diff options
| author | Al Viro <viro@zeniv.linux.org.uk> | 2013-10-20 08:44:39 -0400 |
|---|---|---|
| committer | Al Viro <viro@zeniv.linux.org.uk> | 2013-10-20 08:44:39 -0400 |
| commit | c7314d74fcb089b127ef5753b5263ac8473f33bc (patch) | |
| tree | 0ca6b0cc8afcc0145d5eeaed90469470e60cf173 | |
| parent | 43ae9e3fc70ca0057ae0a24ef5eedff05e3fae06 (diff) | |
| download | lwn-c7314d74fcb089b127ef5753b5263ac8473f33bc.tar.gz lwn-c7314d74fcb089b127ef5753b5263ac8473f33bc.zip | |
nfsd regression since delayed fput()
Background: nfsd v[23] had throughput regression since delayed fput
went in; every read or write ends up doing fput() and we get a pair
of extra context switches out of that (plus quite a bit of work
in queue_work itselfi, apparently). Use of schedule_delayed_work()
gives it a chance to accumulate a bit before we do __fput() on all
of them. I'm not too happy about that solution, but... on at least
one real-world setup it reverts about 10% throughput loss we got from
switch to delayed fput.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
| -rw-r--r-- | fs/file_table.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/fs/file_table.c b/fs/file_table.c index abdd15ad13c9..e900ca518635 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -297,7 +297,7 @@ void flush_delayed_fput(void) delayed_fput(NULL); } -static DECLARE_WORK(delayed_fput_work, delayed_fput); +static DECLARE_DELAYED_WORK(delayed_fput_work, delayed_fput); void fput(struct file *file) { @@ -317,7 +317,7 @@ void fput(struct file *file) } if (llist_add(&file->f_u.fu_llist, &delayed_fput_list)) - schedule_work(&delayed_fput_work); + schedule_delayed_work(&delayed_fput_work, 1); } } |