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


Groups > linux.kernel > #1237371 > unrolled thread

Re: [RFC v2 5/7] powerpc: atomic: Implement cmpxchg{,64}_* and atomic{,64}_cmpxchg_* variants

Started byPeter Zijlstra <peterz@infradead.org>
First post2015-10-01 14:30 +0200
Last post2015-10-01 17:20 +0200
Articles 5 — 2 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [RFC v2 5/7] powerpc: atomic: Implement cmpxchg{,64}_* and  atomic{,64}_cmpxchg_* variants Peter Zijlstra <peterz@infradead.org> - 2015-10-01 14:30 +0200
    Re: [RFC v2 5/7] powerpc: atomic: Implement cmpxchg{,64}_* and  atomic{,64}_cmpxchg_* variants Peter Zijlstra <peterz@infradead.org> - 2015-10-01 14:40 +0200
      Re: [RFC v2 5/7] powerpc: atomic: Implement cmpxchg{,64}_* and  atomic{,64}_cmpxchg_* variants "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-10-01 17:20 +0200
        Re: [RFC v2 5/7] powerpc: atomic: Implement cmpxchg{,64}_* and  atomic{,64}_cmpxchg_* variants Peter Zijlstra <peterz@infradead.org> - 2015-10-01 19:20 +0200
      Re: [RFC v2 5/7] powerpc: atomic: Implement cmpxchg{,64}_* and  atomic{,64}_cmpxchg_* variants "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-10-01 17:20 +0200

#1237371 — Re: [RFC v2 5/7] powerpc: atomic: Implement cmpxchg{,64}_* and atomic{,64}_cmpxchg_* variants

FromPeter Zijlstra <peterz@infradead.org>
Date2015-10-01 14:30 +0200
SubjectRe: [RFC v2 5/7] powerpc: atomic: Implement cmpxchg{,64}_* and atomic{,64}_cmpxchg_* variants
Message-ID<qeL9M-3Vj-11@gated-at.bofh.it>
On Wed, Sep 16, 2015 at 11:49:33PM +0800, Boqun Feng wrote:
> Unlike other atomic operation variants, cmpxchg{,64}_acquire and
> atomic{,64}_cmpxchg_acquire don't have acquire semantics if the cmp part
> fails, so we need to implement these using assembly.

I think that is actually expected and documented. That is, a cmpxchg
only implies barriers on success. See:

  ed2de9f74ecb ("locking/Documentation: Clarify failed cmpxchg() memory ordering semantics")
--
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/

[toc] | [next] | [standalone]


#1237382

FromPeter Zijlstra <peterz@infradead.org>
Date2015-10-01 14:40 +0200
Message-ID<qeLjs-46x-17@gated-at.bofh.it>
In reply to#1237371
On Thu, Oct 01, 2015 at 02:27:15PM +0200, Peter Zijlstra wrote:
> On Wed, Sep 16, 2015 at 11:49:33PM +0800, Boqun Feng wrote:
> > Unlike other atomic operation variants, cmpxchg{,64}_acquire and
> > atomic{,64}_cmpxchg_acquire don't have acquire semantics if the cmp part
> > fails, so we need to implement these using assembly.
> 
> I think that is actually expected and documented. That is, a cmpxchg
> only implies barriers on success. See:
> 
>   ed2de9f74ecb ("locking/Documentation: Clarify failed cmpxchg() memory ordering semantics")

Also:

654672d4ba1a6 (Will Deacon     2015-08-06 17:54:37 +0100  28)  * store portion of the operation. Note that a failed cmpxchg_acquire
654672d4ba1a6 (Will Deacon     2015-08-06 17:54:37 +0100  29)  * does -not- imply any memory ordering constraints.
--
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/

[toc] | [prev] | [next] | [standalone]


#1237501

From"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date2015-10-01 17:20 +0200
Message-ID<qeNOj-8mD-25@gated-at.bofh.it>
In reply to#1237382
On Thu, Oct 01, 2015 at 02:36:26PM +0200, Peter Zijlstra wrote:
> On Thu, Oct 01, 2015 at 02:27:15PM +0200, Peter Zijlstra wrote:
> > On Wed, Sep 16, 2015 at 11:49:33PM +0800, Boqun Feng wrote:
> > > Unlike other atomic operation variants, cmpxchg{,64}_acquire and
> > > atomic{,64}_cmpxchg_acquire don't have acquire semantics if the cmp part
> > > fails, so we need to implement these using assembly.
> > 
> > I think that is actually expected and documented. That is, a cmpxchg
> > only implies barriers on success. See:
> > 
> >   ed2de9f74ecb ("locking/Documentation: Clarify failed cmpxchg() memory ordering semantics")
> 
> Also:
> 
> 654672d4ba1a6 (Will Deacon     2015-08-06 17:54:37 +0100  28)  * store portion of the operation. Note that a failed cmpxchg_acquire
> 654672d4ba1a6 (Will Deacon     2015-08-06 17:54:37 +0100  29)  * does -not- imply any memory ordering constraints.

What C11 does is to allow the developer to specify different orderings
on success and failure.  But it is no harder to supply a barrier (if
needed) on the failure path, right?

							Thanx, Paul

--
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/

[toc] | [prev] | [next] | [standalone]


#1237584

FromPeter Zijlstra <peterz@infradead.org>
Date2015-10-01 19:20 +0200
Message-ID<qePGr-2Bk-41@gated-at.bofh.it>
In reply to#1237501
On Thu, Oct 01, 2015 at 08:12:19AM -0700, Paul E. McKenney wrote:

> What C11 does is to allow the developer to specify different orderings
> on success and failure.  But it is no harder to supply a barrier (if
> needed) on the failure path, right?

Quite right.
--
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/

[toc] | [prev] | [next] | [standalone]


#1237504

From"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date2015-10-01 17:20 +0200
Message-ID<qeNOj-8mD-31@gated-at.bofh.it>
In reply to#1237382
On Thu, Oct 01, 2015 at 02:36:26PM +0200, Peter Zijlstra wrote:
> On Thu, Oct 01, 2015 at 02:27:15PM +0200, Peter Zijlstra wrote:
> > On Wed, Sep 16, 2015 at 11:49:33PM +0800, Boqun Feng wrote:
> > > Unlike other atomic operation variants, cmpxchg{,64}_acquire and
> > > atomic{,64}_cmpxchg_acquire don't have acquire semantics if the cmp part
> > > fails, so we need to implement these using assembly.
> > 
> > I think that is actually expected and documented. That is, a cmpxchg
> > only implies barriers on success. See:
> > 
> >   ed2de9f74ecb ("locking/Documentation: Clarify failed cmpxchg() memory ordering semantics")
> 
> Also:
> 
> 654672d4ba1a6 (Will Deacon     2015-08-06 17:54:37 +0100  28)  * store portion of the operation. Note that a failed cmpxchg_acquire
> 654672d4ba1a6 (Will Deacon     2015-08-06 17:54:37 +0100  29)  * does -not- imply any memory ordering constraints.

Agreed, no need for ordering on failed cmpxchg.

							Thanx, Paul

--
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/

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web