs(KERNEL_DS); - host_err = inode->i_op->readlink(path.dentry, buf, *lenp); + host_err = inode->i_op->readlink(path.dentry, (char __user *)buf, *lenp); set_fs(oldfs); if (host_err < 0) diff --git a/fs/nilfs2/file.c b/fs/nilfs2/file.c index a4d56ac02e6c..16f35f7423c5 100644 --- a/fs/nilfs2/file.c +++ b/fs/nilfs2/file.c @@ -116,6 +116,7 @@ static int nilfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) if (unlikely(ret)) goto out; + file_update_time(vma->vm_file); ret = __block_page_mkwrite(vma, vmf, nilfs_get_block); if (ret) { nilfs_transaction_abort(inode->i_sb); @@ -134,13 +135,13 @@ static int nilfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) static const struct vm_operations_struct nilfs_file_vm_ops = { .fault = filemap_fault, .page_mkwrite = nilfs_page_mkwrite, + .remap_pages = generic_file_remap_pages, }; static int nilfs_file_mmap(struct file *file, struct vm_area_struct *vma) { file_accessed(file); vma->vm_ops = &nilfs_file_vm_ops; - vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } diff --git a/fs/ocfs2/mmap.c b/fs/ocfs2/mmap.c index d150372fd81d..47a87dda54ce 100644 --- a/fs/ocfs2/mmap.c +++ b/fs/ocfs2/mmap.c @@ -173,6 +173,7 @@ out: static const struct vm_operations_struct ocfs2_file_vm_ops = { .fault = ocfs2_fault, .page_mkwrite = ocfs2_page_mkwrite, + .remap_pages = generic_file_remap_pages, }; int ocfs2_mmap(struct file *file, struct vm_area_struct *vma) @@ -188,7 +189,6 @@ int ocfs2_mmap(struct file *file, struct vm_area_struct *vma) ocfs2_inode_unlock(file->f_dentry->d_inode, lock_level); out: vma->vm_ops = &ocfs2_file_vm_ops; - vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } diff --git a/fs/omfs/file.c b/fs/omfs/file.c index 2c6d95257a4d..77e3cb2962b4 100644 --- a/fs/omfs/file.c +++ b/fs/omfs/file.c @@ -146,8 +146,7 @@ static int omfs_grow_extent(struct inode *inode, struct omfs_extent *oe, be64_to_cpu(entry->e_blocks); if (omfs_allocate_block(inode->i_sb, new_block)) { - entry->e_blocks = - cpu_to_be64(be64_to_cpu(entry->e_blocks) + 1); + be64_add_cpu(&entry->e_blocks, 1); terminator->e_blocks = ~(cpu_to_be64( be64_to_cpu(~terminator->e_blocks) + 1)); goto out; @@ -177,7 +176,7 @@ static int omfs_grow_extent(struct inode *inode, struct omfs_extent *oe, be64_to_cpu(~terminator->e_blocks) + (u64) new_count)); /* write in new entry */ - oe->e_extent_count = cpu_to_be32(1 + be32_to_cpu(oe->e_extent_count)); + be32_add_cpu(&oe->e_extent_count, 1); out: *ret_block = new_block; diff --git a/fs/open.c b/fs/open.c index 44da0feeca2c..59071f55bf7f 100644 --- a/fs/open.c +++ b/fs/open.c @@ -478,7 +478,7 @@ SYSCALL_DEFINE2(fchmod, unsigned int, fd, umode_t, mode) file = fget(fd); if (file) { - audit_inode(NULL, file->f_path.dentry); + audit_inode(NULL, file->f_path.dentry, 0); err = chmod_common(&file->f_path, mode); fput(file); } @@ -588,7 +588,7 @@ SYSCALL_DEFINE3(fchown, unsigned int, fd, uid_t, user, gid_t, group) error = mnt_want_write_file(f.file); if (error) goto out_fput; - audit_inode(NULL, f.file->f_path.dentry); + audit_inode(NULL, f.file->f_path.dentry, 0); error = chown_common(&f.file->f_path, user, group); mnt_drop_write_file(f.file); out_fput: @@ -859,6 +859,24 @@ static inline int build_open_flags(int flags, umode_t mode, struct open_flags *o } /** + * file_open_name - open file and return file pointer + * + * @name: struct filename containing path to open + * @flags: open flags as per the open(2) second argument + * @mode: mode for the new file if O_CREAT is set, else ignored + * + * This is the helper to open a file from kernelspace if you really + * have to. But in generally you should not do this, so please move + * along, nothing to see here.. + */ +struct file *file_open_name(struct filename *name, int flags, umode_t mode) +{ + struct open_flags op; + int lookup = build_open_flags(flags, mode, &op); + return do_filp_open(AT_FDCWD, name, &op, lookup); +} + +/** * filp_open - open file and return file pointer * * @filename: path to open @@ -871,9 +889,8 @@ static inline int build_open_flags(int flags, umode_t mode, struct open_flags *o */ struct file *filp_open(const char *filename, int flags, umode_t mode) { - struct open_flags op; - int lookup = build_open_flags(flags, mode, &op); - return do_filp_open(AT_FDCWD, filename, &op, lookup); + struct filename name = {.name = filename}; + return file_open_name(&name, flags, mode); } EXPORT_SYMBOL(filp_open); @@ -895,7 +912,7 @@ long do_sys_open(int dfd, const char __user *filename, int flags, umode_t mode) { struct open_flags op; int lookup = build_open_flags(flags, mode, &op); - char *tmp = getname(filename); + struct filename *tmp = getname(filename); int fd = PTR_ERR(tmp); if (!IS_ERR(tmp)) { diff --git a/fs/proc/base.c b/fs/proc/base.c index d295af993677..144a96732dd7 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -873,111 +873,6 @@ static const struct file_operations proc_environ_operations = { .release = mem_release, }; -static ssize_t oom_adjust_read(struct file *file, char __user *buf, - size_t count, loff_t *ppos) -{ - struct task_struct *task = get_proc_task(file->f_path.dentry->d_inode); - char buffer[PROC_NUMBUF]; - size_t len; - int oom_adjust = OOM_DISABLE; - unsigned long flags; - - if (!task) - return -ESRCH; - - if (lock_task_sighand(task, &flags)) { - oom_adjust = task->signal->oom_adj; - unlock_task_sighand(task, &flags); - } - - put_task_struct(task); - - len = snprintf(buffer, sizeof(buffer), "%i\n", oom_adjust); - - return simple_read_from_buffer(buf, count, ppos, buffer, len); -} - -static ssize_t oom_adjust_write(struct file *file, const char __user *buf, - size_t count, loff_t *ppos) -{ - struct task_struct *task; - char buffer[PROC_NUMBUF]; - int oom_adjust; - unsigned long flags; - int err; - - memset(buffer, 0, sizeof(buffer)); - if (count > sizeof(buffer) - 1) - count = sizeof(buffer) - 1; - if (copy_from_user(buffer, buf, count)) { - err = -EFAULT; - goto out; - } - - err = kstrtoint(strstrip(buffer), 0, &oom_adjust); - if (err) - goto out; - if ((oom_adjust < OOM_ADJUST_MIN || oom_adjust > OOM_ADJUST_MAX) && - oom_adjust != OOM_DISABLE) { - err = -EINVAL; - goto out; - } - - task = get_proc_task(file->f_path.dentry->d_inode); - if (!task) { - err = -ESRCH; - goto out; - } - - task_lock(task); - if (!task->mm) { - err = -EINVAL; - goto err_task_lock; - } - - if (!lock_task_sighand(task, &flags)) { - err = -ESRCH; - goto err_task_lock; - } - - if (oom_adjust < task->signal->oom_adj && !capable(CAP_SYS_RESOURCE)) { - err = -EACCES; - goto err_sighand; - } - - /* - * Warn that /proc/pid/oom_adj is deprecated, see - * Documentation/feature-removal-schedule.txt. - */ - printk_once(KERN_WARNING "%s (%d): /proc/%d/oom_adj is deprecated, please use /proc/%d/oom_score_adj instead.\n", - current->comm, task_pid_nr(current), task_pid_nr(task), - task_pid_nr(task)); - task->signal->oom_adj = oom_adjust; - /* - * Scale /proc/pid/oom_score_adj appropriately ensuring that a maximum - * value is always attainable. - */ - if (task->signal->oom_adj == OOM_ADJUST_MAX) - task->signal->oom_score_adj = OOM_SCORE_ADJ_MAX; - else - task->signal->oom_score_adj = (oom_adjust * OOM_SCORE_ADJ_MAX) / - -OOM_DISABLE; - trace_oom_score_adj_update(task); -err_sighand: - unlock_task_sighand(task, &flags); -err_task_lock: - task_unlock(task); - put_task_struct(task); -out: - return err < 0 ? err : count; -} - -static const struct file_operations proc_oom_adjust_operations = { - .read = oom_adjust_read, - .write = oom_adjust_write, - .llseek = generic_file_llseek, -}; - static ssize_t oom_score_adj_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { @@ -1051,15 +946,7 @@ static ssize_t oom_score_adj_write(struct file *file, const char __user *buf, if (has_capability_noaudit(current, CAP_SYS_RESOURCE)) task->signal->oom_score_adj_min = oom_score_adj; trace_oom_score_adj_update(task); - /* - * Scale /proc/pid/oom_adj appropriately ensuring that OOM_DISABLE is - * always attainable. - */ - if (task->signal->oom_score_adj == OOM_SCORE_ADJ_MIN) - task->signal->oom_adj = OOM_DISABLE; - else - task->signal->oom_adj = (oom_score_adj * OOM_ADJUST_MAX) / - OOM_SCORE_ADJ_MAX; + err_sighand: unlock_task_sighand(task, &flags); err_task_lock: @@ -2371,7 +2258,8 @@ static void *proc_self_follow_link(struct dentry *dentry, struct nameidata *nd) pid_t tgid = task_tgid_nr_ns(current, ns); char *name = ERR_PTR(-ENOENT); if (tgid) { - name = __getname(); + /* 11 for max length of signed int in decimal + NULL term */ + name = kmalloc(12, GFP_KERNEL); if (!name) name = ERR_PTR(-ENOMEM); else @@ -2386,7 +2274,7 @@ static void proc_self_put_link(struct dentry *dentry, struct nameidata *nd, { char *s = nd_get_link(nd); if (!IS_ERR(s)) - __putname(s); + kfree(s); } static const struct inode_operations proc_self_inode_operations = { @@ -2710,7 +2598,6 @@ static const struct pid_entry tgid_base_stuff[] = { REG("cgroup", S_IRUGO, proc_cgroup_operations), #endif INF("oom_score", S_IRUGO, proc_oom_score), - REG("oom_adj", S_IRUGO|S_IWUSR, proc_oom_adjust_operations), REG("oom_score_adj", S_IRUGO|S_IWUSR, proc_oom_score_adj_operations), #ifdef CONFIG_AUDITSYSCALL REG("loginuid", S_IWUSR|S_IRUGO, proc_loginuid_operations), @@ -3077,7 +2964,6 @@ static const struct pid_entry tid_base_stuff[] = { REG("cgroup", S_IRUGO, proc_cgroup_operations), #endif INF("oom_score", S_IRUGO, proc_oom_score), - REG("oom_adj", S_IRUGO|S_IWUSR, proc_oom_adjust_operations), REG("oom_score_adj", S_IRUGO|S_IWUSR, proc_oom_score_adj_operations), #ifdef CONFIG_AUDITSYSCALL REG("loginuid", S_IWUSR|S_IRUGO, proc_loginuid_operations), diff --git a/fs/proc/generic.c b/fs/proc/generic.c index b3647fe6a608..0d80cef4cfb9 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -427,7 +427,7 @@ struct dentry *proc_lookup_de(struct proc_dir_entry *de, struct inode *dir, if (!memcmp(dentry->d_name.name, de->name, de->namelen)) { pde_get(de); spin_unlock(&proc_subdir_lock); - error = -EINVAL; + error = -ENOMEM; inode = proc_get_inode(dir->i_sb, de); goto out_unlock; } @@ -605,7 +605,8 @@ static struct proc_dir_entry *__proc_create(struct proc_dir_entry **parent, unsigned int len; /* make sure name is valid */ - if (!name || !strlen(name)) goto out; + if (!name || !strlen(name)) + goto out; if (xlate_proc_name(name, parent, &fn) != 0) goto out; @@ -616,20 +617,18 @@ static struct proc_dir_entry *__proc_create(struct proc_dir_entry **parent, len = strlen(fn); - ent = kmalloc(sizeof(struct proc_dir_entry) + len + 1, GFP_KERNEL); - if (!ent) goto out; + ent = kzalloc(sizeof(struct proc_dir_entry) + len + 1, GFP_KERNEL); + if (!ent) + goto out; - memset(ent, 0, sizeof(struct proc_dir_entry)); memcpy(ent->name, fn, len + 1); ent->namelen = len; ent->mode = mode; ent->nlink = nlink; atomic_set(&ent->count, 1); - ent->pde_users = 0; spin_lock_init(&ent->pde_unload_lock); - ent->pde_unload_completion = NULL; INIT_LIST_HEAD(&ent->pde_openers); - out: +out: return ent; } diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 7ac817b64a71..3b22bbdee9ec 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -450,7 +450,6 @@ struct inode *proc_get_inode(struct super_block *sb, struct proc_dir_entry *de) return NULL; if (inode->i_state & I_NEW) { inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; - PROC_I(inode)->fd = 0; PROC_I(inode)->pde = de; if (de->mode) { diff --git a/fs/proc/internal.h b/fs/proc/internal.h index 67925a7bd8cb..43973b084abf 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -12,6 +12,7 @@ #include <linux/sched.h> #include <linux/proc_fs.h> struct ctl_table_header; +struct mempolicy; extern struct proc_dir_entry proc_root; #ifdef CONFIG_PROC_SYSCTL @@ -74,6 +75,9 @@ struct proc_maps_private { #ifdef CONFIG_MMU struct vm_area_struct *tail_vma; #endif +#ifdef CONFIG_NUMA + struct mempolicy *task_mempolicy; +#endif }; void proc_init_inodecache(void); @@ -103,7 +107,7 @@ static inline int task_dumpable(struct task_struct *task) if (mm) dumpable = get_dumpable(mm); task_unlock(task); - if(dumpable == 1) + if (dumpable == SUID_DUMPABLE_ENABLED) return 1; return 0; } diff --git a/fs/proc/page.c b/fs/proc/page.c index 7fcd0d60a968..b8730d9ebaee 100644 --- a/fs/proc/page.c +++ b/fs/proc/page.c @@ -115,7 +115,13 @@ u64 stable_page_flags(struct page *page) u |= 1 << KPF_COMPOUND_TAIL; if (PageHuge(page)) u |= 1 << KPF_HUGE; - else if (PageTransCompound(page)) + /* + * PageTransCompound can be true for non-huge compound pages (slab + * pages or pages allocated by drivers with __GFP_COMP) because it + * just checks PG_head/PG_tail, so we need to check PageLRU to make + * sure a given page is a thp, not a non-huge compound page. + */ + else if (PageTransCompound(page) && PageLRU(compound_trans_head(page))) u |= 1 << KPF_THP; /* diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index eb7cc91b7258..a781bdf06694 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -142,6 +142,7 @@ static int insert_entry(struct ctl_table_header *head, struct ctl_table *entry) } rb_link_node(node, parent, p); + rb_insert_color(node, &head->parent->root); return 0; } @@ -168,10 +169,8 @@ static void init_header(struct ctl_table_header *head, head->node = node; if (node) { struct ctl_table *entry; - for (entry = table; entry->procname; entry++, node++) { - rb_init_node(&node->node); + for (entry = table; entry->procname; entry++, node++) node->header = head; - } } } @@ -266,8 +265,7 @@ void sysctl_head_put(struct ctl_table_header *head) static struct ctl_table_header *sysctl_head_grab(struct ctl_table_header *head) { - if (!head) - BUG(); + BUG_ON(!head); spin_lock(&sysctl_lock); if (!use_table(head)) head = ERR_PTR(-ENOENT); diff --git a/fs/proc/root.c b/fs/proc/root.c index 9a2d9fd7cadd..9889a92d2e01 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -61,7 +61,7 @@ static int proc_parse_options(char *options, struct pid_namespace *pid) if (!*p) continue; - args[0].to = args[0].from = 0; + args[0].to = args[0].from = NULL; token = match_token(p, tokens, args); switch (token) { case Opt_gid: diff --git a/fs/proc/stat.c b/fs/proc/stat.c index 64c3b3172367..e296572c73ed 100644 --- a/fs/proc/stat.c +++ b/fs/proc/stat.c @@ -45,10 +45,13 @@ static cputime64_t get_iowait_time(int cpu) static u64 get_idle_time(int cpu) { - u64 idle, idle_time = get_cpu_idle_time_us(cpu, NULL); + u64 idle, idle_time = -1ULL; + + if (cpu_online(cpu)) + idle_time = get_cpu_idle_time_us(cpu, NULL); if (idle_time == -1ULL) - /* !NO_HZ so we can rely on cpustat.idle */ + /* !NO_HZ or cpu offline so we can rely on cpustat.idle */ idle = kcpustat_cpu(cpu).cpustat[CPUTIME_IDLE]; else idle = usecs_to_cputime64(idle_time); @@ -58,10 +61,13 @@ static u64 get_idle_time(int cpu) static u64 get_iowait_time(int cpu) { - u64 iowait, iowait_time = get_cpu_iowait_time_us(cpu, NULL); + u64 iowait, iowait_time = -1ULL; + + if (cpu_online(cpu)) + iowait_time = get_cpu_iowait_time_us(cpu, NULL); if (iowait_time == -1ULL) - /* !NO_HZ so we can rely on cpustat.iowait */ + /* !NO_HZ or cpu offline so we can rely on cpustat.iowait */ iowait = kcpustat_cpu(cpu).cpustat[CPUTIME_IOWAIT]; else iowait = usecs_to_cputime64(iowait_time); diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 4540b8f76f16..90c63f9392a5 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -54,7 +54,7 @@ void task_mem(struct seq_file *m, struct mm_struct *mm) "VmPTE:\t%8lu kB\n" "VmSwap:\t%8lu kB\n", hiwater_vm << (PAGE_SHIFT-10), - (total_vm - mm->reserved_vm) << (PAGE_SHIFT-10), + total_vm << (PAGE_SHIFT-10), mm->locked_vm << (PAGE_SHIFT-10), mm->pinned_vm << (PAGE_SHIFT-10), hiwater_rss << (PAGE_SHIFT-10), @@ -90,10 +90,55 @@ static void pad_len_spaces(struct seq_file *m, int len) seq_printf(m, "%*c", len, ' '); } +#ifdef CONFIG_NUMA +/* + * These functions are for numa_maps but called in generic **maps seq_file + * ->start(), ->stop() ops. + * + * numa_maps scans all vmas under mmap_sem and checks their mempolicy. + * Each mempolicy object is controlled by reference counting. The problem here + * is how to avoid accessing dead mempolicy object. + * + * Because we're holding mmap_sem while reading seq_file, it's safe to access + * each vma's mempolicy, no vma objects will never drop refs to mempolicy. + * + * A task's mempolicy (task->mempolicy) has different behavior. task->mempolicy + * is set and replaced under mmap_sem but unrefed and cleared under task_lock(). + * So, without task_lock(), we cannot trust get_vma_policy() because we cannot + * gurantee the task never exits under us. But taking task_lock() around + * get_vma_plicy() causes lock order problem. + * + * To access task->mempolicy without lock, we hold a reference count of an + * object pointed by task->mempolicy and remember it. This will guarantee + * that task->mempolicy points to an alive object or NULL in numa_maps accesses. + */ +static void hold_task_mempolicy(struct proc_maps_private *priv) +{ + struct task_struct *task = priv->task; + + task_lock(task); + priv->task_mempolicy = task->mempolicy; + mpol_get(priv->task_mempolicy); + task_unlock(task); +} +static void release_task_mempolicy(struct proc_maps_private *priv) +{ + mpol_put(priv->task_mempolicy); +} +#else +static void hold_task_mempolicy(struct proc_maps_private *priv) +{ +} +static void release_task_mempolicy(struct proc_maps_private *priv) +{ +} +#endif + static void vma_stop(struct proc_maps_private *priv, struct vm_area_struct *vma) { if (vma && vma != priv->tail_vma) { struct mm_struct *mm = vma->vm_mm; + release_task_mempolicy(priv); up_read(&mm->mmap_sem); mmput(mm); } @@ -132,7 +177,7 @@ static void *m_start(struct seq_file *m, loff_t *pos) tail_vma = get_gate_vma(priv->task->mm); priv->tail_vma = tail_vma; - + hold_task_mempolicy(priv); /* Start with last addr hint */ vma = find_vma(mm, last_addr); if (last_addr && vma) { @@ -159,6 +204,7 @@ out: if (vma) return vma; + release_task_mempolicy(priv); /* End of vmas has been reached */ m->version = (tail_vma != NULL)? 0: -1UL; up_read(&mm->mmap_sem); @@ -1158,6 +1204,7 @@ static int show_numa_map(struct seq_file *m, void *v, int is_pid) struct vm_area_struct *vma = v; struct numa_maps *md = &numa_priv->md; struct file *file = vma->vm_file; + struct task_struct *task = proc_priv->task; struct mm_struct *mm = vma->vm_mm; struct mm_walk walk = {}; struct mempolicy *pol; @@ -1177,7 +1224,7 @@ static int show_numa_map(struct seq_file *m, void *v, int is_pid) walk.private = md; walk.mm = mm; - pol = get_vma_policy(proc_priv->task, vma, vma->vm_start); + pol = get_vma_policy(task, vma, vma->vm_start); mpol_to_str(buffer, sizeof(buffer), pol, 0); mpol_cond_put(pol); @@ -1189,7 +1236,7 @@ static int show_numa_map(struct seq_file *m, void *v, int is_pid) } else if (vma->vm_start <= mm->brk && vma->vm_end >= mm->start_brk) { seq_printf(m, " heap"); } else { - pid_t tid = vm_is_stack(proc_priv->task, vma, is_pid); + pid_t tid = vm_is_stack(task, vma, is_pid); if (tid != 0) { /* * Thread stack in /proc/PID/task/TID/maps or diff --git a/fs/pstore/Kconfig b/fs/pstore/Kconfig index d39bb5cce883..ca71db69da07 100644 --- a/fs/pstore/Kconfig +++ b/fs/pstore/Kconfig @@ -23,6 +23,7 @@ config PSTORE_FTRACE bool "Persistent function tracer" depends on PSTORE depends on FUNCTION_TRACER + depends on DEBUG_FS help With this option kernel traces function calls into a persistent ram buffer that can be decoded and dumped after reboot through diff --git a/fs/pstore/ftrace.c b/fs/pstore/ftrace.c index a130d484b7d3..2d57e1ac0115 100644 --- a/fs/pstore/ftrace.c +++ b/fs/pstore/ftrace.c @@ -17,19 +17,113 @@ #include <linux/percpu.h> #include <linux/smp.h> #include <linux/atomic.h> +#include <linux/types.h> +#include <linux/mutex.h> +#include <linux/ftrace.h> +#include <linux/fs.h> +#include <linux/debugfs.h> +#include <linux/err.h> +#include <linux/cache.h> #include <asm/barrier.h> #include "internal.h" -void notrace pstore_ftrace_call(unsigned long ip, unsigned long parent_ip) +static void notrace pstore_ftrace_call(unsigned long ip, + unsigned long parent_ip) { + unsigned long flags; struct pstore_ftrace_record rec = {}; if (unlikely(oops_in_progress)) return; + local_irq_save(flags); + rec.ip = ip; rec.parent_ip = parent_ip; pstore_ftrace_encode_cpu(&rec, raw_smp_processor_id()); psinfo->write_buf(PSTORE_TYPE_FTRACE, 0, NULL, 0, (void *)&rec, sizeof(rec), psinfo); + + local_irq_restore(flags); +} + +static struct ftrace_ops pstore_ftrace_ops __read_mostly = { + .func = pstore_ftrace_call, +}; + +static DEFINE_MUTEX(pstore_ftrace_lock); +static bool pstore_ftrace_enabled; + +static ssize_t pstore_ftrace_knob_write(struct file *f, const char __user *buf, + size_t count, loff_t *ppos) +{ + u8 on; + ssize_t ret; + + ret = kstrtou8_from_user(buf, count, 2, &on); + if (ret) + return ret; + + mutex_lock(&pstore_ftrace_lock); + + if (!on ^ pstore_ftrace_enabled) + goto out; + + if (on) + ret = register_ftrace_function(&pstore_ftrace_ops); + else + ret = unregister_ftrace_function(&pstore_ftrace_ops); + if (ret) { + pr_err("%s: unable to %sregister ftrace ops: %zd\n", + __func__, on ? "" : "un", ret); + goto err; + } + + pstore_ftrace_enabled = on; +out: + ret = count; +err: + mutex_unlock(&pstore_ftrace_lock); + + return ret; +} + +static ssize_t pstore_ftrace_knob_read(struct file *f, char __user *buf, + size_t count, loff_t *ppos) +{ + char val[] = { '0' + pstore_ftrace_enabled, '\n' }; + + return simple_read_from_buffer(buf, count, ppos, val, sizeof(val)); +} + +static const struct file_operations pstore_knob_fops = { + .open = simple_open, + .read = pstore_ftrace_knob_read, + .write = pstore_ftrace_knob_write, +}; + +void pstore_register_ftrace(void) +{ + struct dentry *dir; + struct dentry *file; + + if (!psinfo->write_buf) + return; + + dir = debugfs_create_dir("pstore", NULL); + if (!dir) { + pr_err("%s: unable to create pstore directory\n", __func__); + return; + } + + file = debugfs_create_file("record_ftrace", 0600, dir, NULL, + &pstore_knob_fops); + if (!file) { + pr_err("%s: unable to create record_ftrace file\n", __func__); + goto err_file; + } + + return; +err_file: + debugfs_remove(dir); } diff --git a/fs/pstore/internal.h b/fs/pstore/internal.h index 0d0d3b7d5f12..4847f588b7d5 100644 --- a/fs/pstore/internal.h +++ b/fs/pstore/internal.h @@ -39,6 +39,12 @@ pstore_ftrace_decode_cpu(struct pstore_ftrace_record *rec) #endif } +#ifdef CONFIG_PSTORE_FTRACE +extern void pstore_register_ftrace(void); +#else +static inline void pstore_register_ftrace(void) {} +#endif + extern struct pstore_info *psinfo; extern void pstore_set_kmsg_bytes(int); diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c index 29996e8793a7..a40da07e93d6 100644 --- a/fs/pstore/platform.c +++ b/fs/pstore/platform.c @@ -164,7 +164,13 @@ static void pstore_console_write(struct console *con, const char *s, unsigned c) if (c > psinfo->bufsize) c = psinfo->bufsize; - spin_lock_irqsave(&psinfo->buf_lock, flags); + + if (oops_in_progress) { + if (!spin_trylock_irqsave(&psinfo->buf_lock, flags)) + break; + } else { + spin_lock_irqsave(&psinfo->buf_lock, flags); + } memcpy(psinfo->buf, s, c); psinfo->write(PSTORE_TYPE_CONSOLE, 0, NULL, 0, c, psinfo); spin_unlock_irqrestore(&psinfo->buf_lock, flags); @@ -236,6 +242,7 @@ int pstore_register(struct pstore_info *psi) kmsg_dump_register(&pstore_dumper); pstore_register_console(); + pstore_register_ftrace(); if (pstore_update_ms >= 0) { pstore_timer.expires = jiffies + diff --git a/fs/pstore/ram.c b/fs/pstore/ram.c index 0b311bc18916..1a4f6da58eab 100644 --- a/fs/pstore/ram.c +++ b/fs/pstore/ram.c @@ -32,6 +32,7 @@ #include <linux/ioport.h> #include <linux/platform_device.h> #include <linux/slab.h> +#include <linux/compiler.h> #include <linux/pstore_ram.h> #define RAMOOPS_KERNMSG_HDR "====" @@ -181,12 +182,11 @@ static size_t ramoops_write_kmsg_hdr(struct persistent_ram_zone *prz) return len; } - -static int ramoops_pstore_write_buf(enum pstore_type_id type, - enum kmsg_dump_reason reason, - u64 *id, unsigned int part, - const char *buf, size_t size, - struct pstore_info *psi) +static int notrace ramoops_pstore_write_buf(enum pstore_type_id type, + enum kmsg_dump_reason reason, + u64 *id, unsigned int part, + const char *buf, size_t size, + struct pstore_info *psi) { struct ramoops_context *cxt = psi->data; struct persistent_ram_zone *prz = cxt->przs[cxt->dump_write_cnt]; @@ -406,7 +406,7 @@ static int __devinit ramoops_probe(struct platform_device *pdev) goto fail_init_fprz; if (!cxt->przs && !cxt->cprz && !cxt->fprz) { - pr_err("memory size too small, minimum is %lu\n", + pr_err("memory size too small, minimum is %zu\n", cxt->console_size + cxt->record_size + cxt->ftrace_size); goto fail_cnt; @@ -414,13 +414,14 @@ static int __devinit ramoops_probe(struct platform_device *pdev) cxt->pstore.data = cxt; /* - * Console can handle any buffer size, so prefer dumps buffer - * size since usually it is smaller. + * Console can handle any buffer size, so prefer LOG_LINE_MAX. If we + * have to handle dumps, we must have at least record_size buffer. And + * for ftrace, bufsize is irrelevant (if bufsize is 0, buf will be + * ZERO_SIZE_PTR). */ - if (cxt->przs) - cxt->pstore.bufsize = cxt->przs[0]->buffer_size; - else - cxt->pstore.bufsize = cxt->cprz->buffer_size; + if (cxt->console_size) + cxt->pstore.bufsize = 1024; /* LOG_LINE_MAX */ + cxt->pstore.bufsize = max(cxt->record_size, cxt->pstore.bufsize); cxt->pstore.buf = kmalloc(cxt->pstore.bufsize, GFP_KERNEL); spin_lock_init(&cxt->pstore.buf_lock); if (!cxt->pstore.buf) { @@ -537,6 +538,7 @@ postcore_initcall(ramoops_init); static void __exit ramoops_exit(void) { platform_driver_unregister(&ramoops_driver); + platform_device_unregister(dummy); kfree(dummy_data); } module_exit(ramoops_exit); diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index 557a9c20a215..05ae3c97f7a5 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -1160,6 +1160,8 @@ static int need_print_warning(struct dquot_warn *warn) return uid_eq(current_fsuid(), warn->w_dq_id.uid); case GRPQUOTA: return in_group_p(warn->w_dq_id.gid); + case PRJQUOTA: /* Never taken... Just make gcc happy */ + return 0; } return 0; } diff --git a/fs/quota/quota.c b/fs/quota/quota.c index ff0135d6bc51..af1661f7a54f 100644 --- a/fs/quota/quota.c +++ b/fs/quota/quota.c @@ -331,11 +331,11 @@ static struct super_block *quotactl_block(const char __user *special, int cmd) #ifdef CONFIG_BLOCK struct block_device *bdev; struct super_block *sb; - char *tmp = getname(special); + struct filename *tmp = getname(special); if (IS_ERR(tmp)) return ERR_CAST(tmp); - bdev = lookup_bdev(tmp); + bdev = lookup_bdev(tmp->name); putname(tmp); if (IS_ERR(bdev)) return ERR_CAST(bdev); diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 46485557cdc6..f27f01a98aa2 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -1573,8 +1573,10 @@ struct dentry *reiserfs_fh_to_dentry(struct super_block *sb, struct fid *fid, reiserfs_warning(sb, "reiserfs-13077", "nfsd/reiserfs, fhtype=%d, len=%d - odd", fh_type, fh_len); - fh_type = 5; + fh_type = fh_len; } + if (fh_len < 2) + return NULL; return reiserfs_get_dentry(sb, fid->raw[0], fid->raw[1], (fh_type == 3 || fh_type >= 5) ? fid->raw[2] : 0); @@ -1583,6 +1585,8 @@ struct dentry *reiserfs_fh_to_dentry(struct super_block *sb, struct fid *fid, struct dentry *reiserfs_fh_to_parent(struct super_block *sb, struct fid *fid, int fh_len, int fh_type) { + if (fh_type > fh_len) + fh_type = fh_len; if (fh_type < 4) return NULL; diff --git a/fs/super.c b/fs/super.c index 5fdf7ff32c4e..12f123712161 100644 --- a/fs/super.c +++ b/fs/super.c @@ -186,15 +186,8 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags) spin_lock_init(&s->s_inode_lru_lock); INIT_LIST_HEAD(&s->s_mounts); init_rwsem(&s->s_umount); - mutex_init(&s->s_lock); lockdep_set_class(&s->s_umount, &type->s_umount_key); /* - * The locking rules for s_lock are up to the - * filesystem. For example ext3fs has different - * lock ordering than usbfs: - */ - lockdep_set_class(&s->s_lock, &type->s_lock_key); - /* * sget() can have s_umount recursion. * * When it cannot find a suitable sb, it allocates a new @@ -394,22 +387,6 @@ bool grab_super_passive(struct super_block *sb) return false; } -/* - * Superblock locking. We really ought to get rid of these two. - */ -void lock_super(struct super_block * sb) -{ - mutex_lock(&sb->s_lock); -} - -void unlock_super(struct super_block * sb) -{ - mutex_unlock(&sb->s_lock); -} - -EXPORT_SYMBOL(lock_super); -EXPORT_SYMBOL(unlock_super); - /** * generic_shutdown_super - common helper for ->kill_sb() * @sb: superblock to kill @@ -865,7 +842,7 @@ int get_anon_bdev(dev_t *p) else if (error) return -EAGAIN; - if ((dev & MAX_ID_MASK) == (1 << MINORBITS)) { + if ((dev & MAX_IDR_MASK) == (1 << MINORBITS)) { spin_lock(&unnamed_dev_lock); ida_remove(&unnamed_dev_ida, dev); if (unnamed_dev_start > dev) diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 6b0bb00d4d2b..2fbdff6be25c 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -485,20 +485,18 @@ int __sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd) /** * sysfs_pathname - return full path to sysfs dirent * @sd: sysfs_dirent whose path we want - * @path: caller allocated buffer + * @path: caller allocated buffer of size PATH_MAX * * Gives the name "/" to the sysfs_root entry; any path returned * is relative to wherever sysfs is mounted. - * - * XXX: does no error checking on @path size */ static char *sysfs_pathname(struct sysfs_dirent *sd, char *path) { if (sd->s_parent) { sysfs_pathname(sd->s_parent, path); - strcat(path, "/"); + strlcat(path, "/", PATH_MAX); } - strcat(path, sd->s_name); + strlcat(path, sd->s_name, PATH_MAX); return path; } @@ -531,9 +529,11 @@ int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd) char *path = kzalloc(PATH_MAX, GFP_KERNEL); WARN(1, KERN_WARNING "sysfs: cannot create duplicate filename '%s'\n", - (path == NULL) ? sd->s_name : - strcat(strcat(sysfs_pathname(acxt->parent_sd, path), "/"), - sd->s_name)); + (path == NULL) ? sd->s_name + : (sysfs_pathname(acxt->parent_sd, path), + strlcat(path, "/", PATH_MAX), + strlcat(path, sd->s_name, PATH_MAX), + path)); kfree(path); } diff --git a/fs/sysv/balloc.c b/fs/sysv/balloc.c index 9a6ad96acf27..921c053fc052 100644 --- a/fs/sysv/balloc.c +++ b/fs/sysv/balloc.c @@ -60,12 +60,12 @@ void sysv_free_block(struct super_block * sb, sysv_zone_t nr) return; } - lock_super(sb); + mutex_lock(&sbi->s_lock); count = fs16_to_cpu(sbi, *sbi->s_bcache_count); if (count > sbi->s_flc_size) { printk("sysv_free_block: flc_count > flc_size\n"); - unlock_super(sb); + mutex_unlock(&sbi->s_lock); return; } /* If the free list head in super-block is full, it is copied @@ -77,7 +77,7 @@ void sysv_free_block(struct super_block * sb, sysv_zone_t nr) bh = sb_getblk(sb, block); if (!bh) { printk("sysv_free_block: getblk() failed\n"); - unlock_super(sb); + mutex_unlock(&sbi->s_lock); return; } memset(bh->b_data, 0, sb->s_blocksize); @@ -93,7 +93,7 @@ void sysv_free_block(struct super_block * sb, sysv_zone_t nr) *sbi->s_bcache_count = cpu_to_fs16(sbi, count); fs32_add(sbi, sbi->s_free_blocks, 1); dirty_sb(sb); - unlock_super(sb); + mutex_unlock(&sbi->s_lock); } sysv_zone_t sysv_new_block(struct super_block * sb) @@ -104,7 +104,7 @@ sysv_zone_t sysv_new_block(struct super_block * sb) struct buffer_head * bh; unsigned count; - lock_super(sb); + mutex_lock(&sbi->s_lock); count = fs16_to_cpu(sbi, *sbi->s_bcache_count); if (count == 0) /* Applies only to Coherent FS */ @@ -147,11 +147,11 @@ sysv_zone_t sysv_new_block(struct super_block * sb) /* Now the free list head in the superblock is valid again. */ fs32_add(sbi, sbi->s_free_blocks, -1); dirty_sb(sb); - unlock_super(sb); + mutex_unlock(&sbi->s_lock); return nr; Enospc: - unlock_super(sb); + mutex_unlock(&sbi->s_lock); return 0; } @@ -173,7 +173,7 @@ unsigned long sysv_count_free_blocks(struct super_block * sb) if (sbi->s_type == FSTYPE_AFS) return 0; - lock_super(sb); + mutex_lock(&sbi->s_lock); sb_count = fs32_to_cpu(sbi, *sbi->s_free_blocks); if (0) @@ -211,7 +211,7 @@ unsigned long sysv_count_free_blocks(struct super_block * sb) if (count != sb_count) goto Ecount; done: - unlock_super(sb); + mutex_unlock(&sbi->s_lock); return count; Einval: diff --git a/fs/sysv/ialloc.c b/fs/sysv/ialloc.c index 8233b02eccae..f9db4eb31db4 100644 --- a/fs/sysv/ialloc.c +++ b/fs/sysv/ialloc.c @@ -118,7 +118,7 @@ void sysv_free_inode(struct inode * inode) "%s\n", inode->i_sb->s_id); return; } - lock_super(sb); + mutex_lock(&sbi->s_lock); count = fs16_to_cpu(sbi, *sbi->s_sb_fic_count); if (count < sbi->s_fic_size) { *sv_sb_fic_inode(sb,count++) = cpu_to_fs16(sbi, ino); @@ -128,7 +128,7 @@ void sysv_free_inode(struct inode * inode) dirty_sb(sb); memset(raw_inode, 0, sizeof(struct sysv_inode)); mark_buffer_dirty(bh); - unlock_super(sb); + mutex_unlock(&sbi->s_lock); brelse(bh); } @@ -147,13 +147,13 @@ struct inode * sysv_new_inode(const struct inode * dir, umode_t mode) if (!inode) return ERR_PTR(-ENOMEM); - lock_super(sb); + mutex_lock(&sbi->s_lock); count = fs16_to_cpu(sbi, *sbi->s_sb_fic_count); if (count == 0 || (*sv_sb_fic_inode(sb,count-1) == 0)) { count = refill_free_cache(sb); if (count == 0) { iput(inode); - unlock_super(sb); + mutex_unlock(&sbi->s_lock); return ERR_PTR(-ENOSPC); } } @@ -174,7 +174,7 @@ struct inode * sysv_new_inode(const struct inode * dir, umode_t mode) sysv_write_inode(inode, &wbc); /* ensure inode not allocated again */ mark_inode_dirty(inode); /* cleared by sysv_write_inode() */ /* That's it. */ - unlock_super(sb); + mutex_unlock(&sbi->s_lock); return inode; } @@ -185,7 +185,7 @@ unsigned long sysv_count_free_inodes(struct super_block * sb) struct sysv_inode * raw_inode; int ino, count, sb_count; - lock_super(sb); + mutex_lock(&sbi->s_lock); sb_count = fs16_to_cpu(sbi, *sbi->s_sb_total_free_inodes); @@ -213,7 +213,7 @@ unsigned long sysv_count_free_inodes(struct super_block * sb) if (count != sb_count) goto Einval; out: - unlock_super(sb); + mutex_unlock(&sbi->s_lock); return count; Einval: diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index d33e506c1eac..c327d4ee1235 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -36,7 +36,7 @@ static int sysv_sync_fs(struct super_block *sb, int wait) struct sysv_sb_info *sbi = SYSV_SB(sb); unsigned long time = get_seconds(), old_time; - lock_super(sb); + mutex_lock(&sbi->s_lock); /* * If we are going to write out the super block, @@ -51,7 +51,7 @@ static int sysv_sync_fs(struct super_block *sb, int wait) mark_buffer_dirty(sbi->s_bh2); } - unlock_super(sb); + mutex_unlock(&sbi->s_lock); return 0; } diff --git a/fs/sysv/super.c b/fs/sysv/super.c index 7491c33b6468..a38e87bdd78d 100644 --- a/fs/sysv/super.c +++ b/fs/sysv/super.c @@ -368,6 +368,7 @@ static int sysv_fill_super(struct super_block *sb, void *data, int silent) sbi->s_sb = sb; sbi->s_block_base = 0; + mutex_init(&sbi->s_lock); sb->s_fs_info = sbi; sb_set_blocksize(sb, BLOCK_SIZE); diff --git a/fs/sysv/sysv.h b/fs/sysv/sysv.h index 0bc35fdc58e2..69d488986cce 100644 --- a/fs/sysv/sysv.h +++ b/fs/sysv/sysv.h @@ -58,6 +58,7 @@ struct sysv_sb_info { u32 s_nzones; /* same as s_sbd->s_fsize */ u16 s_namelen; /* max length of dir entry */ int s_forced_ro; + struct mutex s_lock; }; /* diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index ff48c5a85309..5bc77817f382 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -1536,6 +1536,7 @@ out_unlock: static const struct vm_operations_struct ubifs_file_vm_ops = { .fault = filemap_fault, .page_mkwrite = ubifs_vm_page_mkwrite, + .remap_pages = generic_file_remap_pages, }; static int ubifs_file_mmap(struct file *file, struct vm_area_struct *vma) diff --git a/fs/ufs/balloc.c b/fs/ufs/balloc.c index 1b3e410bf334..a7ea492ae660 100644 --- a/fs/ufs/balloc.c +++ b/fs/ufs/balloc.c @@ -54,7 +54,7 @@ void ufs_free_fragments(struct inode *inode, u64 fragment, unsigned count) if (ufs_fragnum(fragment) + count > uspi->s_fpg) ufs_error (sb, "ufs_free_fragments", "internal error"); - lock_super(sb); + mutex_lock(&UFS_SB(sb)->s_lock); cgno = ufs_dtog(uspi, fragment); bit = ufs_dtogd(uspi, fragment); @@ -118,12 +118,12 @@ void ufs_free_fragments(struct inode *inode, u64 fragment, unsigned count) ubh_sync_block(UCPI_UBH(ucpi)); ufs_mark_sb_dirty(sb); - unlock_super (sb); + mutex_unlock(&UFS_SB(sb)->s_lock); UFSD("EXIT\n"); return; failed: - unlock_super (sb); + mutex_unlock(&UFS_SB(sb)->s_lock); UFSD("EXIT (FAILED)\n"); return; } @@ -155,7 +155,7 @@ void ufs_free_blocks(struct inode *inode, u64 fragment, unsigned count) goto failed; } - lock_super(sb); + mutex_lock(&UFS_SB(sb)->s_lock); do_more: overflow = 0; @@ -215,12 +215,12 @@ do_more: } ufs_mark_sb_dirty(sb); - unlock_super (sb); + mutex_unlock(&UFS_SB(sb)->s_lock); UFSD("EXIT\n"); return; failed_unlock: - unlock_super (sb); + mutex_unlock(&UFS_SB(sb)->s_lock); failed: UFSD("EXIT (FAILED)\n"); return; @@ -361,7 +361,7 @@ u64 ufs_new_fragments(struct inode *inode, void *p, u64 fragment, usb1 = ubh_get_usb_first(uspi); *err = -ENOSPC; - lock_super (sb); + mutex_lock(&UFS_SB(sb)->s_lock); tmp = ufs_data_ptr_to_cpu(sb, p); if (count + ufs_fragnum(fragment) > uspi->s_fpb) { @@ -382,19 +382,19 @@ u64 ufs_new_fragments(struct inode *inode, void *p, u64 fragment, "fragment %llu, tmp %llu\n", (unsigned long long)fragment, (unsigned long long)tmp); - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); return INVBLOCK; } if (fragment < UFS_I(inode)->i_lastfrag) { UFSD("EXIT (ALREADY ALLOCATED)\n"); - unlock_super (sb); + mutex_unlock(&UFS_SB(sb)->s_lock); return 0; } } else { if (tmp) { UFSD("EXIT (ALREADY ALLOCATED)\n"); - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); return 0; } } @@ -403,7 +403,7 @@ u64 ufs_new_fragments(struct inode *inode, void *p, u64 fragment, * There is not enough space for user on the device */ if (!capable(CAP_SYS_RESOURCE) && ufs_freespace(uspi, UFS_MINFREE) <= 0) { - unlock_super (sb); + mutex_unlock(&UFS_SB(sb)->s_lock); UFSD("EXIT (FAILED)\n"); return 0; } @@ -428,7 +428,7 @@ u64 ufs_new_fragments(struct inode *inode, void *p, u64 fragment, ufs_clear_frags(inode, result + oldcount, newcount - oldcount, locked_page != NULL); } - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); UFSD("EXIT, result %llu\n", (unsigned long long)result); return result; } @@ -443,7 +443,7 @@ u64 ufs_new_fragments(struct inode *inode, void *p, u64 fragment, fragment + count); ufs_clear_frags(inode, result + oldcount, newcount - oldcount, locked_page != NULL); - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); UFSD("EXIT, result %llu\n", (unsigned long long)result); return result; } @@ -481,7 +481,7 @@ u64 ufs_new_fragments(struct inode *inode, void *p, u64 fragment, *err = 0; UFS_I(inode)->i_lastfrag = max(UFS_I(inode)->i_lastfrag, fragment + count); - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); if (newcount < request) ufs_free_fragments (inode, result + newcount, request - newcount); ufs_free_fragments (inode, tmp, oldcount); @@ -489,7 +489,7 @@ u64 ufs_new_fragments(struct inode *inode, void *p, u64 fragment, return result; } - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); UFSD("EXIT (FAILED)\n"); return 0; } diff --git a/fs/ufs/ialloc.c b/fs/ufs/ialloc.c index e84cbe21b986..d0426d74817b 100644 --- a/fs/ufs/ialloc.c +++ b/fs/ufs/ialloc.c @@ -71,11 +71,11 @@ void ufs_free_inode (struct inode * inode) ino = inode->i_ino; - lock_super (sb); + mutex_lock(&UFS_SB(sb)->s_lock); if (!((ino > 1) && (ino < (uspi->s_ncg * uspi->s_ipg )))) { ufs_warning(sb, "ufs_free_inode", "reserved inode or nonexistent inode %u\n", ino); - unlock_super (sb); + mutex_unlock(&UFS_SB(sb)->s_lock); return; } @@ -83,7 +83,7 @@ void ufs_free_inode (struct inode * inode) bit = ufs_inotocgoff (ino); ucpi = ufs_load_cylinder (sb, cg); if (!ucpi) { - unlock_super (sb); + mutex_unlock(&UFS_SB(sb)->s_lock); return; } ucg = ubh_get_ucg(UCPI_UBH(ucpi)); @@ -117,7 +117,7 @@ void ufs_free_inode (struct inode * inode) ubh_sync_block(UCPI_UBH(ucpi)); ufs_mark_sb_dirty(sb); - unlock_super (sb); + mutex_unlock(&UFS_SB(sb)->s_lock); UFSD("EXIT\n"); } @@ -197,7 +197,7 @@ struct inode *ufs_new_inode(struct inode *dir, umode_t mode) uspi = sbi->s_uspi; usb1 = ubh_get_usb_first(uspi); - lock_super (sb); + mutex_lock(&sbi->s_lock); /* * Try to place the inode in its parent directory @@ -333,20 +333,20 @@ cg_found: brelse(bh); } - unlock_super (sb); + mutex_unlock(&sbi->s_lock); UFSD("allocating inode %lu\n", inode->i_ino); UFSD("EXIT\n"); return inode; fail_remove_inode: - unlock_super(sb); + mutex_unlock(&sbi->s_lock); clear_nlink(inode); iput(inode); UFSD("EXIT (FAILED): err %d\n", err); return ERR_PTR(err); failed: - unlock_super (sb); + mutex_unlock(&sbi->s_lock); make_bad_inode(inode); iput (inode); UFSD("EXIT (FAILED): err %d\n", err); diff --git a/fs/ufs/super.c b/fs/ufs/super.c index f7cfecfe1cab..dc8e3a861d0f 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -699,7 +699,7 @@ static int ufs_sync_fs(struct super_block *sb, int wait) unsigned flags; lock_ufs(sb); - lock_super(sb); + mutex_lock(&UFS_SB(sb)->s_lock); UFSD("ENTER\n"); @@ -717,7 +717,7 @@ static int ufs_sync_fs(struct super_block *sb, int wait) ufs_put_cstotal(sb); UFSD("EXIT\n"); - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); unlock_ufs(sb); return 0; @@ -805,6 +805,7 @@ static int ufs_fill_super(struct super_block *sb, void *data, int silent) } #endif mutex_init(&sbi->mutex); + mutex_init(&sbi->s_lock); spin_lock_init(&sbi->work_lock); INIT_DELAYED_WORK(&sbi->sync_work, delayed_sync_fs); /* @@ -1280,7 +1281,7 @@ static int ufs_remount (struct super_block *sb, int *mount_flags, char *data) unsigned flags; lock_ufs(sb); - lock_super(sb); + mutex_lock(&UFS_SB(sb)->s_lock); uspi = UFS_SB(sb)->s_uspi; flags = UFS_SB(sb)->s_flags; usb1 = ubh_get_usb_first(uspi); @@ -1294,7 +1295,7 @@ static int ufs_remount (struct super_block *sb, int *mount_flags, char *data) new_mount_opt = 0; ufs_set_opt (new_mount_opt, ONERROR_LOCK); if (!ufs_parse_options (data, &new_mount_opt)) { - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); unlock_ufs(sb); return -EINVAL; } @@ -1302,14 +1303,14 @@ static int ufs_remount (struct super_block *sb, int *mount_flags, char *data) new_mount_opt |= ufstype; } else if ((new_mount_opt & UFS_MOUNT_UFSTYPE) != ufstype) { printk("ufstype can't be changed during remount\n"); - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); unlock_ufs(sb); return -EINVAL; } if ((*mount_flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY)) { UFS_SB(sb)->s_mount_opt = new_mount_opt; - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); unlock_ufs(sb); return 0; } @@ -1334,7 +1335,7 @@ static int ufs_remount (struct super_block *sb, int *mount_flags, char *data) #ifndef CONFIG_UFS_FS_WRITE printk("ufs was compiled with read-only support, " "can't be mounted as read-write\n"); - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); unlock_ufs(sb); return -EINVAL; #else @@ -1344,13 +1345,13 @@ static int ufs_remount (struct super_block *sb, int *mount_flags, char *data) ufstype != UFS_MOUNT_UFSTYPE_SUNx86 && ufstype != UFS_MOUNT_UFSTYPE_UFS2) { printk("this ufstype is read-only supported\n"); - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); unlock_ufs(sb); return -EINVAL; } if (!ufs_read_cylinder_structures(sb)) { printk("failed during remounting\n"); - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); unlock_ufs(sb); return -EPERM; } @@ -1358,7 +1359,7 @@ static int ufs_remount (struct super_block *sb, int *mount_flags, char *data) #endif } UFS_SB(sb)->s_mount_opt = new_mount_opt; - unlock_super(sb); + mutex_unlock(&UFS_SB(sb)->s_lock); unlock_ufs(sb); return 0; } diff --git a/fs/ufs/ufs.h b/fs/ufs/ufs.h index 343e6fc571e5..ff2c15ab81aa 100644 --- a/fs/ufs/ufs.h +++ b/fs/ufs/ufs.h @@ -24,6 +24,7 @@ struct ufs_sb_info { int work_queued; /* non-zero if the delayed work is queued */ struct delayed_work sync_work; /* FS sync delayed work */ spinlock_t work_lock; /* protects sync_work and work_queued */ + struct mutex s_lock; }; struct ufs_inode_info { diff --git a/fs/xattr.c b/fs/xattr.c index 1780f062dbaf..e21c119f4f99 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -412,7 +412,7 @@ SYSCALL_DEFINE5(fsetxattr, int, fd, const char __user *, name, if (!f.file) return error; dentry = f.file->f_path.dentry; - audit_inode(NULL, dentry); + audit_inode(NULL, dentry, 0); error = mnt_want_write_file(f.file); if (!error) { error = setxattr(dentry, name, value, size, flags); @@ -507,7 +507,7 @@ SYSCALL_DEFINE4(fgetxattr, int, fd, const char __user *, name, if (!f.file) return error; - audit_inode(NULL, f.file->f_path.dentry); + audit_inode(NULL, f.file->f_path.dentry, 0); error = getxattr(f.file->f_path.dentry, name, value, size); fdput(f); return error; @@ -586,7 +586,7 @@ SYSCALL_DEFINE3(flistxattr, int, fd, char __user *, list, size_t, size) if (!f.file) return error; - audit_inode(NULL, f.file->f_path.dentry); + audit_inode(NULL, f.file->f_path.dentry, 0); error = listxattr(f.file->f_path.dentry, list, size); fdput(f); return error; @@ -655,7 +655,7 @@ SYSCALL_DEFINE2(fremovexattr, int, fd, const char __user *, name) if (!f.file) return error; dentry = f.file->f_path.dentry; - audit_inode(NULL, dentry); + audit_inode(NULL, dentry, 0); error = mnt_want_write_file(f.file); if (!error) { error = removexattr(dentry, name); @@ -846,7 +846,7 @@ static int __simple_xattr_set(struct simple_xattrs *xattrs, const char *name, const void *value, size_t size, int flags) { struct simple_xattr *xattr; - struct simple_xattr *uninitialized_var(new_xattr); + struct simple_xattr *new_xattr = NULL; int err = 0; /* value == NULL means remove */ diff --git a/fs/xattr_acl.c b/fs/xattr_acl.c index 11efd830b5f5..9fbea87fdb6e 100644 --- a/fs/xattr_acl.c +++ b/fs/xattr_acl.c @@ -45,7 +45,7 @@ static void posix_acl_fix_xattr_userns( break; case ACL_GROUP: gid = make_kgid(from, le32_to_cpu(entry->e_id)); - entry->e_id = cpu_to_le32(from_kuid(to, uid)); + entry->e_id = cpu_to_le32(from_kgid(to, gid)); break; default: break; diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c index 42679223a0fd..8c6d1d70278c 100644 --- a/fs/xfs/xfs_export.c +++ b/fs/xfs/xfs_export.c @@ -189,6 +189,9 @@ xfs_fs_fh_to_parent(struct super_block *sb, struct fid *fid, struct xfs_fid64 *fid64 = (struct xfs_fid64 *)fid; struct inode *inode = NULL; + if (fh_len < xfs_fileid_length(fileid_type)) + return NULL; + switch (fileid_type) { case FILEID_INO32_GEN_PARENT: inode = xfs_nfs_get_inode(sb, fid->i32.parent_ino, diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 1eaeb8be3aae..aa473fa640a2 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -940,7 +940,6 @@ xfs_file_mmap( struct vm_area_struct *vma) { vma->vm_ops = &xfs_file_vm_ops; - vma->vm_flags |= VM_CAN_NONLINEAR; file_accessed(filp); return 0; @@ -1443,4 +1442,5 @@ const struct file_operations xfs_dir_file_operations = { static const struct vm_operations_struct xfs_file_vm_ops = { .fault = filemap_fault, .page_mkwrite = xfs_vm_page_mkwrite, + .remap_pages = generic_file_remap_pages, }; |