diff options
| author | Andrew Gallatin <gallatin@FreeBSD.org> | 2022-12-14 19:34:07 +0000 |
|---|---|---|
| committer | Andrew Gallatin <gallatin@FreeBSD.org> | 2022-12-14 19:34:07 +0000 |
| commit | 1cac76c93fb7f627fd9e304cbd99e8c8a2b8fce8 (patch) | |
| tree | 478ab936c759d5b0ac77c1c066319635d63eeb70 /sys/amd64/include | |
| parent | c4a4b2633d975bd0813afca6b8e23ead29d80e82 (diff) | |
| download | src-1cac76c93fb7f627fd9e304cbd99e8c8a2b8fce8.tar.gz src-1cac76c93fb7f627fd9e304cbd99e8c8a2b8fce8.zip | |
vm: reduce lock contention when processing vm batchqueues
Rather than waiting until the batchqueue is full to acquire the lock &
process the queue, we now start trying to acquire the lock using trylocks
when the batchqueue is 1/2 full. This removes almost all contention on the
vm pagequeue mutex for for our busy sendfile() based web workload.
It also greadly reduces the amount of time a network driver ithread
remains blocked on a mutex, and eliminates some packet drops under
heavy load.
So that the system does not loose the benefit of processing large
batchqueues, I've doubled the size of the batchqueues. This way, when
there is no contention, we process the same batch size as before.
This has been run for several months on a busy Netflix server, as well
as on my personal desktop.
Reviewed by: markj
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D37305
Diffstat (limited to 'sys/amd64/include')
| -rw-r--r-- | sys/amd64/include/vmparam.h | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/sys/amd64/include/vmparam.h b/sys/amd64/include/vmparam.h index fc88296f754c..205848489644 100644 --- a/sys/amd64/include/vmparam.h +++ b/sys/amd64/include/vmparam.h @@ -293,7 +293,7 @@ * Use a fairly large batch size since we expect amd64 systems to have lots of * memory. */ -#define VM_BATCHQUEUE_SIZE 31 +#define VM_BATCHQUEUE_SIZE 63 /* * The pmap can create non-transparent large page mappings. |