linux-kernel - Re: [for-next][PATCH 05/31] tracing: Have existing event_command.parse() implementations use helpers

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <388b9922-4231-6e34-1305-f0b439d9d07c@kernel.org>
Date:   Thu, 13 Jan 2022 18:03:07 +0100
From:   Daniel Bristot de Oliveira <bristot@...nel.org>
To:     Steven Rostedt <rostedt@...dmis.org>,
        Tom Zanussi <zanussi@...nel.org>
Cc:     Ingo Molnar <mingo@...nel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        linux-kernel@...r.kernel.org
Subject: Re: [for-next][PATCH 05/31] tracing: Have existing
 event_command.parse() implementations use helpers

Hi

On 1/11/22 18:30, Steven Rostedt wrote:
> From: Tom Zanussi <zanussi@...nel.org>
> 
> Simplify the existing event_command.parse() implementations by having
> them make use of the helper functions previously introduced.

While testing rtla with all for-next changes, I noticed this patch breaks:

# rtla osnoise hist -d 5

Before this patch, osnoise hist is able to enable histograms and collect data.
After this patch it does not work.

The event I am creating the histogram is osnoise:sample_threshold, and the
trigger command is:

hist:keys=duration.buckets=1000,common_cpu:vals=hitcount:sort=hitcount:size=2048

I did some debug, and found that the histogram is working. The problem is that,
to read the histogram I pause it to have consistent data:

in tools/tracing/rtla/osnoise_hist.c:
osnoise_read_trace_hist() {
 [...]
        tracefs_hist_pause(tool->trace.inst, data->trace_hist);

        content = tracefs_event_file_read(tool->trace.inst, "osnoise",
                                          "sample_threshold",
                                          "hist", NULL);
 [...]
}

and, as far as I got, after this patch, pausing the histogram makes it to clear
up. If I comment the "tracefs_hist_pause" line, "rtla osnoise hist" start
working back again.

Thoughts?

-- Daniel