Michael Ellerman <m...@ellerman.id.au> writes: > On Tue, 2015-01-20 at 17:05 +0530, Aneesh Kumar K.V wrote: >> This enables us to understand how many hash fault we are taking >> when running benchmarks. >> >> For ex: >> -bash-4.2# ./perf stat -e powerpc:hash_fault -e page-faults >> /tmp/ebizzy.ppc64 -S 30 -P -n 1000 >> ... >> >> Performance counter stats for '/tmp/ebizzy.ppc64 -S 30 -P -n 1000': >> >> 1,10,04,075 powerpc:hash_fault >> 1,10,03,429 page-faults >> >> 30.865978991 seconds time elapsed > > Looks good. > > Can you attach some test results that show it's not hurting performance when > it's disabled. >
ebizzy with -S 30 -t 1 -P gave 13627 records/s -> Without patch 13546 records/s -> With patch with tracepoint disabled 12408 records/s -> With patch with tracepoint enabled. perf stat gave the below data for the above run. 22,38,284 page-faults 22,38,291 powerpc:hash_fault I also used random_access_bench that Anton wrote, it actually create lots of hash fault. A simple run gives. (random_access_bench -o load -g -i -t 10 16G) 1,888 page-faults 2,64,283 powerpc:hash_fault random_access_bench gave: 1435.979 MB/s -> Without patch 1435.29 MB/s -> With patch with tracepoint disabled 1434.75 MB/s -> With patch with tracepoint enabled. -aneesh _______________________________________________ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev