linux-kernel - Re: [PATCH 4/4] rtla: Report missed event count

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] [day] [month] [year] [list]

Message-ID: <3568976320e31a90e36f7d2a17ec7d42e1f2f34a.camel@redhat.com>
Date: Fri, 24 Jan 2025 10:24:29 +0100
From: Gabriele Monaco <gmonaco@...hat.com>
To: Tomas Glozar <tglozar@...hat.com>, Steven Rostedt <rostedt@...dmis.org>
Cc: linux-trace-kernel@...r.kernel.org, linux-kernel@...r.kernel.org, John
 Kacur <jkacur@...hat.com>, Luis Goncalves <lgoncalv@...hat.com>
Subject: Re: [PATCH 4/4] rtla: Report missed event count

On Thu, 2025-01-23 at 15:23 +0100, Tomas Glozar wrote:
> Print how many events were missed by trace buffer overflow in the
> main
> instance at the end of the run (for hist) or during the run (for
> top).
> 
> Signed-off-by: Tomas Glozar <tglozar@...hat.com>
> ---

The patchset seems to work as expected, I get the number of dropped
events and the percentage, of course if I let it run too long, I get
unknown number.
Tested on a 128 cores arm64 box with -p100, known to miss events:

run for 5s
4611870 (24.12%) events missed, results might not be accurate

run for 1m
104653605 (45.44%) events missed, results might not be accurate

run for 5m
539717695 (46.85%) events missed, results might not be accurate

run for 10m
1098569155 (47.68%) events missed, results might not be accurate

run for 1h
6749574550 (48.83%) events missed, results might not be accurate

Tested-by: Gabriele Monaco <gmonaco@...hat.com>