Implementation of AVX2 optimized Fletcher-4

[mirror_zfs.git] / man / man5 / zfs-module-parameters.5

diff --git a/man/man5/zfs-module-parameters.5 b/man/man5/zfs-module-parameters.5
index 58b8742ee51a26289b5450604eff8026aa740c2d..664a806c0c0128fab8871ac8176619b598fcae6f 100644 (file)
--- a/man/man5/zfs-module-parameters.5
+++ b/man/man5/zfs-module-parameters.5
@@ -830,6 +830,23 @@ Start syncing out a transaction group if there is at least this much dirty data.
 Default value: \fB67,108,864\fR.
 .RE
 
+.sp
+.ne 2
+.na
+\fBzfs_fletcher_4_impl\fR (string)
+.ad
+.RS 12n
+Select a fletcher 4 implementation.
+.sp
+Supported selectors are: \fBfastest\fR, \fBscalar\fR, and \fBavx2\fR when
+AVX2 is supported by the processor.  If multiple implementations of fletcher 4
+are available the \fBfastest\fR will be chosen using a micro benchmark.
+Selecting \fBscalar\fR results in the original CPU based calculation being
+used, \fBavx2\fR uses the AVX2 vector instructions to compute a fletcher 4.
+.sp
+Default value: \fBfastest\fR.
+.RE
+
 .sp
 .ne 2
 .na