From: Trond Myklebust <trond.myklebust@primarydata.com>
To: linux-nfs@vger.kernel.org
Subject: [PATCH v4 04/17] NFS: Kill NFS_INO_NFS_INO_FLUSHING: it is a performance killer
Date: Thu, 23 Jun 2016 16:00:36 -0400 [thread overview]
Message-ID: <1466712049-5326-4-git-send-email-trond.myklebust@primarydata.com> (raw)
In-Reply-To: <1466712049-5326-3-git-send-email-trond.myklebust@primarydata.com>
filemap_datawrite() and friends already deal just fine with livelock.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
---
fs/nfs/file.c | 8 --------
fs/nfs/nfstrace.h | 1 -
fs/nfs/write.c | 11 -----------
include/linux/nfs_fs.h | 1 -
4 files changed, 21 deletions(-)
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 2d39d9f9da7d..29d7477a62e8 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -360,14 +360,6 @@ static int nfs_write_begin(struct file *file, struct address_space *mapping,
start:
/*
- * Prevent starvation issues if someone is doing a consistency
- * sync-to-disk
- */
- ret = wait_on_bit_action(&NFS_I(mapping->host)->flags, NFS_INO_FLUSHING,
- nfs_wait_bit_killable, TASK_KILLABLE);
- if (ret)
- return ret;
- /*
* Wait for O_DIRECT to complete
*/
inode_dio_wait(mapping->host);
diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h
index 0b9e5cc9a747..fe80a1c26340 100644
--- a/fs/nfs/nfstrace.h
+++ b/fs/nfs/nfstrace.h
@@ -37,7 +37,6 @@
{ 1 << NFS_INO_ADVISE_RDPLUS, "ADVISE_RDPLUS" }, \
{ 1 << NFS_INO_STALE, "STALE" }, \
{ 1 << NFS_INO_INVALIDATING, "INVALIDATING" }, \
- { 1 << NFS_INO_FLUSHING, "FLUSHING" }, \
{ 1 << NFS_INO_FSCACHE, "FSCACHE" }, \
{ 1 << NFS_INO_LAYOUTCOMMIT, "NEED_LAYOUTCOMMIT" }, \
{ 1 << NFS_INO_LAYOUTCOMMITTING, "LAYOUTCOMMIT" })
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index e1c74d3db64d..980d44f3a84c 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -657,16 +657,9 @@ static int nfs_writepages_callback(struct page *page, struct writeback_control *
int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
{
struct inode *inode = mapping->host;
- unsigned long *bitlock = &NFS_I(inode)->flags;
struct nfs_pageio_descriptor pgio;
int err;
- /* Stop dirtying of new pages while we sync */
- err = wait_on_bit_lock_action(bitlock, NFS_INO_FLUSHING,
- nfs_wait_bit_killable, TASK_KILLABLE);
- if (err)
- goto out_err;
-
nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGES);
nfs_pageio_init_write(&pgio, inode, wb_priority(wbc), false,
@@ -674,10 +667,6 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
err = write_cache_pages(mapping, wbc, nfs_writepages_callback, &pgio);
nfs_pageio_complete(&pgio);
- clear_bit_unlock(NFS_INO_FLUSHING, bitlock);
- smp_mb__after_atomic();
- wake_up_bit(bitlock, NFS_INO_FLUSHING);
-
if (err < 0)
goto out_err;
err = pgio.pg_error;
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index d71278c3c5bd..120dd04b553c 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -205,7 +205,6 @@ struct nfs_inode {
#define NFS_INO_STALE (1) /* possible stale inode */
#define NFS_INO_ACL_LRU_SET (2) /* Inode is on the LRU list */
#define NFS_INO_INVALIDATING (3) /* inode is being invalidated */
-#define NFS_INO_FLUSHING (4) /* inode is flushing out data */
#define NFS_INO_FSCACHE (5) /* inode can be cached by FS-Cache */
#define NFS_INO_FSCACHE_LOCK (6) /* FS-Cache cookie management lock */
#define NFS_INO_LAYOUTCOMMIT (9) /* layoutcommit required */
--
2.7.4
next prev parent reply other threads:[~2016-06-23 20:01 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-23 20:00 [PATCH v4 01/17] NFS: Don't flush caches for a getattr that races with writeback Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 02/17] NFS: Cache access checks more aggressively Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 03/17] NFS: Cache aggressively when file is open for writing Trond Myklebust
2016-06-23 20:00 ` Trond Myklebust [this message]
2016-06-23 20:00 ` [PATCH v4 05/17] NFS: writepage of a single page should not be synchronous Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 06/17] NFS: Don't hold the inode lock across fsync() Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 07/17] NFS: Don't call COMMIT in ->releasepage() Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 08/17] NFS: Fix O_DIRECT verifier problems Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 09/17] NFS: Ensure we reset the write verifier 'committed' value on resend Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 10/17] NFS: Remove racy size manipulations in O_DIRECT Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 11/17] NFS Cleanup: move call to generic_write_checks() into fs/nfs/direct.c Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 12/17] NFS: Move buffered I/O locking into nfs_file_write() Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 13/17] NFS: Do not serialise O_DIRECT reads and writes Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 14/17] NFS: Cleanup nfs_direct_complete() Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 15/17] NFS: Remove redundant waits for O_DIRECT in fsync() and write_begin() Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 16/17] NFS: Remove unused function nfs_revalidate_mapping_protected() Trond Myklebust
2016-06-23 20:00 ` [PATCH v4 17/17] NFS nfs_vm_page_mkwrite: Don't freeze me, Bro Trond Myklebust
2016-06-28 12:12 ` [PATCH v4 01/17] NFS: Don't flush caches for a getattr that races with writeback Christoph Hellwig
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=1466712049-5326-4-git-send-email-trond.myklebust@primarydata.com \
--to=trond.myklebust@primarydata.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).