s390/kernel: add system calls for PCI memory access

Add the new __NR_s390_pci_mmio_write and __NR_s390_pci_mmio_read
system calls to allow user space applications to access device PCI I/O
memory pages on s390x platform.

[ Martin Schwidefsky: some code beautification ]

Signed-off-by: Alexey Ishchuk <aishchuk@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

7 files changed, 127 insertions, 2 deletions

diff --git a/arch/s390/include/uapi/asm/unistd.h b/arch/s390/include/uapi/asm/unistd.h
index 4197c89c52d4..2b446cf0cc65 100644
--- a/arch/s390/include/uapi/asm/unistd.h
+++ b/arch/s390/include/uapi/asm/unistd.h
@@ -287,7 +287,9 @@
 #define __NR_getrandom		349
 #define __NR_memfd_create	350
 #define __NR_bpf		351
-#define NR_syscalls 352
+#define __NR_s390_pci_mmio_write	352
+#define __NR_s390_pci_mmio_read		353
+#define NR_syscalls 354
 
 /* 
  * There are some system calls that are not present on 64 bit, some
diff --git a/arch/s390/kernel/compat_wrapper.c b/arch/s390/kernel/compat_wrapper.c
index c4f7a3d655b8..d7fa2f0f1425 100644
--- a/arch/s390/kernel/compat_wrapper.c
+++ b/arch/s390/kernel/compat_wrapper.c
@@ -218,3 +218,5 @@ COMPAT_SYSCALL_WRAP3(seccomp, unsigned int, op, unsigned int, flags, const char
 COMPAT_SYSCALL_WRAP3(getrandom, char __user *, buf, size_t, count, unsigned int, flags)
 COMPAT_SYSCALL_WRAP2(memfd_create, const char __user *, uname, unsigned int, flags)
 COMPAT_SYSCALL_WRAP3(bpf, int, cmd, union bpf_attr *, attr, unsigned int, size);
+COMPAT_SYSCALL_WRAP3(s390_pci_mmio_write, const unsigned long, mmio_addr, const void __user *, user_buffer, const size_t, length);
+COMPAT_SYSCALL_WRAP3(s390_pci_mmio_read, const unsigned long, mmio_addr, void __user *, user_buffer, const size_t, length);
diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h
index 0554b9771c9f..8e61393c8275 100644
--- a/arch/s390/kernel/entry.h
+++ b/arch/s390/kernel/entry.h
@@ -74,4 +74,6 @@ struct old_sigaction;
 long sys_s390_personality(unsigned int personality);
 long sys_s390_runtime_instr(int command, int signum);
 
+long sys_s390_pci_mmio_write(unsigned long, const void __user *, size_t);
+long sys_s390_pci_mmio_read(unsigned long, void __user *, size_t);
 #endif /* _ENTRY_H */
diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S
index 9f7087fd58de..a2987243bc76 100644
--- a/arch/s390/kernel/syscalls.S
+++ b/arch/s390/kernel/syscalls.S
@@ -360,3 +360,5 @@ SYSCALL(sys_seccomp,sys_seccomp,compat_sys_seccomp)
 SYSCALL(sys_getrandom,sys_getrandom,compat_sys_getrandom)
 SYSCALL(sys_memfd_create,sys_memfd_create,compat_sys_memfd_create) /* 350 */
 SYSCALL(sys_bpf,sys_bpf,compat_sys_bpf)
+SYSCALL(sys_ni_syscall,sys_s390_pci_mmio_write,compat_sys_s390_pci_mmio_write)
+SYSCALL(sys_ni_syscall,sys_s390_pci_mmio_read,compat_sys_s390_pci_mmio_read)
diff --git a/arch/s390/pci/Makefile b/arch/s390/pci/Makefile
index a9e1dc4ae442..805d8b29193a 100644
--- a/arch/s390/pci/Makefile
+++ b/arch/s390/pci/Makefile
@@ -3,4 +3,4 @@
 #
 
 obj-$(CONFIG_PCI)	+= pci.o pci_dma.o pci_clp.o pci_sysfs.o \
-			   pci_event.o pci_debug.o pci_insn.o
+			   pci_event.o pci_debug.o pci_insn.o pci_mmio.o
diff --git a/arch/s390/pci/pci_mmio.c b/arch/s390/pci/pci_mmio.c
new file mode 100644
index 000000000000..62c5ea6d8682
--- /dev/null
+++ b/arch/s390/pci/pci_mmio.c
@@ -0,0 +1,115 @@
+/*
+ * Access to PCI I/O memory from user space programs.
+ *
+ * Copyright IBM Corp. 2014
+ * Author(s): Alexey Ishchuk <aishchuk@linux.vnet.ibm.com>
+ */
+#include <linux/kernel.h>
+#include <linux/syscalls.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/errno.h>
+#include <linux/pci.h>
+
+static long get_pfn(unsigned long user_addr, unsigned long access,
+		    unsigned long *pfn)
+{
+	struct vm_area_struct *vma;
+	long ret;
+
+	down_read(&current->mm->mmap_sem);
+	ret = -EINVAL;
+	vma = find_vma(current->mm, user_addr);
+	if (!vma)
+		goto out;
+	ret = -EACCES;
+	if (!(vma->vm_flags & access))
+		goto out;
+	ret = follow_pfn(vma, user_addr, pfn);
+out:
+	up_read(&current->mm->mmap_sem);
+	return ret;
+}
+
+SYSCALL_DEFINE3(s390_pci_mmio_write, unsigned long, mmio_addr,
+		const void __user *, user_buffer, size_t, length)
+{
+	u8 local_buf[64];
+	void __iomem *io_addr;
+	void *buf;
+	unsigned long pfn;
+	long ret;
+
+	if (!zpci_is_enabled())
+		return -ENODEV;
+
+	if (length <= 0 || PAGE_SIZE - (mmio_addr & ~PAGE_MASK) < length)
+		return -EINVAL;
+	if (length > 64) {
+		buf = kmalloc(length, GFP_KERNEL);
+		if (!buf)
+			return -ENOMEM;
+	} else
+		buf = local_buf;
+
+	ret = get_pfn(mmio_addr, VM_WRITE, &pfn);
+	if (ret)
+		goto out;
+	io_addr = (void *)((pfn << PAGE_SHIFT) | (mmio_addr & ~PAGE_MASK));
+
+	ret = -EFAULT;
+	if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE)
+		goto out;
+
+	if (copy_from_user(buf, user_buffer, length))
+		goto out;
+
+	memcpy_toio(io_addr, buf, length);
+	ret = 0;
+out:
+	if (buf != local_buf)
+		kfree(buf);
+	return ret;
+}
+
+SYSCALL_DEFINE3(s390_pci_mmio_read, unsigned long, mmio_addr,
+		void __user *, user_buffer, size_t, length)
+{
+	u8 local_buf[64];
+	void __iomem *io_addr;
+	void *buf;
+	unsigned long pfn;
+	long ret;
+
+	if (!zpci_is_enabled())
+		return -ENODEV;
+
+	if (length <= 0 || PAGE_SIZE - (mmio_addr & ~PAGE_MASK) < length)
+		return -EINVAL;
+	if (length > 64) {
+		buf = kmalloc(length, GFP_KERNEL);
+		if (!buf)
+			return -ENOMEM;
+	} else
+		buf = local_buf;
+
+	ret = get_pfn(mmio_addr, VM_READ, &pfn);
+	if (ret)
+		goto out;
+	io_addr = (void *)((pfn << PAGE_SHIFT) | (mmio_addr & ~PAGE_MASK));
+
+	ret = -EFAULT;
+	if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE)
+		goto out;
+
+	memcpy_fromio(buf, io_addr, length);
+
+	if (copy_to_user(user_buffer, buf, length))
+		goto out;
+
+	ret = 0;
+out:
+	if (buf != local_buf)
+		kfree(buf);
+	return ret;
+}
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 02aa4185b17e..61eea02b53f5 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -169,6 +169,8 @@ cond_syscall(ppc_rtas);
 cond_syscall(sys_spu_run);
 cond_syscall(sys_spu_create);
 cond_syscall(sys_subpage_prot);
+cond_syscall(sys_s390_pci_mmio_read);
+cond_syscall(sys_s390_pci_mmio_write);
 
 /* mmu depending weak syscall entries */
 cond_syscall(sys_mprotect);