Re: [PATCH 01/11] perf stat: Fix AGGR_CORE segfault on multi-socket system

From: Arnaldo Carvalho de Melo
Date: Wed Oct 01 2014 - 13:08:37 EST


Em Wed, Oct 01, 2014 at 03:36:22PM +0100, Matt Fleming escreveu:
> From: Matt Fleming <matt.fleming@xxxxxxxxx>
>
> When printing the stats associated with a counter in AGGR_MODE mode, the
> 'cpu' argument represents an encoded socket and core_id, not a 'cpu'.
> Using it as an index into the any of the *_stats[MAX_NR_CPUS] arrays
> generates a SIGSEGV if the encoded socket id is non-zero.
>
> Follow the AGGR_GLOBAL case and reset the cpu index to 0.

So, is this still needed after applying Andi's patch:

perf stat: Fix --per-core on multi socket systems

https://git.kernel.org/cgit/linux/kernel/git/acme/linux.git/commit/?h=perf/core&id=da88c7f78d842a6938d9adde6af87a2ce262051d

Or should this be applied on top of that one, after fixing up the
conflicts?

- Arnaldo

> Cc: Jiri Olsa <jolsa@xxxxxxxxxx>
> Cc: Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>
> Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> Signed-off-by: Matt Fleming <matt.fleming@xxxxxxxxx>
> ---
> tools/perf/builtin-stat.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
> index b22c62f80078..53bd6b729498 100644
> --- a/tools/perf/builtin-stat.c
> +++ b/tools/perf/builtin-stat.c
> @@ -965,7 +965,7 @@ static void abs_printout(int id, int nr, struct perf_evsel *evsel, double avg)
>
> aggr_printout(evsel, id, nr);
>
> - if (aggr_mode == AGGR_GLOBAL)
> + if (aggr_mode == AGGR_GLOBAL || aggr_mode == AGGR_CORE)
> cpu = 0;
>
> fprintf(output, fmt, avg, csv_sep);
> --
> 1.9.3
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/