lkml.org 
[lkml]   [2016]   [Oct]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
SubjectRe: [f2fs-dev] [PATCH 04/28] f2fs: replace a build-time warning with runtime WARN_ON
Date
On Wednesday, October 26, 2016 10:05:00 PM CEST Chao Yu wrote:
> On 2016/10/18 6:05, Arnd Bergmann wrote:
> > gcc is unsure about the use of last_ofs_in_node, which might happen
> > without a prior initialization:
> >
> > fs/f2fs//git/arm-soc/fs/f2fs/data.c: In function ‘f2fs_map_blocks’:
> > fs/f2fs/data.c:799:54: warning: ‘last_ofs_in_node’ may be used uninitialized in this function [-Wmaybe-uninitialized]
> > if (prealloc && dn.ofs_in_node != last_ofs_in_node + 1) {
>
> In each round of dnode block traverse, we will init 'prealloc' and then update
> 'prealloc' and 'last_ofs_in_node' together in below lines of f2fs_map_blocks:
> if (flag == F2FS_GET_BLOCK_PRE_AIO) {
> if (blkaddr == NULL_ADDR) {
> prealloc++;
> last_ofs_in_node = dn.ofs_in_node;
> }
> }
>
> Then in below codes, it is safe to use 'last_ofs_in_node' since we will check
> 'prealloc' firstly, so if 'prealloc' is non-zero, 'last_ofs_in_node' must be valid.
> if (prealloc && dn.ofs_in_node != last_ofs_in_node + 1) {
>
> So I think we should not add WARN_ON there.

Ok, that make sense. Thanks for taking a closer look!

Should we just set last_ofs_in_node to the same value as ofs_in_node
before the loop?

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 9ae194f..14db4b7 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -716,7 +716,7 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map,
}

prealloc = 0;
- ofs_in_node = dn.ofs_in_node;
+ last_ofs_in_node = ofs_in_node = dn.ofs_in_node;
end_offset = ADDRS_PER_PAGE(dn.node_page, inode);

next_block:
Arnd

\
 
 \ /
  Last update: 2016-10-26 16:58    [W:0.137 / U:0.460 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site