[lkml]   [2012]   [May]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
    SubjectRe: [BUG] perf stat: useless output for raw events with new event parser

    I will try to experiment with: cpu/event=INST_RETIRED:ANY:c=1/
    But for now with libpfm4 as it is. Then will try to come up with an
    external file format that encapsulates what I already have in the
    C-table. There is more than just what the hardware exports. You
    have to encode certain constraints, e.g., how umasks can be
    combined. I don't even use the same format for all arch. On
    X86 the AMD and Intel formats are different.

    On Mon, Apr 23, 2012 at 12:45 PM, Stephane Eranian <> wrote:
    > Hi,
    > With the new event parser, one can express raw events field by field:
    > $ perf stat -e cpu/event=0x3c,umask=0x0/,cpu/event=0xc5,umask=0x0/ noploop 1
    > The problem with this is that the output of perf stat becomes useless:
    > $ perf stat -e cpu/event=0x3c,umask=0x0/,cpu/event=0xc5,umask=0x0/ noploop 1
    > noploop for 1 seconds
    >  Performance counter stats for 'noploop 1':
    >        2395038678 pmu
    >            10787 pmu
    >                       ^^^^^^
    >       1.000802603 seconds time elapsed
    > We lose the event names or encoding completely. Now for all events
    > expressed via this
    > new syntax , all we see is 'pmu'. That is pretty useless. It is hard
    > to decrypt the results
    > without some serious scripting.
    > Not sure how to solve this given how the parser works. This looks like
    > a regression to me.
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to
    More majordomo info at
    Please read the FAQ at

     \ /
      Last update: 2012-05-02 13:41    [W:0.032 / U:32.084 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site