From: Trond Myklebust <trond.myklebust@primarydata.com>
To: Benjamin Coddington <bcodding@redhat.com>
Cc: linux-nfs@vger.kernel.org
Subject: [PATCH v2 1/3] NFS: Fix a performance regression in readdir
Date: Fri, 2 Dec 2016 09:21:33 -0500 [thread overview]
Message-ID: <20161202142135.11515-2-trond.myklebust@primarydata.com> (raw)
In-Reply-To: <20161202142135.11515-1-trond.myklebust@primarydata.com>
Ben Coddington reports that commit 311324ad1713, by adding the function
nfs_dir_mapping_need_revalidate() that checks page cache validity on
each call to nfs_readdir() causes a performance regression when
the directory is being modified.
If the directory is changing while we're iterating through the directory,
POSIX does not require us to invalidate the page cache unless the user
calls rewinddir(). However, we still do want to ensure that we use
readdirplus in order to avoid a load of stat() calls when the user
is doing an 'ls -l' workload.
The fix should be to invalidate the page cache immediately when we're
setting the NFS_INO_ADVISE_RDPLUS bit.
Reported-by: Benjamin Coddington <bcodding@redhat.com>
Fixes: 311324ad1713 ("NFS: Be more aggressive in using readdirplus...")
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
fs/nfs/dir.c | 15 ++-------------
1 file changed, 2 insertions(+), 13 deletions(-)
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index f1ecfcc632eb..9220bc7b9cd7 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -477,7 +477,7 @@ void nfs_force_use_readdirplus(struct inode *dir)
{
if (!list_empty(&NFS_I(dir)->open_files)) {
nfs_advise_use_readdirplus(dir);
- nfs_zap_mapping(dir, dir->i_mapping);
+ invalidate_mapping_pages(dir->i_mapping, 0, -1);
}
}
@@ -886,17 +886,6 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc)
goto out;
}
-static bool nfs_dir_mapping_need_revalidate(struct inode *dir)
-{
- struct nfs_inode *nfsi = NFS_I(dir);
-
- if (nfs_attribute_cache_expired(dir))
- return true;
- if (nfsi->cache_validity & NFS_INO_INVALID_DATA)
- return true;
- return false;
-}
-
/* The file offset position represents the dirent entry number. A
last cookie cache takes care of the common case of reading the
whole directory.
@@ -928,7 +917,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx)
desc->decode = NFS_PROTO(inode)->decode_dirent;
desc->plus = nfs_use_readdirplus(inode, ctx) ? 1 : 0;
- if (ctx->pos == 0 || nfs_dir_mapping_need_revalidate(inode))
+ if (ctx->pos == 0 || nfs_attribute_cache_expired(inode))
res = nfs_revalidate_mapping(inode, file->f_mapping);
if (res < 0)
goto out;
--
2.9.3
next prev parent reply other threads:[~2016-12-02 14:21 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-12-02 14:21 [PATCH v2 0/3] Readdirplus performance improvements for 4.10 Trond Myklebust
2016-12-02 14:21 ` Trond Myklebust [this message]
2016-12-02 14:21 ` [PATCH v2 2/3] NFS: Be more targeted about readdirplus use when doing lookup/revalidation Trond Myklebust
2016-12-02 14:21 ` [PATCH v2 3/3] NFS: Allow getattr to also report readdirplus cache hits Trond Myklebust
2016-12-02 15:42 ` [PATCH v2 0/3] Readdirplus performance improvements for 4.10 Benjamin Coddington
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20161202142135.11515-2-trond.myklebust@primarydata.com \
--to=trond.myklebust@primarydata.com \
--cc=bcodding@redhat.com \
--cc=linux-nfs@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).