summaryrefslogtreecommitdiffstats
path: root/man/man5
diff options
context:
space:
mode:
authorAndrey Vesnovaty <[email protected]>2013-08-28 05:09:25 +0300
committerBrian Behlendorf <[email protected]>2014-04-25 15:29:18 -0700
commit703371d8c734bc2cc6350f1bca014f08245dcc69 (patch)
tree593c5e83bacee874a2fa25b1745a711fd5f30edf /man/man5
parentae16ed992bd0ef5a55b04d9edaaa6456674315f9 (diff)
Evenly distribute the taskq threads across available CPUs
The problem is described in commit aeeb4e0c0ae75b99ebbaa3056f0afc8e12949532. However, instead of disabling the binding to CPU altogether we just keep the last CPU index across calls to taskq_create() and thus achieve even distribution of the taskq threads across all available CPUs. The implementation based on assumption that task queues initialization performed in serial manner. Signed-off-by: Andrey Vesnovaty <[email protected]> Signed-off-by: Andrey Vesnovaty <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #336
Diffstat (limited to 'man/man5')
-rw-r--r--man/man5/spl-module-parameters.512
1 files changed, 12 insertions, 0 deletions
diff --git a/man/man5/spl-module-parameters.5 b/man/man5/spl-module-parameters.5
index 3c134f775..9b351762c 100644
--- a/man/man5/spl-module-parameters.5
+++ b/man/man5/spl-module-parameters.5
@@ -124,3 +124,15 @@ Spin a maximum of N times to acquire lock
.sp
.ne -4
Default value: \fB0\fR.
+.RE
+
+.sp
+.ne 2
+.na
+\fBspl_taskq_thread_bind\fR (int)
+.ad
+.RS 12n
+Bind taskq thread to CPU
+.sp
+Default value: \fB0\fR.
+.RE