LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Waiman Long <waiman.long@hp.com>
Cc: tglx@linutronix.de, mingo@redhat.com, hpa@zytor.com,
	paolo.bonzini@gmail.com, konrad.wilk@oracle.com,
	boris.ostrovsky@oracle.com, paulmck@linux.vnet.ibm.com,
	riel@redhat.com, torvalds@linux-foundation.org,
	raghavendra.kt@linux.vnet.ibm.com, david.vrabel@citrix.com,
	oleg@redhat.com, scott.norton@hp.com, doug.hatch@hp.com,
	linux-arch@vger.kernel.org, x86@kernel.org,
	linux-kernel@vger.kernel.org,
	virtualization@lists.linux-foundation.org,
	xen-devel@lists.xenproject.org, kvm@vger.kernel.org,
	luto@amacapital.net
Subject: Re: [PATCH 8/9] qspinlock: Generic paravirt support
Date: Thu, 2 Apr 2015 19:20:57 +0200	[thread overview]
Message-ID: <20150402172057.GA27490@worktop.programming.kicks-ass.net> (raw)
In-Reply-To: <551D6E2E.1080801@hp.com>

On Thu, Apr 02, 2015 at 12:28:30PM -0400, Waiman Long wrote:
> On 04/01/2015 05:03 PM, Peter Zijlstra wrote:
> >On Wed, Apr 01, 2015 at 03:58:58PM -0400, Waiman Long wrote:
> >>On 04/01/2015 02:48 PM, Peter Zijlstra wrote:
> >>I am sorry that I don't quite get what you mean here. My point is that in
> >>the hashing step, a cpu will need to scan an empty bucket to put the lock
> >>in. In the interim, an previously used bucket before the empty one may get
> >>freed. In the lookup step for that lock, the scanning will stop because of
> >>an empty bucket in front of the target one.
> >Right, that's broken. So we need to do something else to limit the
> >lookup, because without that break, a lookup that needs to iterate the
> >entire array in order to determine -ENOENT, which is expensive.
> >
> >So my alternative proposal is that IFF we can guarantee that every
> >lookup will succeed -- the entry we're looking for is always there, we
> >don't need the break on empty but can probe until we find the entry.
> >This will be bound in cost to the same number if probes we required for
> >insertion and avoids the full array scan.
> >
> >Now I think we can indeed do this, if as said earlier we do not clear
> >the bucket on insert if the cmpxchg succeeds, in that case the unlock
> >will observe _Q_SLOW_VAL and do the lookup, the lookup will then find
> >the entry. And we then need the unlock to clear the entry.
> >_Q_SLOW_VAL
> >Does that explain this? Or should I try again with code?
> 
> OK, I got your proposal now. However, there is still the issue that setting
> the _Q_SLOW_VAL flag and the hash bucket are not atomic wrt each other.

So? They're strictly ordered, that's sufficient. We first hash the lock,
then we set _Q_SLOW_VAL. There's a full memory barrier between them.

> It
> is possible a CPU has set the _Q_SLOW_VAL flag but not yet filling in the
> hash bucket while another one is trying to look for it.

Nope. The unlock side does an xchg() on the locked value first, xchg
also implies a full barrier, so that guarantees that if we observe
_Q_SLOW_VAL we must also observe the hash bucket with the lock value.

> So we need to have
> some kind of synchronization mechanism to let the lookup CPU know when is a
> good time to look up.

No, its all already ordered and working.

pv_wait_head():

	pv_hash()
	/* MB as per cmpxchg */
	cmpxchg(&l->locked, _Q_LOCKED_VAL, _Q_SLOW_VAL);

VS

__pv_queue_spin_unlock():

	if (xchg(&l->locked, 0) != _Q_SLOW_VAL)
		return;

	/* MB as per xchg */
	pv_hash_find(lock);



  reply	other threads:[~2015-04-02 17:22 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-03-16 13:16 [PATCH 0/9] qspinlock stuff -v15 Peter Zijlstra
2015-03-16 13:16 ` [PATCH 1/9] qspinlock: A simple generic 4-byte queue spinlock Peter Zijlstra
2015-03-16 13:16 ` [PATCH 2/9] qspinlock, x86: Enable x86-64 to use " Peter Zijlstra
2015-03-16 13:16 ` [PATCH 3/9] qspinlock: Add pending bit Peter Zijlstra
2015-03-16 13:16 ` [PATCH 4/9] qspinlock: Extract out code snippets for the next patch Peter Zijlstra
2015-03-16 13:16 ` [PATCH 5/9] qspinlock: Optimize for smaller NR_CPUS Peter Zijlstra
2015-03-16 13:16 ` [PATCH 6/9] qspinlock: Use a simple write to grab the lock Peter Zijlstra
2015-03-16 13:16 ` [PATCH 7/9] qspinlock: Revert to test-and-set on hypervisors Peter Zijlstra
2015-03-16 13:16 ` [PATCH 8/9] qspinlock: Generic paravirt support Peter Zijlstra
     [not found]   ` <5509E51D.7040909@hp.com>
2015-03-19 10:12     ` Peter Zijlstra
2015-03-19 12:25       ` Peter Zijlstra
2015-03-19 13:43         ` Peter Zijlstra
2015-03-19 23:25         ` Waiman Long
2015-04-01 16:20         ` Waiman Long
2015-04-01 17:12           ` Peter Zijlstra
2015-04-01 17:42             ` Peter Zijlstra
2015-04-01 18:17               ` Peter Zijlstra
2015-04-01 18:54                 ` Waiman Long
2015-04-01 18:48                   ` Peter Zijlstra
2015-04-01 19:58                     ` Waiman Long
2015-04-01 21:03                       ` Peter Zijlstra
2015-04-02 16:28                         ` Waiman Long
2015-04-02 17:20                           ` Peter Zijlstra [this message]
2015-04-02 19:48                             ` Peter Zijlstra
2015-04-03  3:39                               ` Waiman Long
2015-04-03 13:43                               ` Peter Zijlstra
2015-04-01 20:10             ` Waiman Long
2015-03-16 13:16 ` [PATCH 9/9] qspinlock,x86,kvm: Implement KVM support for paravirt qspinlock Peter Zijlstra
     [not found]   ` <550A3863.2060808@hp.com>
2015-03-19 10:01     ` Peter Zijlstra
2015-03-19 21:08       ` Waiman Long
2015-03-20  7:43         ` Raghavendra K T
2015-03-16 14:08 ` [Xen-devel] [PATCH 0/9] qspinlock stuff -v15 David Vrabel
2015-03-18 20:36 ` Waiman Long
2015-03-19 18:01 ` [Xen-devel] " David Vrabel
2015-03-19 18:32   ` Peter Zijlstra
2015-03-25 19:47 ` Konrad Rzeszutek Wilk
2015-03-26 20:21   ` Peter Zijlstra
2015-03-27 14:07     ` Konrad Rzeszutek Wilk
2015-03-30 16:41       ` Waiman Long
2015-03-30 16:25   ` Waiman Long
2015-03-30 16:29     ` Peter Zijlstra
2015-03-30 16:43       ` Waiman Long
2015-03-27  6:40 ` Raghavendra K T

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=20150402172057.GA27490@worktop.programming.kicks-ass.net \
    --to=peterz@infradead.org \
    --cc=boris.ostrovsky@oracle.com \
    --cc=david.vrabel@citrix.com \
    --cc=doug.hatch@hp.com \
    --cc=hpa@zytor.com \
    --cc=konrad.wilk@oracle.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@amacapital.net \
    --cc=mingo@redhat.com \
    --cc=oleg@redhat.com \
    --cc=paolo.bonzini@gmail.com \
    --cc=paulmck@linux.vnet.ibm.com \
    --cc=raghavendra.kt@linux.vnet.ibm.com \
    --cc=riel@redhat.com \
    --cc=scott.norton@hp.com \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.org \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=waiman.long@hp.com \
    --cc=x86@kernel.org \
    --cc=xen-devel@lists.xenproject.org \
    --subject='Re: [PATCH 8/9] qspinlock: Generic paravirt support' \
    /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

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).