forked from TencentOS/TencentOS-kernel
bea022cf22
[ Upstream commit 1beaef29c34154ccdcb3f1ae557f6883eda18840 ]
For memcpy, the source pages are memset to zero only when --cycles is
used. This leads to wildly different results with or without --cycles,
since all sources pages are likely to be mapped to the same zero page
without explicit writes.
Before this fix:
$ export cmd="./perf stat -e LLC-loads -- ./perf bench \
mem memcpy -s 1024MB -l 100 -f default"
$ $cmd
2,935,826 LLC-loads
3.821677452 seconds time elapsed
$ $cmd --cycles
217,533,436 LLC-loads
8.616725985 seconds time elapsed
After this fix:
$ $cmd
214,459,686 LLC-loads
8.674301124 seconds time elapsed
$ $cmd --cycles
214,758,651 LLC-loads
8.644480006 seconds time elapsed
Fixes: 47b5757bac03c338 ("perf bench mem: Move boilerplate memory allocation to the infrastructure")
Signed-off-by: Vincent Whitchurch <vincent.whitchurch@axis.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: kernel@axis.com
Link: http://lore.kernel.org/lkml/20200810133404.30829-1-vincent.whitchurch@axis.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
||
|---|---|---|
| .. | ||
| Build | ||
| bench.h | ||
| epoll-ctl.c | ||
| epoll-wait.c | ||
| futex-hash.c | ||
| futex-lock-pi.c | ||
| futex-requeue.c | ||
| futex-wake-parallel.c | ||
| futex-wake.c | ||
| futex.h | ||
| mem-functions.c | ||
| mem-memcpy-arch.h | ||
| mem-memcpy-x86-64-asm-def.h | ||
| mem-memcpy-x86-64-asm.S | ||
| mem-memcpy-x86-64-lib.c | ||
| mem-memset-arch.h | ||
| mem-memset-x86-64-asm-def.h | ||
| mem-memset-x86-64-asm.S | ||
| numa.c | ||
| sched-messaging.c | ||
| sched-pipe.c | ||