MIME-Version: 1.0
In-Reply-To: <553977AA.9040802@zytor.com>
References: <1429792491-5978-1-git-send-email-dvlasenk@redhat.com>
 <CA+55aFwDP1QxvHs9wKE9ZpjAzSHhzc+YQS1Mhwt=oYLrB+Rp1A@mail.gmail.com>
 <CAMzpN2j3K9B4UDPB8suz15PjoDY2aZBr8bvEZ5r4ag5FLgCVrw@mail.gmail.com>
 <CA+55aFxre3Jk-CG+mBs6Bp9a1J=riCQaA3dwUTFRph1OsCMSuA@mail.gmail.com>
 <CALCETrX_RCGH0C4GJ0Zxm1emcasMbZ5s29HCfFy=GrEC9w28+A@mail.gmail.com>
 <CAK1hOcPp3Xav4iR2mZ4oq4zB5XY+ZNuEds32Rf0Z3fB8A7h1PQ@mail.gmail.com>
 <20150423211015.GS28327@pd.tnic> <55396617.4020303@zytor.com>
 <CALCETrVX_O+vu+GJXJKJa7w83DnKn0nrjp2_9xtNdhEJW79XQA@mail.gmail.com>
 <553972D5.2070104@zytor.com> <CALCETrV2t1ynQ0KpRv71+_1wqUg6d2=k-guqHatrEHJLeJ9d6Q@mail.gmail.com>
 <553977AA.9040802@zytor.com>
From: Andy Lutomirski <luto@amacapital.net>
Date: Thu, 23 Apr 2015 15:55:26 -0700
Message-ID: <CALCETrXQnb6jPuXkjXmg5yYCKOEyCL3hZ_sDrD7=Af1idr5zBw@mail.gmail.com>
Subject: Re: [PATCH] x86/asm/entry/32: Restore %ss before SYSRETL if necessary
To: "H. Peter Anvin" <hpa@zytor.com>
Cc: Borislav Petkov <bp@alien8.de>, Denys Vlasenko <vda.linux@googlemail.com>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        Brian Gerst <brgerst@gmail.com>, Denys Vlasenko <dvlasenk@redhat.com>,
        Ingo Molnar <mingo@kernel.org>, Steven Rostedt <rostedt@goodmis.org>,
        Oleg Nesterov <oleg@redhat.com>,
        Frederic Weisbecker <fweisbec@gmail.com>,
        Alexei Starovoitov <ast@plumgrid.com>, Will Drewry <wad@chromium.org>,
        Kees Cook <keescook@chromium.org>,
        "the arch/x86 maintainers" <x86@kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset=UTF-8
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1286
Lines: 34

On Thu, Apr 23, 2015 at 3:52 PM, H. Peter Anvin <hpa@zytor.com> wrote:
> On 04/23/2015 03:38 PM, Andy Lutomirski wrote:
>>>
>>> Because there are way more sysrets than context switches, and Linux is
>>> particularly sensitive to system call latency, by design.
>>
>
> Just to clarify: why would Linux be more sensitive to system call by
> design?  It enables much simpler APIs and avoids hacks like sending down
> a syscall task list (which was genuinely proposed at one point.)  If
> kernel entry/exit is too expensive, then the APIs get more complex
> because they *have* to do everything in the smallest number of system calls.
>

It's a matter of the ratio, right?  One cycle of syscall overhead
saved is worth some number of context switch cycles added, and the
ratio probably varies by workload.

If we do syscall, two context switches, and sysret, then we wouldn't
have been better off fixing it on sysret.  But maybe most workloads
still prefer the fixup on context switch.

--Andy


-- 
Andy Lutomirski
AMA Capital Management, LLC
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/