linux-kernel - Re: RFC: Petition Intel/AMD to add POPF

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: <0ad8824d-7e4a-6c72-43e0-9ecf7a43f86e@redhat.com>
Date:   Thu, 18 Aug 2016 19:22:16 +0200
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     Denys Vlasenko <dvlasenk@...hat.com>,
        Linus Torvalds <torvalds@...ux-foundation.org>
Cc:     Andy Lutomirski <luto@...capital.net>,
        Sara Sharon <sara.sharon@...el.com>,
        Dan Williams <dan.j.williams@...el.com>,
        Christian König <christian.koenig@....com>,
        Vinod Koul <vinod.koul@...el.com>,
        Alex Deucher <alexander.deucher@....com>,
        Johannes Berg <johannes.berg@...el.com>,
        "Rafael J. Wysocki" <rafael.j.wysocki@...el.com>,
        Andy Lutomirski <luto@...nel.org>,
        the arch/x86 maintainers <x86@...nel.org>,
        Ingo Molnar <mingo@...nel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Adrian Hunter <adrian.hunter@...el.com>
Subject: Re: RFC: Petition Intel/AMD to add POPF_IF insn



On 18/08/2016 14:18, Denys Vlasenko wrote:
> 
> -     2,588,839,023      branches                  #  977.555
> M/sec                    ( +-  0.02% )
> +     2,599,319,615      branches                  # 1046.786
> M/sec                    ( +-  0.04% )
> -         3,620,273      branch-misses             #    0.14% of all
> branches          ( +-  0.67% )
> +         3,577,771      branch-misses             #    0.14% of all
> branches          ( +-  0.69% )
> -       2.648799072 seconds time
> elapsed                                          ( +-  0.24% )
> +       2.487452268 seconds time
> elapsed                                          ( +-  0.31% )
> 
> Good, we run more insns/cycle, as expected. However, a bit more branches.

Can you see where the missed branches are?  Assuming branch misses are
the case where IF=0, perhaps there are a few places that can be changed
to spin_lock/unlock_irq or local_irq_disable/enable.

Paolo