From: Will Deacon <will.deacon@arm.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Waiman Long <waiman.long@hpe.com>,
linux-kernel@vger.kernel.org, torvalds@linux-foundation.org,
manfred@colorfullife.com, dave@stgolabs.net,
paulmck@linux.vnet.ibm.com, boqun.feng@gmail.com, tj@kernel.org,
pablo@netfilter.org, kaber@trash.net, davem@davemloft.net,
oleg@redhat.com, netfilter-devel@vger.kernel.org,
sasha.levin@oracle.com, hofrat@osadl.org
Subject: Re: [PATCH -v3 7/8] locking: Move smp_cond_load_acquire() and friends into asm-generic/barrier.h
Date: Wed, 1 Jun 2016 13:00:10 +0100 [thread overview]
Message-ID: <20160601120009.GB355@arm.com> (raw)
In-Reply-To: <20160601093158.GN3190@twins.programming.kicks-ass.net>
On Wed, Jun 01, 2016 at 11:31:58AM +0200, Peter Zijlstra wrote:
> On Tue, May 31, 2016 at 04:01:06PM -0400, Waiman Long wrote:
> > You are doing two READ_ONCE's in the smp_cond_load_acquire loop. Can we
> > change it to do just one READ_ONCE, like
> >
> > --- a/include/asm-generic/barrier.h
> > +++ b/include/asm-generic/barrier.h
> > @@ -229,12 +229,18 @@ do {
> > * value; some architectures can do this in hardware.
> > */
> > #ifndef cmpwait
> > +#define cmpwait(ptr, val) ({ \
> > typeof (ptr) __ptr = (ptr); \
> > + typeof (val) __old = (val); \
> > + typeof (val) __new; \
> > + for (;;) { \
> > + __new = READ_ONCE(*__ptr); \
> > + if (__new != __old) \
> > + break; \
> > cpu_relax(); \
> > + } \
> > + __new; \
> > +})
> > #endif
> >
> > /**
> > @@ -251,12 +257,11 @@ do {
> > #ifndef smp_cond_load_acquire
> > #define smp_cond_load_acquire(ptr, cond_expr) ({ \
> > typeof(ptr) __PTR = (ptr); \
> > + typeof(*ptr) VAL = READ_ONCE(*__PTR); \
> > for (;;) { \
> > if (cond_expr) \
> > break; \
> > + VAL = cmpwait(__PTR, VAL); \
> > } \
> > smp_acquire__after_ctrl_dep(); \
> > VAL; \
>
> Yes, that generates slightly better code, but now that you made me look
> at it, I think we need to kill the cmpwait() in the generic version and
> only keep it for arch versions.
>
> /me ponders...
>
> So cmpwait() as implemented here has strict semantics; but arch
> implementations as previously proposed have less strict semantics; and
> the use here follows that less strict variant.
>
> The difference being that the arch implementations of cmpwait can have
> false positives (ie. return early, without a changed value)
> smp_cond_load_acquire() can deal with these false positives seeing how
> its in a loop and does its own (more specific) comparison.
>
> Exposing cmpwait(), with the documented semantics, means that arch
> versions need an additional loop inside to match these strict semantics,
> or we need to weaken the cmpwait() semantics, at which point I'm not
> entirely sure its worth keeping as a generic primitive...
>
> Hmm, so if we can find a use for the weaker cmpwait() outside of
> smp_cond_load_acquire() I think we can make a case for keeping it, and
> looking at qspinlock.h there's two sites we can replace cpu_relax() with
> it.
>
> Will, since ARM64 seems to want to use this, does the below make sense
> to you?
Not especially -- I was going to override smp_cond_load_acquire anyway
because I want to build it using cmpwait_acquire and get rid of the
smp_acquire__after_ctrl_dep trick, which is likely slower on arm64.
So I'd be happier nuking cmpwait from the generic interfaces and using
smp_cond_load_acquire everywhere, if that's possible.
Will
next prev parent reply other threads:[~2016-06-01 11:59 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-31 9:41 [PATCH -v3 0/8] spin_unlock_wait borkage and assorted bits Peter Zijlstra
2016-05-31 9:41 ` [PATCH -v3 1/8] locking: Replace smp_cond_acquire with smp_cond_load_acquire Peter Zijlstra
2016-05-31 9:41 ` [PATCH -v3 2/8] locking: Introduce cmpwait() Peter Zijlstra
2016-05-31 9:41 ` [PATCH -v3 3/8] locking: Introduce smp_acquire__after_ctrl_dep Peter Zijlstra
2016-06-01 13:52 ` Boqun Feng
2016-06-01 16:22 ` Peter Zijlstra
2016-06-01 23:19 ` Boqun Feng
2016-05-31 9:41 ` [PATCH -v3 4/8] locking, arch: Update spin_unlock_wait() Peter Zijlstra
2016-06-01 11:24 ` Will Deacon
2016-06-01 11:37 ` Peter Zijlstra
2016-05-31 9:41 ` [PATCH -v3 5/8] locking: Update spin_unlock_wait users Peter Zijlstra
2016-05-31 9:41 ` [PATCH -v3 6/8] locking,netfilter: Fix nf_conntrack_lock() Peter Zijlstra
2016-05-31 9:41 ` [PATCH -v3 7/8] locking: Move smp_cond_load_acquire() and friends into asm-generic/barrier.h Peter Zijlstra
2016-05-31 20:01 ` Waiman Long
2016-06-01 9:31 ` Peter Zijlstra
2016-06-01 12:00 ` Will Deacon [this message]
2016-06-01 12:06 ` Peter Zijlstra
2016-06-01 12:13 ` Will Deacon
2016-06-01 12:45 ` Peter Zijlstra
2016-06-01 14:07 ` Will Deacon
2016-06-01 17:13 ` Peter Zijlstra
2016-06-01 16:53 ` Waiman Long
2016-05-31 9:41 ` [PATCH -v3 8/8] locking, tile: Provide TILE specific smp_acquire__after_ctrl_dep Peter Zijlstra
2016-05-31 15:32 ` Chris Metcalf
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=20160601120009.GB355@arm.com \
--to=will.deacon@arm.com \
--cc=boqun.feng@gmail.com \
--cc=dave@stgolabs.net \
--cc=davem@davemloft.net \
--cc=hofrat@osadl.org \
--cc=kaber@trash.net \
--cc=linux-kernel@vger.kernel.org \
--cc=manfred@colorfullife.com \
--cc=netfilter-devel@vger.kernel.org \
--cc=oleg@redhat.com \
--cc=pablo@netfilter.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
--cc=sasha.levin@oracle.com \
--cc=tj@kernel.org \
--cc=torvalds@linux-foundation.org \
--cc=waiman.long@hpe.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
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;
as well as URLs for NNTP newsgroup(s).