[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <63f2abf4-9a3e-a655-dbb7-301ad42a78f3@linux.intel.com>
Date: Tue, 9 Jan 2018 10:02:32 +0800
From: "Jin, Yao" <yao.jin@...ux.intel.com>
To: Arnaldo Carvalho de Melo <acme@...nel.org>
Cc: jolsa@...nel.org, peterz@...radead.org, mingo@...hat.com,
alexander.shishkin@...ux.intel.com, Linux-kernel@...r.kernel.org,
ak@...ux.intel.com, kan.liang@...el.com, yao.jin@...el.com
Subject: Re: [PATCH v7 5/6] perf report: support time percent and multiple
time ranges
On 1/8/2018 11:04 PM, Arnaldo Carvalho de Melo wrote:
> One more thing to consider:
>
> When you use:
>
> perf report --time 10%/1
>
> it will do what is asked but there is no indication of what percentage
> ranges are in place, it would be nice to have this in the first line in
> the TUI, right after this:
>
> Samples: 128 of event 'cycles:ppp', Event count (approx.): 21386169 (10%/1)
>
> - Arnaldo
>
Hi Arnaldo,
Thanks for merging this feature and thanks so much for providing
comments for how to improve this feature in next step.
Please let me summarize what the follow-up patch needs to do:
1. Improve the error message, something like:
"HINT: use 'perf record --foobar' to record the first/last sample
timestamps in the perf.data file header or enable build-id processing."
2. Currently it uses magic number 10 to limit the number of time slices.
But it's not good. The best thing would be to have no such limitations,
i.e. to lift this limitation from the code and docs.
3. 'perf report --time 10%' should be equivalent to 'perf report --time
10%/1'. Or at least say something like:
"percent slices need an index to specify which one is wanted" when
you notice a % in the --time string, etc.
4. Add an indication of what percentage ranges are being used. For example,
Samples: 128 of event 'cycles:ppp', Event count (approx.): 21386169 (10%/1)
I will develop a follow-up patchset to improve this feature.
Thanks
Jin Yao
Powered by blists - more mailing lists