[PATCH v2 1/2] msync: support syncing a small part of the file

From: Paolo Bonzini
Date: Fri Jun 15 2012 - 11:13:14 EST


msync does not need to flush changes to the entire file, even with MS_SYNC.
Instead, it can use vfs_fsync_range to only synchronize a part of the file.
This is part of the specification; expecting msync to synchronize all the
file would take a very creative interpretation of the manual page as well
as the specification.

In addition, not all metadata has to be synced; msync is closer to
fdatasync than it is to fsync. So, pass 1 to vfs_fsync_range.

Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Cc: Hugh Dickins <hughd@xxxxxxxxxx>
Cc: Chris Friesen <chris.friesen@xxxxxxxxxxx>
Signed-off-by: Paolo Bonzini <pbonzini@xxxxxxxxxx>
---
v1->v2: fixed off-by-one in vall to vfs_fsync_range.

mm/msync.c | 13 ++++++++++---
1 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/mm/msync.c b/mm/msync.c
index 632df45..505fe99 100644
--- a/mm/msync.c
+++ b/mm/msync.c
@@ -15,7 +15,7 @@
#include <linux/sched.h>

/*
- * MS_SYNC syncs the entire file - including mappings.
+ * MS_SYNC syncs the specified range - including mappings.
*
* MS_ASYNC does not start I/O (it used to, up to 2.5.67).
* Nor does it marks the relevant pages dirty (it used to up to 2.6.17).
@@ -58,6 +58,8 @@ SYSCALL_DEFINE3(msync, unsigned long, start, size_t, len, int, flags)
vma = find_vma(mm, start);
for (;;) {
struct file *file;
+ unsigned long next;
+ loff_t file_offset;

/* Still start < end. */
error = -ENOMEM;
@@ -77,18 +79,23 @@ SYSCALL_DEFINE3(msync, unsigned long, start, size_t, len, int, flags)
goto out_unlock;
}
file = vma->vm_file;
- start = vma->vm_end;
+ next = min(end, vma->vm_end);
if ((flags & MS_SYNC) && file &&
(vma->vm_flags & VM_SHARED)) {
+ file_offset = vma->vm_pgoff * PAGE_SIZE;
get_file(file);
up_read(&mm->mmap_sem);
- error = vfs_fsync(file, 0);
+ error = vfs_fsync_range(file,
+ start - vma->vm_start + file_offset,
+ next - vma->vm_start + file_offset - 1, 1);
fput(file);
+ start = next;
if (error || start >= end)
goto out;
down_read(&mm->mmap_sem);
vma = find_vma(mm, start);
} else {
+ start = next;
if (start >= end) {
error = 0;
goto out_unlock;
--
1.7.1



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/