Techwizz/kernel-fxtec-pro1x
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Btrfs: kmap_atomic(KM_USER0) is safe for btrfs_readpage_end_io_hook

Browse source 
None of the checksum verification code schedules, so we can use the faster
kmap_atomic

Signed-off-by: Chris Mason <chris.mason@oracle.com>
This commit is contained in:
Chris Mason 2009-01-07 09:48:51 -05:00
parent cc7172defc
commit 9ab86c8e01
1 changed files with 3 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
fs/btrfs/inode.c
View file

@ -1748,7 +1748,7 @@ static int btrfs_readpage_end_io_hook(struct page *page, u64 start, u64 end,
} else {
ret = get_state_private(io_tree, start, &private);
}
kaddr = kmap(page);
kaddr = kmap_atomic(page, KM_USER0);
if (ret)
goto zeroit;
@ -1757,7 +1757,7 @@ static int btrfs_readpage_end_io_hook(struct page *page, u64 start, u64 end,
if (csum != private)
goto zeroit;
kunmap(page);
kunmap_atomic(kaddr, KM_USER0);
good:
/* if the io failure tree for this inode is non-empty,
* check to see if we've recovered from a failed IO
@ -1772,7 +1772,7 @@ static int btrfs_readpage_end_io_hook(struct page *page, u64 start, u64 end,
(unsigned long long)private);
memset(kaddr + offset, 1, end - start + 1);
flush_dcache_page(page);
kunmap(page);
kunmap_atomic(kaddr, KM_USER0);
if (private == 0)
return 0;
return -EIO;