linux-kernel - Re: [PATCH 1/2] IB/hfi1: Use preempt_{dis,en}able

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: <20171011134300.GS28623@kernel.org>
Date:   Wed, 11 Oct 2017 10:43:00 -0300
From:   Arnaldo Carvalho de Melo <acme@...nel.org>
To:     Sebastian Andrzej Siewior <sebastian.siewior@...utronix.de>
Cc:     Dennis Dalessandro <dennis.dalessandro@...el.com>,
        linux-rt-users@...r.kernel.org, linux-kernel@...r.kernel.org,
        Arnaldo Carvalho de Melo <acme@...hat.com>,
        Clark Williams <williams@...hat.com>,
        Dean Luick <dean.luick@...el.com>,
        Doug Ledford <dledford@...hat.com>,
        Kaike Wan <kaike.wan@...el.com>,
        Leon Romanovsky <leonro@...lanox.com>,
        linux-rdma@...r.kernel.org, Peter Zijlstra <peterz@...radead.org>,
        Sebastian Sanchez <sebastian.sanchez@...el.com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [PATCH 1/2] IB/hfi1: Use preempt_{dis,en}able_nort()

Em Wed, Oct 11, 2017 at 01:03:55PM +0200, Sebastian Andrzej Siewior escreveu:
> On 2017-10-10 16:02:18 [-0300], Arnaldo Carvalho de Melo wrote:
> > 
> > Right, and so far there were no strong objection for this one to be
> > merged on the -rt tree, Sebastian, can you do it please? Adding Dennis'
> > reviewed-by, one of maintainers for this driver, ok?
> 
> I am still curious about the performance improvement that is with this
> preempt disable section compared to without it compared to !PREEMPT
> kernel..
> If that is important then migrate_disable() would do that on RT.

I can try that

> I guess that there were no splat with CONFIG_DEBUG_PREEMPT ?

I haven't tried that

> If that is all okay, please resend the patch with the explanation why
> this preempt_disable() does not matter and I pick it up.

I can just pick the explanation given by the authors and stash it there,
I guess.
 
> > > tree? Anyway, for this patch:
> > > 
> > > Reviewed-by: Dennis Dalessandro <dennis.dalessandro@...el.com>
> 
> Sebastian