public inbox for linuxppc-dev@ozlabs.org 
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead•org>
To: Pan Xinhui <xinhui@linux•vnet.ibm.com>
Cc: linux-kernel@vger•kernel.org, linuxppc-dev@lists•ozlabs.org,
	benh@kernel•crashing.org, paulus@samba•org, mpe@ellerman•id.au,
	boqun.feng@gmail•com, paulmck@linux•vnet.ibm.com,
	tglx@linutronix•de
Subject: Re: [PATCH V3] powerpc: Implement {cmp}xchg for u8 and u16
Date: Thu, 21 Apr 2016 18:13:54 +0200	[thread overview]
Message-ID: <20160421161354.GI3430@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <5718F32B.3050409@linux.vnet.ibm.com>

On Thu, Apr 21, 2016 at 11:35:07PM +0800, Pan Xinhui wrote:
> yes, you are right. more load/store will be done in C code.
> However such xchg_u8/u16 is just used by qspinlock now. and I did not see any performance regression.
> So just wrote in C, for simple. :)

Which is fine; but worthy of a note in your Changelog.

> Of course I have done xchg tests.
> we run code just like xchg((u8*)&v, j++); in several threads.
> and the result is,
> [  768.374264] use time[1550072]ns in xchg_u8_asm
> [  768.377102] use time[2826802]ns in xchg_u8_c
> 
> I think this is because there is one more load in C.
> If possible, we can move such code in asm-generic/.

So I'm not actually _that_ familiar with the PPC LL/SC implementation;
but there are things a CPU can do to optimize these loops.

For example, a CPU might choose to not release the exclusive hold of the
line for a number of cycles, except when it passes SC or an interrupt
happens. This way there's a smaller chance the SC fails and inhibits
forward progress.

By doing the modification outside of the LL/SC you loose such
advantages.

And yes, doing a !exclusive load prior to the exclusive load leads to an
even bigger window where the data can get changed out from under you.

  parent reply	other threads:[~2016-04-21 16:14 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-19  6:29 [PATCH V2] powerpc: Implement {cmp}xchg for u8 and u16 Pan Xinhui
2016-04-19  9:18 ` Boqun Feng
2016-04-20  3:39   ` Pan Xinhui
2016-04-20 13:24 ` [PATCH V3] " Pan Xinhui
2016-04-20 14:24   ` Peter Zijlstra
2016-04-21 15:35     ` Pan Xinhui
2016-04-21 15:52       ` Boqun Feng
2016-04-22  1:59         ` Pan Xinhui
2016-04-22  3:16           ` Boqun Feng
2016-04-21 16:13       ` Peter Zijlstra [this message]
2016-04-25 10:10         ` Pan Xinhui
2016-04-25 15:37           ` Peter Zijlstra
2016-04-26 11:35             ` Pan Xinhui
2016-04-27  9:16   ` [PATCH V4] " Pan Xinhui
2016-04-27 13:58     ` Boqun Feng
2016-04-27 14:16       ` Boqun Feng
2016-04-27 14:50       ` Boqun Feng
2016-04-27 14:59         ` Boqun Feng
2016-04-28 10:21           ` Pan Xinhui
2016-04-28  7:59     ` Peter Zijlstra
2016-04-28 10:21       ` Pan Xinhui
2016-11-25  0:04     ` [V4] " Michael Ellerman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20160421161354.GI3430@twins.programming.kicks-ass.net \
    --to=peterz@infradead$(echo .)org \
    --cc=benh@kernel$(echo .)crashing.org \
    --cc=boqun.feng@gmail$(echo .)com \
    --cc=linux-kernel@vger$(echo .)kernel.org \
    --cc=linuxppc-dev@lists$(echo .)ozlabs.org \
    --cc=mpe@ellerman$(echo .)id.au \
    --cc=paulmck@linux$(echo .)vnet.ibm.com \
    --cc=paulus@samba$(echo .)org \
    --cc=tglx@linutronix$(echo .)de \
    --cc=xinhui@linux$(echo .)vnet.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox