Messages in this thread | | | Date | Wed, 18 Jan 2012 13:27:34 +0000 | From | "Jan Beulich" <> | Subject | [PATCH 0/4] perf/x86-64: "bench mem" improvements |
| |
In the context of building up confidence in the proposed memset() and to be proposed memcpy() adjustments (to make them support regions of 4Gb and more), a number of limitations were found in the current "bench mem" functionality specific to x86-64. Those are being addressed in this series:
1: make "default" memcpy() selection actually use glibc's implementation 2: also allow measuring alternative memcpy implementations 3: also allow measuring memset() 4: allow passing an iteration count to "bench mem mem{cpy,set}"
Signed-off-by: Jan Beulich <jbeulich@suse.com>
| |