lkml.org 
[lkml]   [2020]   [Aug]   [20]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    Patch in this message
    /
    From
    Subject[PATCH 5.7 185/204] perf bench mem: Always memset source before memcpy
    Date
    From: Vincent Whitchurch <vincent.whitchurch@axis.com>

    [ 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>
    ---
    tools/perf/bench/mem-functions.c | 21 +++++++++++----------
    1 file changed, 11 insertions(+), 10 deletions(-)

    diff --git a/tools/perf/bench/mem-functions.c b/tools/perf/bench/mem-functions.c
    index 9235b76501be8..19d45c377ac18 100644
    --- a/tools/perf/bench/mem-functions.c
    +++ b/tools/perf/bench/mem-functions.c
    @@ -223,12 +223,8 @@ static int bench_mem_common(int argc, const char **argv, struct bench_mem_info *
    return 0;
    }

    -static u64 do_memcpy_cycles(const struct function *r, size_t size, void *src, void *dst)
    +static void memcpy_prefault(memcpy_t fn, size_t size, void *src, void *dst)
    {
    - u64 cycle_start = 0ULL, cycle_end = 0ULL;
    - memcpy_t fn = r->fn.memcpy;
    - int i;
    -
    /* Make sure to always prefault zero pages even if MMAP_THRESH is crossed: */
    memset(src, 0, size);

    @@ -237,6 +233,15 @@ static u64 do_memcpy_cycles(const struct function *r, size_t size, void *src, vo
    * to not measure page fault overhead:
    */
    fn(dst, src, size);
    +}
    +
    +static u64 do_memcpy_cycles(const struct function *r, size_t size, void *src, void *dst)
    +{
    + u64 cycle_start = 0ULL, cycle_end = 0ULL;
    + memcpy_t fn = r->fn.memcpy;
    + int i;
    +
    + memcpy_prefault(fn, size, src, dst);

    cycle_start = get_cycles();
    for (i = 0; i < nr_loops; ++i)
    @@ -252,11 +257,7 @@ static double do_memcpy_gettimeofday(const struct function *r, size_t size, void
    memcpy_t fn = r->fn.memcpy;
    int i;

    - /*
    - * We prefault the freshly allocated memory range here,
    - * to not measure page fault overhead:
    - */
    - fn(dst, src, size);
    + memcpy_prefault(fn, size, src, dst);

    BUG_ON(gettimeofday(&tv_start, NULL));
    for (i = 0; i < nr_loops; ++i)
    --
    2.25.1


    \
     
     \ /
      Last update: 2020-08-20 11:46    [W:3.081 / U:0.020 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site