Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1217114

Re: [PATCH 0/7] x86 vdso32 cleanups

From Andy Lutomirski <luto@amacapital.net>
Newsgroups linux.kernel
Subject Re: [PATCH 0/7] x86 vdso32 cleanups
Date 2015-09-02 00:00 +0200
Message-ID <q41KW-2cN-19@gated-at.bofh.it> (permalink)
References (1 earlier) <q2R1g-6Wa-9@gated-at.bofh.it> <q3ib7-3Zh-9@gated-at.bofh.it> <q3nua-39B-3@gated-at.bofh.it> <q3IIi-eh-29@gated-at.bofh.it> <q3IIi-eh-27@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On Mon, Aug 31, 2015 at 6:37 PM, Andy Lutomirski <luto@amacapital.net> wrote:
> I got random errors from perf kvm, but I think I found at least part
> of the issue.  The two irqs_disabled() calls in common.c are kind of
> expensive.  I should disable them on non-lockdep kernels.
>
> The context tracking hooks are also too expensive, even when disabled.
> I should do something to optimize those.  Hello, static keys?  This
> doesn't affect syscalls, though.
>
> With context tracking off and the irqs_disabled checks commented out,
> we're probably doing well enough.  We can always tweak the C code and
> aggressively force inlining if we want a few cycles back.

Currently, a compat AT_SYSINFO syscall (getpid) is 171 cycles for me.
With my patches, it's 196 cycles, so it's really not that bad.  The
impact will probably be slightly worse on native 32-bit because of
increased register pressure and because one of the micro-optimizations
I threw in are 64-bit specific.  We could probably tune the C code a
bit more to get a few of the cycles back.

On the flip side, the rewrite is *far* faster in some of the slow path
cases because the slow path no longer forces IRET.

On 32-bit, there's the added benefit that we could drop asmlinkage
from the syscall bodies on top of the rewrite.

--Andy

>
> --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/

Back to linux.kernel | Previous | NextPrevious in thread | Find similar | Unroll thread


Thread

[PATCH 0/7] x86 vdso32 cleanups Brian Gerst <brgerst@gmail.com> - 2015-08-29 17:30 +0200
  [PATCH 3/7] x86/vdso32: Remove unused vdso-fakesections.c Brian Gerst <brgerst@gmail.com> - 2015-08-29 17:30 +0200
    Re: [PATCH 3/7] x86/vdso32: Remove unused vdso-fakesections.c Andy Lutomirski <luto@amacapital.net> - 2015-08-30 18:50 +0200
  [PATCH 4/7] x86/vdso32: Build single vdso32 image Brian Gerst <brgerst@gmail.com> - 2015-08-29 17:30 +0200
  [PATCH 6/7] x86/vdso32/xen: Move VDSO_NOTE_NONEGSEG_BIT define Brian Gerst <brgerst@gmail.com> - 2015-08-29 17:30 +0200
  [PATCH 5/7] x86/vdso: Merge 32-bit and 64-bit source files Brian Gerst <brgerst@gmail.com> - 2015-08-29 17:30 +0200
  Re: [PATCH 0/7] x86 vdso32 cleanups Andy Lutomirski <luto@amacapital.net> - 2015-08-29 18:20 +0200
    Re: [PATCH 0/7] x86 vdso32 cleanups Brian Gerst <brgerst@gmail.com> - 2015-08-30 23:20 +0200
      Re: [PATCH 0/7] x86 vdso32 cleanups Andy Lutomirski <luto@amacapital.net> - 2015-08-31 05:00 +0200
        Re: [PATCH 0/7] x86 vdso32 cleanups Andy Lutomirski <luto@amacapital.net> - 2015-09-01 03:40 +0200
          Re: [PATCH 0/7] x86 vdso32 cleanups Andy Lutomirski <luto@amacapital.net> - 2015-09-02 00:00 +0200

csiph-web