[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CABPqkBS+WaLfeEox8ZHQhdxz3k45S2qXAPqURecr2V+AKpoHaA@mail.gmail.com>
Date: Wed, 8 Nov 2017 19:42:28 -0800
From: Stephane Eranian <eranian@...gle.com>
To: Andi Kleen <ak@...ux.intel.com>
Cc: LKML <linux-kernel@...r.kernel.org>,
Arnaldo Carvalho de Melo <acme@...hat.com>,
Peter Zijlstra <peterz@...radead.org>, mingo@...e.hu,
"Liang, Kan" <kan.liang@...el.com>, Jiri Olsa <jolsa@...hat.com>
Subject: Re: [PATCH v3 4/5] perf/record: add documentation for sampling skid ip
On Wed, Nov 8, 2017 at 12:56 PM, Andi Kleen <ak@...ux.intel.com> wrote:
>> diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
>> index 5a626ef666c2..3b156fa03c99 100644
>> --- a/tools/perf/Documentation/perf-record.txt
>> +++ b/tools/perf/Documentation/perf-record.txt
>> @@ -57,6 +57,14 @@ OPTIONS
>> FP mode, "dwarf" for DWARF mode, "lbr" for LBR mode and
>> "no" for disable callgraph.
>> - 'stack-size': user stack size for dwarf mode
>> + - 'skid_ip' : boolean, captures the unmodified interrupt instruction pointer
>> + (IP) in each sample. Usually with event-based sampling, the IP
>> + has skid and rarely point to the instruction which caused the
>> + event to overflow. On some architectures, the hardware can eliminate
>> + the skid and perf_events returns it as the IP with precise sampling is
>> + enabled. But for certain measurements, it may be useful to have both
>> + the correct and skid ip. This option enable capturing the skid ip in
>> + additional to the corrected ip. Default is: false
>
> Actually in the scanner you use 'skid-ip' (with minus). Documentation should at least match
> the scanner.
>
I have fixed the documentation. I used - because many other terms and
options use - instead of _.
> _ is usually safer than -
>
I tend to agree with you.
> -Andi
Powered by blists - more mailing lists