diff options
| author | Kent Overstreet <kent.overstreet@linux.dev> | 2023-11-23 00:42:53 +0100 |
|---|---|---|
| committer | Christian Brauner <brauner@kernel.org> | 2023-11-28 14:08:46 +0100 |
| commit | 71eb6b6b0ba93b1467bccff57b5de746b09113d2 (patch) | |
| tree | 50d8617b7f3cc91869e92507787303e110635722 /security | |
| parent | eventfd: make eventfd_signal{_mask}() void (diff) | |
| download | linux-71eb6b6b0ba93b1467bccff57b5de746b09113d2.tar.xz linux-71eb6b6b0ba93b1467bccff57b5de746b09113d2.zip | |
fs/aio: obey min_nr when doing wakeups
I've been observing workloads where IPIs due to wakeups in
aio_complete() are ~15% of total CPU time in the profile. Most of those
wakeups are unnecessary when completion batching is in use in
io_getevents().
This plumbs min_nr through via the wait eventry, so that aio_complete()
can avoid doing unnecessary wakeups.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Link: https://lore.kernel.org/r/20231122234257.179390-1-kent.overstreet@linux.dev
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Christian Brauner <brauner@kernel.org>
Cc: <linux-aio@kvack.org>
Cc: <linux-fsdevel@vger.kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Diffstat (limited to 'security')
0 files changed, 0 insertions, 0 deletions