diff options
| author | NeilBrown <neilb@suse.de> | 2014-09-24 11:28:32 +1000 |
|---|---|---|
| committer | Trond Myklebust <trond.myklebust@primarydata.com> | 2014-09-25 08:25:38 -0400 |
| commit | 353db7966288a2f18da22438aeec2b4862c0b241 (patch) | |
| tree | 959b0fe5a246747af6db55bb7652d4d35b738ba8 /fs/nfs/file.c | |
| parent | 9590544694becc64f4874963dbfc4b4d391024b7 (diff) | |
| download | linux-353db7966288a2f18da22438aeec2b4862c0b241.tar.gz linux-353db7966288a2f18da22438aeec2b4862c0b241.tar.bz2 linux-353db7966288a2f18da22438aeec2b4862c0b241.zip | |
NFS: avoid waiting at all in nfs_release_page when congested.
If nfs_release_page() is called on a sequence of pages which are all
in the same file which is blocked on COMMIT, each page could
contribute a 1 second delay which could be come excessive. I have
seen delays of as much as 208 seconds.
To keep the delay to one second, mark the bdi as write-congested
if the commit didn't finished. Once it does finish, the
write-congested flag will be cleared by nfs_commit_release_pages().
With this, the longest total delay in try_to_free_pages that I have
seen is under 3 seconds. With no waiting in nfs_release_page at all
I have seen delays of nearly 1.5 seconds.
Signed-off-by: NeilBrown <neilb@suse.de>
Acked-by: Jeff Layton <jlayton@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Diffstat (limited to 'fs/nfs/file.c')
| -rw-r--r-- | fs/nfs/file.c | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 325df0aeab05..24832d040eb8 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -477,7 +477,8 @@ static int nfs_release_page(struct page *page, gfp_t gfp) /* Always try to initiate a 'commit' if relevant, but only * wait for it if __GFP_WAIT is set and the calling process is - * allowed to block. Even then, only wait 1 second. + * allowed to block. Even then, only wait 1 second and only + * if the 'bdi' is not congested. * Waiting indefinitely can cause deadlocks when the NFS * server is on this machine, and there is no particular need * to wait extensively here. A short wait has the benefit @@ -488,9 +489,13 @@ static int nfs_release_page(struct page *page, gfp_t gfp) nfs_commit_inode(mapping->host, 0); if ((gfp & __GFP_WAIT) && !current_is_kswapd() && - !(current->flags & PF_FSTRANS)) { + !(current->flags & PF_FSTRANS) && + !bdi_write_congested(&nfss->backing_dev_info)) { wait_on_page_bit_killable_timeout(page, PG_private, HZ); + if (PagePrivate(page)) + set_bdi_congested(&nfss->backing_dev_info, + BLK_RW_ASYNC); } } /* If PagePrivate() is set, then the page is not freeable */ |