linux-kernel - Re: [PATCH 1/2] locking: Add lock contention tracepoints

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]

Date:   Thu, 17 Mar 2022 12:37:41 -0400 (EDT)
From:   Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
To:     rostedt <rostedt@...dmis.org>
Cc:     Namhyung Kim <namhyung@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...nel.org>, Will Deacon <will@...nel.org>,
        Waiman Long <longman@...hat.com>,
        Boqun Feng <boqun.feng@...il.com>,
        linux-kernel <linux-kernel@...r.kernel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Byungchul Park <byungchul.park@....com>,
        paulmck <paulmck@...nel.org>, Arnd Bergmann <arnd@...db.de>,
        Radoslaw Burny <rburny@...gle.com>,
        linux-arch <linux-arch@...r.kernel.org>,
        bpf <bpf@...r.kernel.org>
Subject: Re: [PATCH 1/2] locking: Add lock contention tracepoints

----- On Mar 17, 2022, at 12:07 PM, rostedt rostedt@...dmis.org wrote:

> On Thu, 17 Mar 2022 09:32:55 -0400 (EDT)
> Mathieu Desnoyers <mathieu.desnoyers@...icios.com> wrote:
> 
>> Unless there is a particular reason for using preprocessor defines here, the
>> following form is typically better because it does not pollute the preprocessor
>> defines, e.g.:
>> 
>> enum lock_contention_flags {
>>         LCB_F_SPIN =   1U << 0;
>>         LCB_F_READ =   1U << 1;
>>         LCB_F_WRITE =  1U << 2;
>>         LCB_F_RT =     1U << 3;
>>         LCB_F_PERCPU = 1U << 4;
>> };
> 
> If you do this, then to use the __print_flags(), You'll also need to add:
> 
> TRACE_DEFINE_ENUM(LCB_F_SPIN);
> TRACE_DEFINE_ENUM(LCB_F_READ);
> TRACE_DEFINE_ENUM(LCB_F_WRITE);
> TRACE_DEFINE_ENUM(LCB_F_RT);
> TRACE_DEFINE_ENUM(LCB_F_PERCPU);
> 
> Which does slow down boot up slightly.

So it looks like there is (currently) a good reason for going with the #define.

As a side-discussion, I keep finding it odd that this adds overhead on boot. I suspect
this is also implemented as a linked list which needs to be iterated over at boot-time.

With a few changes to these macros, these linked lists could be turned into arrays,
and thus remove the boot-time overhead.

Thanks,

Mathieu

> 
> -- Steve

-- 
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com