linux-kernel - Re: [RFC PATCH 1/1] smp: Add tracepoints for functions called with smp_call

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 for Android: free password hash cracker in your pocket

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

Date:   Thu, 06 Apr 2023 05:42:13 -0300
From:   Leonardo Brás <leobras@...hat.com>
To:     Sebastian Andrzej Siewior <bigeasy@...utronix.de>
Cc:     Steven Rostedt <rostedt@...dmis.org>,
        Masami Hiramatsu <mhiramat@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Yury Norov <yury.norov@...il.com>,
        "Paul E. McKenney" <paulmck@...nel.org>,
        "Peter Zijlstra (Intel)" <peterz@...radead.org>,
        Nadav Amit <namit@...are.com>,
        Zhen Lei <thunder.leizhen@...wei.com>,
        Chen Zhongjin <chenzhongjin@...wei.com>,
        linux-kernel@...r.kernel.org, linux-trace-kernel@...r.kernel.org,
        Marcelo Tosatti <mtosatti@...hat.com>
Subject: Re: [RFC PATCH 1/1] smp: Add tracepoints for functions called with
 smp_call_function*()

On Thu, 2023-04-06 at 10:15 +0200, Sebastian Andrzej Siewior wrote:
> On 2023-04-06 04:57:18 [-0300], Leonardo Bras wrote:
> > When running RT workloads in isolated CPUs, many cases of deadline misses
> > are caused by remote CPU requests such as smp_call_function*().
> > 
> > For those cases, having the names of those functions running around the
> > deadline miss moment could help finding a target for the next improvements.
> > 
> > Add tracepoints for acquiring the funtion name & argument before entry and
> > after exitting the called function.
> > 
> > Signed-off-by: Leonardo Bras <leobras@...hat.com>
> 
> Check
> 	https://lore.kernel.org/20230307143558.294354-1-vschneid@redhat.com
> 
> Sebastian
> 

Hello Sebastian, thanks for the heads up!

IIUC Valentin is adding/improving tracepoints that are collected in the
requesting CPU, at the moment of scheduling the IPI, which are also useful.

But on my scenario, would be nice to actually have the information on the time
window in which the requested function is running in the requested CPU. 

This would allow us to check how many and which IPIs took too much time to
complete, and caused the RT deadline to miss. 

(I noticed there is an include/trace/events/ipi.h that could be used instead of
creating a new include/trace/events/smp.h, what is your opinion on this ? )

Please let me know if I am missing something.

Thank you!
Leo