[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1fb2247a-debe-cdc4-2fec-76526dd07dab@gmail.com>
Date: Sat, 24 Nov 2018 15:17:12 -0700
From: David Ahern <dsahern@...il.com>
To: Alexei Starovoitov <alexei.starovoitov@...il.com>,
Song Liu <songliubraving@...com>
Cc: netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
ast@...nel.org, daniel@...earbox.net, acme@...nel.org,
peterz@...radead.org, kernel-team@...com
Subject: Re: [PATCH perf,bpf 5/5] perf util: generate bpf_prog_info_event for
short living bpf programs
On 11/21/18 3:11 PM, Alexei Starovoitov wrote:
> On Wed, Nov 21, 2018 at 11:55:02AM -0800, Song Liu wrote:
>> This patch enables perf-record to listen to bpf_event and generate
>> bpf_prog_info_event for bpf programs loaded and unloaded during
>> perf-record run.
>>
>> To minimize latency between bpf_event and following bpf calls, separate
>> mmap with watermark of 1 is created to process these vip events. Then
>> a separate dummy event is attached to the special mmap. A separate thread
>> is used to only poll bpf events.
>>
>> By default, perf-record will listen to bpf_event. Option no-bpf-event is
>> added in case the user would opt out.
>
> I think only default perf-record for sampling should include it.
> perf record -e for tracepoints, kprobes and anything else should imply no-bpf-event.
>
It should default off for those, but as I have said before users should
be able to add bpf load and unload events to the mix of tracepoints,
probes, etc they are monitoring.
Powered by blists - more mailing lists