From e7fd37ba12170cc414be8b639dfc2c5f7172fac2 Mon Sep 17 00:00:00 2001 From: Ma Shimiao Date: Tue, 12 Dec 2017 09:43:49 +0800 Subject: cgroup: avoid copying strings longer than the buffers cgroup root name and file name have max length limit, we should avoid copying longer name than that to the name. tj: minor update to $SUBJ. Signed-off-by: Ma Shimiao Signed-off-by: Tejun Heo --- kernel/cgroup/cgroup.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 0b1ffe147f24..18d71fbd3923 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -1397,7 +1397,7 @@ static char *cgroup_file_name(struct cgroup *cgrp, const struct cftype *cft, cgroup_on_dfl(cgrp) ? ss->name : ss->legacy_name, cft->name); else - strncpy(buf, cft->name, CGROUP_FILE_NAME_MAX); + strscpy(buf, cft->name, CGROUP_FILE_NAME_MAX); return buf; } @@ -1864,9 +1864,9 @@ void init_cgroup_root(struct cgroup_root *root, struct cgroup_sb_opts *opts) root->flags = opts->flags; if (opts->release_agent) - strcpy(root->release_agent_path, opts->release_agent); + strscpy(root->release_agent_path, opts->release_agent, PATH_MAX); if (opts->name) - strcpy(root->name, opts->name); + strscpy(root->name, opts->name, MAX_CGROUP_ROOT_NAMELEN); if (opts->cpuset_clone_children) set_bit(CGRP_CPUSET_CLONE_CHILDREN, &root->cgrp.flags); } -- cgit v1.2.3 From 50034ed49645463a16327cad05694e201e6b4126 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 15 Dec 2017 05:09:47 -0800 Subject: cgroup: use strlcpy() instead of strscpy() to avoid spurious warning As long as cft->name is guaranteed to be NUL-terminated, using strlcpy() would work just as well and avoid that warning, so the change below could be folded into that commit. Signed-off-by: Arnd Bergmann Signed-off-by: Tejun Heo --- kernel/cgroup/cgroup.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 18d71fbd3923..f4c2f8cb5748 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -1397,7 +1397,7 @@ static char *cgroup_file_name(struct cgroup *cgrp, const struct cftype *cft, cgroup_on_dfl(cgrp) ? ss->name : ss->legacy_name, cft->name); else - strscpy(buf, cft->name, CGROUP_FILE_NAME_MAX); + strlcpy(buf, cft->name, CGROUP_FILE_NAME_MAX); return buf; } @@ -1864,9 +1864,9 @@ void init_cgroup_root(struct cgroup_root *root, struct cgroup_sb_opts *opts) root->flags = opts->flags; if (opts->release_agent) - strscpy(root->release_agent_path, opts->release_agent, PATH_MAX); + strlcpy(root->release_agent_path, opts->release_agent, PATH_MAX); if (opts->name) - strscpy(root->name, opts->name, MAX_CGROUP_ROOT_NAMELEN); + strlcpy(root->name, opts->name, MAX_CGROUP_ROOT_NAMELEN); if (opts->cpuset_clone_children) set_bit(CGRP_CPUSET_CLONE_CHILDREN, &root->cgrp.flags); } -- cgit v1.2.3 From 116d2f7496c51b2e02e8e4ecdd2bdf5fb9d5a641 Mon Sep 17 00:00:00 2001 From: Prateek Sood Date: Tue, 19 Dec 2017 12:56:57 +0530 Subject: cgroup: Fix deadlock in cpu hotplug path Deadlock during cgroup migration from cpu hotplug path when a task T is being moved from source to destination cgroup. kworker/0:0 cpuset_hotplug_workfn() cpuset_hotplug_update_tasks() hotplug_update_tasks_legacy() remove_tasks_in_empty_cpuset() cgroup_transfer_tasks() // stuck in iterator loop cgroup_migrate() cgroup_migrate_add_task() In cgroup_migrate_add_task() it checks for PF_EXITING flag of task T. Task T will not migrate to destination cgroup. css_task_iter_start() will keep pointing to task T in loop waiting for task T cg_list node to be removed. Task T do_exit() exit_signals() // sets PF_EXITING exit_task_namespaces() switch_task_namespaces() free_nsproxy() put_mnt_ns() drop_collected_mounts() namespace_unlock() synchronize_rcu() _synchronize_rcu_expedited() schedule_work() // on cpu0 low priority worker pool wait_event() // waiting for work item to execute Task T inserted a work item in the worklist of cpu0 low priority worker pool. It is waiting for expedited grace period work item to execute. This work item will only be executed once kworker/0:0 complete execution of cpuset_hotplug_workfn(). kworker/0:0 ==> Task T ==>kworker/0:0 In case of PF_EXITING task being migrated from source to destination cgroup, migrate next available task in source cgroup. Signed-off-by: Prateek Sood Signed-off-by: Tejun Heo --- kernel/cgroup/cgroup-v1.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/cgroup/cgroup-v1.c b/kernel/cgroup/cgroup-v1.c index 024085daab1a..a2c05d2476ac 100644 --- a/kernel/cgroup/cgroup-v1.c +++ b/kernel/cgroup/cgroup-v1.c @@ -123,7 +123,11 @@ int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from) */ do { css_task_iter_start(&from->self, 0, &it); - task = css_task_iter_next(&it); + + do { + task = css_task_iter_next(&it); + } while (task && (task->flags & PF_EXITING)); + if (task) get_task_struct(task); css_task_iter_end(&it); -- cgit v1.2.3 From 74d0833c659a8a54735e5efdd44f4b225af68586 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 20 Dec 2017 07:09:19 -0800 Subject: cgroup: fix css_task_iter crash on CSS_TASK_ITER_PROC While teaching css_task_iter to handle skipping over tasks which aren't group leaders, bc2fb7ed089f ("cgroup: add @flags to css_task_iter_start() and implement CSS_TASK_ITER_PROCS") introduced a silly bug. CSS_TASK_ITER_PROCS is implemented by repeating css_task_iter_advance() while the advanced cursor is pointing to a non-leader thread. However, the cursor variable, @l, wasn't updated when the iteration has to advance to the next css_set and the following repetition would operate on the terminal @l from the previous iteration which isn't pointing to a valid task leading to oopses like the following or infinite looping. BUG: unable to handle kernel NULL pointer dereference at 0000000000000254 IP: __task_pid_nr_ns+0xc7/0xf0 PGD 0 P4D 0 Oops: 0000 [#1] SMP ... CPU: 2 PID: 1 Comm: systemd Not tainted 4.14.4-200.fc26.x86_64 #1 Hardware name: System manufacturer System Product Name/PRIME B350M-A, BIOS 3203 11/09/2017 task: ffff88c4baee8000 task.stack: ffff96d5c3158000 RIP: 0010:__task_pid_nr_ns+0xc7/0xf0 RSP: 0018:ffff96d5c315bd50 EFLAGS: 00010206 RAX: 0000000000000000 RBX: ffff88c4b68c6000 RCX: 0000000000000250 RDX: ffffffffa5e47960 RSI: 0000000000000000 RDI: ffff88c490f6ab00 RBP: ffff96d5c315bd50 R08: 0000000000001000 R09: 0000000000000005 R10: ffff88c4be006b80 R11: ffff88c42f1b8004 R12: ffff96d5c315bf18 R13: ffff88c42d7dd200 R14: ffff88c490f6a510 R15: ffff88c4b68c6000 FS: 00007f9446f8ea00(0000) GS:ffff88c4be680000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000254 CR3: 00000007f956f000 CR4: 00000000003406e0 Call Trace: cgroup_procs_show+0x19/0x30 cgroup_seqfile_show+0x4c/0xb0 kernfs_seq_show+0x21/0x30 seq_read+0x2ec/0x3f0 kernfs_fop_read+0x134/0x180 __vfs_read+0x37/0x160 ? security_file_permission+0x9b/0xc0 vfs_read+0x8e/0x130 SyS_read+0x55/0xc0 entry_SYSCALL_64_fastpath+0x1a/0xa5 RIP: 0033:0x7f94455f942d RSP: 002b:00007ffe81ba2d00 EFLAGS: 00000293 ORIG_RAX: 0000000000000000 RAX: ffffffffffffffda RBX: 00005574e2233f00 RCX: 00007f94455f942d RDX: 0000000000001000 RSI: 00005574e2321a90 RDI: 000000000000002b RBP: 0000000000000000 R08: 00005574e2321a90 R09: 00005574e231de60 R10: 00007f94458c8b38 R11: 0000000000000293 R12: 00007f94458c8ae0 R13: 00007ffe81ba3800 R14: 0000000000000000 R15: 00005574e2116560 Code: 04 74 0e 89 f6 48 8d 04 76 48 8d 04 c5 f0 05 00 00 48 8b bf b8 05 00 00 48 01 c7 31 c0 48 8b 0f 48 85 c9 74 18 8b b2 30 08 00 00 <3b> 71 04 77 0d 48 c1 e6 05 48 01 f1 48 3b 51 38 74 09 5d c3 8b RIP: __task_pid_nr_ns+0xc7/0xf0 RSP: ffff96d5c315bd50 Fix it by moving the initialization of the cursor below the repeat label. While at it, rename it to @next for readability. Signed-off-by: Tejun Heo Fixes: bc2fb7ed089f ("cgroup: add @flags to css_task_iter_start() and implement CSS_TASK_ITER_PROCS") Cc: stable@vger.kernel.org # v4.14+ Reported-by: Laura Abbott Reported-by: Bronek Kozicki Reported-by: George Amanakis Signed-off-by: Tejun Heo --- kernel/cgroup/cgroup.c | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index f4c2f8cb5748..2cf06c274e4c 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -4125,26 +4125,24 @@ static void css_task_iter_advance_css_set(struct css_task_iter *it) static void css_task_iter_advance(struct css_task_iter *it) { - struct list_head *l = it->task_pos; + struct list_head *next; lockdep_assert_held(&css_set_lock); - WARN_ON_ONCE(!l); - repeat: /* * Advance iterator to find next entry. cset->tasks is consumed * first and then ->mg_tasks. After ->mg_tasks, we move onto the * next cset. */ - l = l->next; + next = it->task_pos->next; - if (l == it->tasks_head) - l = it->mg_tasks_head->next; + if (next == it->tasks_head) + next = it->mg_tasks_head->next; - if (l == it->mg_tasks_head) + if (next == it->mg_tasks_head) css_task_iter_advance_css_set(it); else - it->task_pos = l; + it->task_pos = next; /* if PROCS, skip over tasks which aren't group leaders */ if ((it->flags & CSS_TASK_ITER_PROCS) && it->task_pos && -- cgit v1.2.3 From 040ee69226f8a96b7943645d68f41d5d44b5ff7d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 2 Dec 2017 20:20:38 -0500 Subject: fix "netfilter: xt_bpf: Fix XT_BPF_MODE_FD_PINNED mode of 'xt_bpf_info_v1'" Descriptor table is a shared object; it's not a place where you can stick temporary references to files, especially when we don't need an opened file at all. Cc: stable@vger.kernel.org # v4.14 Fixes: 98589a0998b8 ("netfilter: xt_bpf: Fix XT_BPF_MODE_FD_PINNED mode of 'xt_bpf_info_v1'") Signed-off-by: Al Viro --- kernel/bpf/inode.c | 40 +++++++++++++++++++++++++++++++++++++++- kernel/bpf/syscall.c | 2 +- 2 files changed, 40 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c index 01aaef1a77c5..5bb5e49ef4c3 100644 --- a/kernel/bpf/inode.c +++ b/kernel/bpf/inode.c @@ -368,7 +368,45 @@ out: putname(pname); return ret; } -EXPORT_SYMBOL_GPL(bpf_obj_get_user); + +static struct bpf_prog *__get_prog_inode(struct inode *inode, enum bpf_prog_type type) +{ + struct bpf_prog *prog; + int ret = inode_permission(inode, MAY_READ | MAY_WRITE); + if (ret) + return ERR_PTR(ret); + + if (inode->i_op == &bpf_map_iops) + return ERR_PTR(-EINVAL); + if (inode->i_op != &bpf_prog_iops) + return ERR_PTR(-EACCES); + + prog = inode->i_private; + + ret = security_bpf_prog(prog); + if (ret < 0) + return ERR_PTR(ret); + + if (!bpf_prog_get_ok(prog, &type, false)) + return ERR_PTR(-EINVAL); + + return bpf_prog_inc(prog); +} + +struct bpf_prog *bpf_prog_get_type_path(const char *name, enum bpf_prog_type type) +{ + struct bpf_prog *prog; + struct path path; + int ret = kern_path(name, LOOKUP_FOLLOW, &path); + if (ret) + return ERR_PTR(ret); + prog = __get_prog_inode(d_backing_inode(path.dentry), type); + if (!IS_ERR(prog)) + touch_atime(&path); + path_put(&path); + return prog; +} +EXPORT_SYMBOL(bpf_prog_get_type_path); static void bpf_evict_inode(struct inode *inode) { diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 2c4cfeaa8d5e..5cb783fc8224 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -1057,7 +1057,7 @@ struct bpf_prog *bpf_prog_inc_not_zero(struct bpf_prog *prog) } EXPORT_SYMBOL_GPL(bpf_prog_inc_not_zero); -static bool bpf_prog_get_ok(struct bpf_prog *prog, +bool bpf_prog_get_ok(struct bpf_prog *prog, enum bpf_prog_type *attach_type, bool attach_drv) { /* not an attachment, just a refcount inc, always allow */ -- cgit v1.2.3 From 5731a879d03bdaa00265f8ebc32dfd0e65d25276 Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Thu, 4 Jan 2018 20:02:09 -0800 Subject: bpf: sockmap missing NULL psock check Add psock NULL check to handle a racing sock event that can get the sk_callback_lock before this case but after xchg happens causing the refcnt to hit zero and sock user data (psock) to be null and queued for garbage collection. Also add a comment in the code because this is a bit subtle and not obvious in my opinion. Signed-off-by: John Fastabend Signed-off-by: Daniel Borkmann --- kernel/bpf/sockmap.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c index 5ee2e41893d9..1712d319c2d8 100644 --- a/kernel/bpf/sockmap.c +++ b/kernel/bpf/sockmap.c @@ -591,8 +591,15 @@ static void sock_map_free(struct bpf_map *map) write_lock_bh(&sock->sk_callback_lock); psock = smap_psock_sk(sock); - smap_list_remove(psock, &stab->sock_map[i]); - smap_release_sock(psock, sock); + /* This check handles a racing sock event that can get the + * sk_callback_lock before this case but after xchg happens + * causing the refcnt to hit zero and sock user data (psock) + * to be null and queued for garbage collection. + */ + if (likely(psock)) { + smap_list_remove(psock, &stab->sock_map[i]); + smap_release_sock(psock, sock); + } write_unlock_bh(&sock->sk_callback_lock); } rcu_read_unlock(); -- cgit v1.2.3 From b2157399cc9898260d6031c5bfe45fe137c1fbe7 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Sun, 7 Jan 2018 17:33:02 -0800 Subject: bpf: prevent out-of-bounds speculation Under speculation, CPUs may mis-predict branches in bounds checks. Thus, memory accesses under a bounds check may be speculated even if the bounds check fails, providing a primitive for building a side channel. To avoid leaking kernel data round up array-based maps and mask the index after bounds check, so speculated load with out of bounds index will load either valid value from the array or zero from the padded area. Unconditionally mask index for all array types even when max_entries are not rounded to power of 2 for root user. When map is created by unpriv user generate a sequence of bpf insns that includes AND operation to make sure that JITed code includes the same 'index & index_mask' operation. If prog_array map is created by unpriv user replace bpf_tail_call(ctx, map, index); with if (index >= max_entries) { index &= map->index_mask; bpf_tail_call(ctx, map, index); } (along with roundup to power 2) to prevent out-of-bounds speculation. There is secondary redundant 'if (index >= max_entries)' in the interpreter and in all JITs, but they can be optimized later if necessary. Other array-like maps (cpumap, devmap, sockmap, perf_event_array, cgroup_array) cannot be used by unpriv, so no changes there. That fixes bpf side of "Variant 1: bounds check bypass (CVE-2017-5753)" on all architectures with and without JIT. v2->v3: Daniel noticed that attack potentially can be crafted via syscall commands without loading the program, so add masking to those paths as well. Signed-off-by: Alexei Starovoitov Acked-by: John Fastabend Signed-off-by: Daniel Borkmann --- kernel/bpf/arraymap.c | 47 ++++++++++++++++++++++++++++++++++++----------- kernel/bpf/verifier.c | 36 ++++++++++++++++++++++++++++++++++++ 2 files changed, 72 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 7c25426d3cf5..aaa319848e7d 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -53,9 +53,10 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) { bool percpu = attr->map_type == BPF_MAP_TYPE_PERCPU_ARRAY; int numa_node = bpf_map_attr_numa_node(attr); + u32 elem_size, index_mask, max_entries; + bool unpriv = !capable(CAP_SYS_ADMIN); struct bpf_array *array; u64 array_size; - u32 elem_size; /* check sanity of attributes */ if (attr->max_entries == 0 || attr->key_size != 4 || @@ -72,11 +73,20 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) elem_size = round_up(attr->value_size, 8); + max_entries = attr->max_entries; + index_mask = roundup_pow_of_two(max_entries) - 1; + + if (unpriv) + /* round up array size to nearest power of 2, + * since cpu will speculate within index_mask limits + */ + max_entries = index_mask + 1; + array_size = sizeof(*array); if (percpu) - array_size += (u64) attr->max_entries * sizeof(void *); + array_size += (u64) max_entries * sizeof(void *); else - array_size += (u64) attr->max_entries * elem_size; + array_size += (u64) max_entries * elem_size; /* make sure there is no u32 overflow later in round_up() */ if (array_size >= U32_MAX - PAGE_SIZE) @@ -86,6 +96,8 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) array = bpf_map_area_alloc(array_size, numa_node); if (!array) return ERR_PTR(-ENOMEM); + array->index_mask = index_mask; + array->map.unpriv_array = unpriv; /* copy mandatory map attributes */ array->map.map_type = attr->map_type; @@ -121,12 +133,13 @@ static void *array_map_lookup_elem(struct bpf_map *map, void *key) if (unlikely(index >= array->map.max_entries)) return NULL; - return array->value + array->elem_size * index; + return array->value + array->elem_size * (index & array->index_mask); } /* emit BPF instructions equivalent to C code of array_map_lookup_elem() */ static u32 array_map_gen_lookup(struct bpf_map *map, struct bpf_insn *insn_buf) { + struct bpf_array *array = container_of(map, struct bpf_array, map); struct bpf_insn *insn = insn_buf; u32 elem_size = round_up(map->value_size, 8); const int ret = BPF_REG_0; @@ -135,7 +148,12 @@ static u32 array_map_gen_lookup(struct bpf_map *map, struct bpf_insn *insn_buf) *insn++ = BPF_ALU64_IMM(BPF_ADD, map_ptr, offsetof(struct bpf_array, value)); *insn++ = BPF_LDX_MEM(BPF_W, ret, index, 0); - *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 3); + if (map->unpriv_array) { + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 4); + *insn++ = BPF_ALU32_IMM(BPF_AND, ret, array->index_mask); + } else { + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 3); + } if (is_power_of_2(elem_size)) { *insn++ = BPF_ALU64_IMM(BPF_LSH, ret, ilog2(elem_size)); @@ -157,7 +175,7 @@ static void *percpu_array_map_lookup_elem(struct bpf_map *map, void *key) if (unlikely(index >= array->map.max_entries)) return NULL; - return this_cpu_ptr(array->pptrs[index]); + return this_cpu_ptr(array->pptrs[index & array->index_mask]); } int bpf_percpu_array_copy(struct bpf_map *map, void *key, void *value) @@ -177,7 +195,7 @@ int bpf_percpu_array_copy(struct bpf_map *map, void *key, void *value) */ size = round_up(map->value_size, 8); rcu_read_lock(); - pptr = array->pptrs[index]; + pptr = array->pptrs[index & array->index_mask]; for_each_possible_cpu(cpu) { bpf_long_memcpy(value + off, per_cpu_ptr(pptr, cpu), size); off += size; @@ -225,10 +243,11 @@ static int array_map_update_elem(struct bpf_map *map, void *key, void *value, return -EEXIST; if (array->map.map_type == BPF_MAP_TYPE_PERCPU_ARRAY) - memcpy(this_cpu_ptr(array->pptrs[index]), + memcpy(this_cpu_ptr(array->pptrs[index & array->index_mask]), value, map->value_size); else - memcpy(array->value + array->elem_size * index, + memcpy(array->value + + array->elem_size * (index & array->index_mask), value, map->value_size); return 0; } @@ -262,7 +281,7 @@ int bpf_percpu_array_update(struct bpf_map *map, void *key, void *value, */ size = round_up(map->value_size, 8); rcu_read_lock(); - pptr = array->pptrs[index]; + pptr = array->pptrs[index & array->index_mask]; for_each_possible_cpu(cpu) { bpf_long_memcpy(per_cpu_ptr(pptr, cpu), value + off, size); off += size; @@ -613,6 +632,7 @@ static void *array_of_map_lookup_elem(struct bpf_map *map, void *key) static u32 array_of_map_gen_lookup(struct bpf_map *map, struct bpf_insn *insn_buf) { + struct bpf_array *array = container_of(map, struct bpf_array, map); u32 elem_size = round_up(map->value_size, 8); struct bpf_insn *insn = insn_buf; const int ret = BPF_REG_0; @@ -621,7 +641,12 @@ static u32 array_of_map_gen_lookup(struct bpf_map *map, *insn++ = BPF_ALU64_IMM(BPF_ADD, map_ptr, offsetof(struct bpf_array, value)); *insn++ = BPF_LDX_MEM(BPF_W, ret, index, 0); - *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 5); + if (map->unpriv_array) { + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 6); + *insn++ = BPF_ALU32_IMM(BPF_AND, ret, array->index_mask); + } else { + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 5); + } if (is_power_of_2(elem_size)) *insn++ = BPF_ALU64_IMM(BPF_LSH, ret, ilog2(elem_size)); else diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 04b24876cd23..b414d6b2d470 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1729,6 +1729,13 @@ static int check_call(struct bpf_verifier_env *env, int func_id, int insn_idx) err = check_func_arg(env, BPF_REG_2, fn->arg2_type, &meta); if (err) return err; + if (func_id == BPF_FUNC_tail_call) { + if (meta.map_ptr == NULL) { + verbose(env, "verifier bug\n"); + return -EINVAL; + } + env->insn_aux_data[insn_idx].map_ptr = meta.map_ptr; + } err = check_func_arg(env, BPF_REG_3, fn->arg3_type, &meta); if (err) return err; @@ -4456,6 +4463,35 @@ static int fixup_bpf_calls(struct bpf_verifier_env *env) */ insn->imm = 0; insn->code = BPF_JMP | BPF_TAIL_CALL; + + /* instead of changing every JIT dealing with tail_call + * emit two extra insns: + * if (index >= max_entries) goto out; + * index &= array->index_mask; + * to avoid out-of-bounds cpu speculation + */ + map_ptr = env->insn_aux_data[i + delta].map_ptr; + if (map_ptr == BPF_MAP_PTR_POISON) { + verbose(env, "tail_call obusing map_ptr\n"); + return -EINVAL; + } + if (!map_ptr->unpriv_array) + continue; + insn_buf[0] = BPF_JMP_IMM(BPF_JGE, BPF_REG_3, + map_ptr->max_entries, 2); + insn_buf[1] = BPF_ALU32_IMM(BPF_AND, BPF_REG_3, + container_of(map_ptr, + struct bpf_array, + map)->index_mask); + insn_buf[2] = *insn; + cnt = 3; + new_prog = bpf_patch_insn_data(env, i + delta, insn_buf, cnt); + if (!new_prog) + return -ENOMEM; + + delta += cnt - 1; + env->prog = prog = new_prog; + insn = new_prog->insnsi + i + delta; continue; } -- cgit v1.2.3 From 290af86629b25ffd1ed6232c4e9107da031705cb Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Tue, 9 Jan 2018 10:04:29 -0800 Subject: bpf: introduce BPF_JIT_ALWAYS_ON config The BPF interpreter has been used as part of the spectre 2 attack CVE-2017-5715. A quote from goolge project zero blog: "At this point, it would normally be necessary to locate gadgets in the host kernel code that can be used to actually leak data by reading from an attacker-controlled location, shifting and masking the result appropriately and then using the result of that as offset to an attacker-controlled address for a load. But piecing gadgets together and figuring out which ones work in a speculation context seems annoying. So instead, we decided to use the eBPF interpreter, which is built into the host kernel - while there is no legitimate way to invoke it from inside a VM, the presence of the code in the host kernel's text section is sufficient to make it usable for the attack, just like with ordinary ROP gadgets." To make attacker job harder introduce BPF_JIT_ALWAYS_ON config option that removes interpreter from the kernel in favor of JIT-only mode. So far eBPF JIT is supported by: x64, arm64, arm32, sparc64, s390, powerpc64, mips64 The start of JITed program is randomized and code page is marked as read-only. In addition "constant blinding" can be turned on with net.core.bpf_jit_harden v2->v3: - move __bpf_prog_ret0 under ifdef (Daniel) v1->v2: - fix init order, test_bpf and cBPF (Daniel's feedback) - fix offloaded bpf (Jakub's feedback) - add 'return 0' dummy in case something can invoke prog->bpf_func - retarget bpf tree. For bpf-next the patch would need one extra hunk. It will be sent when the trees are merged back to net-next Considered doing: int bpf_jit_enable __read_mostly = BPF_EBPF_JIT_DEFAULT; but it seems better to land the patch as-is and in bpf-next remove bpf_jit_enable global variable from all JITs, consolidate in one place and remove this jit_init() function. Signed-off-by: Alexei Starovoitov Signed-off-by: Daniel Borkmann --- kernel/bpf/core.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) (limited to 'kernel') diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index 86b50aa26ee8..51ec2dda7f08 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -767,6 +767,7 @@ noinline u64 __bpf_call_base(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) } EXPORT_SYMBOL_GPL(__bpf_call_base); +#ifndef CONFIG_BPF_JIT_ALWAYS_ON /** * __bpf_prog_run - run eBPF program on a given context * @ctx: is the data we are operating on @@ -1317,6 +1318,14 @@ EVAL6(PROG_NAME_LIST, 224, 256, 288, 320, 352, 384) EVAL4(PROG_NAME_LIST, 416, 448, 480, 512) }; +#else +static unsigned int __bpf_prog_ret0(const void *ctx, + const struct bpf_insn *insn) +{ + return 0; +} +#endif + bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp) { @@ -1364,9 +1373,13 @@ static int bpf_check_tail_call(const struct bpf_prog *fp) */ struct bpf_prog *bpf_prog_select_runtime(struct bpf_prog *fp, int *err) { +#ifndef CONFIG_BPF_JIT_ALWAYS_ON u32 stack_depth = max_t(u32, fp->aux->stack_depth, 1); fp->bpf_func = interpreters[(round_up(stack_depth, 32) / 32) - 1]; +#else + fp->bpf_func = __bpf_prog_ret0; +#endif /* eBPF JITs can rewrite the program in case constant * blinding is active. However, in case of error during @@ -1376,6 +1389,12 @@ struct bpf_prog *bpf_prog_select_runtime(struct bpf_prog *fp, int *err) */ if (!bpf_prog_is_dev_bound(fp->aux)) { fp = bpf_int_jit_compile(fp); +#ifdef CONFIG_BPF_JIT_ALWAYS_ON + if (!fp->jited) { + *err = -ENOTSUPP; + return fp; + } +#endif } else { *err = bpf_prog_offload_compile(fp); if (*err) -- cgit v1.2.3