xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

NFS: fix subtle change in COMMIT behavior

Browse Source 
Recent work in the pgio layer made it possible for there to be more than one
request per page. This caused a subtle change in commit behavior, because
write.c:nfs_commit_unstable_pages compares the number of *pages* waiting for
writeback against the number of requests on a commit list to choose when to
send a COMMIT in a non-blocking flush.

This is probably hard to hit in normal operation - you have to be using
rsize/wsize < PAGE_SIZE, or pnfs with lots of boundaries that are not page
aligned to have a noticeable change in behavior.

Signed-off-by: Weston Andros Adamson <dros@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
This commit is contained in:
Weston Andros Adamson
2014-11-12 12:08:00 -05:00
committed by Trond Myklebust
parent 6a74c0c940
commit cb1410c71e
5 changed files with 27 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
fs/nfs/pagelist.c
View File

@@ -258,6 +258,7 @@ bool nfs_page_group_sync_on_bit(struct nfs_page *req, unsigned int bit)
static inline void
nfs_page_group_init(struct nfs_page *req, struct nfs_page *prev)
{
struct inode *inode;
WARN_ON_ONCE(prev == req);
if (!prev) {
@@ -276,12 +277,16 @@ nfs_page_group_init(struct nfs_page *req, struct nfs_page *prev)
* nfs_page_group_destroy is called */
kref_get(&req->wb_head->wb_kref);
/* grab extra ref if head request has extra ref from
* the write/commit path to handle handoff between write
* and commit lists */
/* grab extra ref and bump the request count if head request
* has extra ref from the write/commit path to handle handoff
* between write and commit lists. */
if (test_bit(PG_INODE_REF, &prev->wb_head->wb_flags)) {
inode = page_file_mapping(req->wb_page)->host;
set_bit(PG_INODE_REF, &req->wb_flags);
kref_get(&req->wb_kref);
spin_lock(&inode->i_lock);
NFS_I(inode)->nrequests++;
spin_unlock(&inode->i_lock);
}
}
}