All of lore.kernel.org
 help / color / mirror / Atom feed
From: David Vrabel <dvrabel@cantab.net>
To: Boris Ostrovsky <boris.ostrovsky@oracle.com>,
	David Vrabel <david.vrabel@citrix.com>,
	xen-devel@lists.xenproject.org
Cc: Keir Fraser <keir@xen.org>, Tim Deegan <tim@xen.org>,
	Ian Campbell <ian.campbell@citrix.com>,
	Jan Beulich <jbeulich@suse.com>,
	Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Subject: Re: [PATCHv2 5/6] xen: use ticket locks for spin locks
Date: Fri, 10 Apr 2015 18:29:17 +0100	[thread overview]
Message-ID: <5528086D.7090407@cantab.net> (raw)
In-Reply-To: <5527FDF6.3030706@oracle.com>



On 10/04/2015 17:44, Boris Ostrovsky wrote:
> On 04/10/2015 10:19 AM, David Vrabel wrote:
> 
>> @@ -138,14 +138,17 @@ void spin_debug_disable(void)
>>
>>   void _spin_lock(spinlock_t *lock)
>>   {
>> +    spinlock_tickets_t tickets = { .tail = 1, };
>>       LOCK_PROFILE_VAR;
>>
>>       check_lock(&lock->debug);
>> -    while ( unlikely(!_raw_spin_trylock(&lock->raw)) )
>> +    tickets.head_tail = xadd(&lock->tickets, tickets.head_tail);
>> +    if ( tickets.tail != read_atomic(&lock->tickets.head) )
>>       {
>>           LOCK_PROFILE_BLOCK;
>> -        while ( likely(_raw_spin_is_locked(&lock->raw)) )
>> +        do {
>>               cpu_relax();
>> +        } while ( tickets.tail != read_atomic(&lock->tickets.head) );
>>       }
> 
> 
> Why do you use both 'if' and 'while"? I.e. why not just
> 
> while ( tickets.tail != read_atomic(&lock->tickets.head) )
> {
>     LOCK_PROFILE_BLOCK;
>     cpu_relax();
> }

We need to only call LOCK_PROFILE_BLOCK once when we start blocking.

>> @@ -194,35 +174,44 @@ void _spin_unlock(spinlock_t *lock)
>>   {
>>       preempt_enable();
>>       LOCK_PROFILE_REL;
>> -    _raw_spin_unlock(&lock->raw);
>> +    lock->tickets.head++;
> 
> 
> Is it safe to do a plain increment here and a locked one?

Yes. Only the lock holder writes to head.

David

  reply	other threads:[~2015-04-10 17:29 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-04-10 14:19 [PATCHv2 0/6] Use ticket locks for spinlocks David Vrabel
2015-04-10 14:19 ` [PATCHv2 1/6] x86/hvm: don't include asm/spinlock.h David Vrabel
2015-04-10 15:24   ` Andrew Cooper
2015-04-13 13:13     ` David Vrabel
2015-04-13 13:15       ` Andrew Cooper
2015-04-16 11:09   ` Tim Deegan
2015-04-10 14:19 ` [PATCHv2 2/6] x86/mtrr: include asm/atomic.h David Vrabel
2015-04-10 15:25   ` Andrew Cooper
2015-04-16 11:10   ` Tim Deegan
2015-04-10 14:19 ` [PATCHv2 3/6] xen: generic xadd() for ticket locks David Vrabel
2015-04-14 13:17   ` Ian Campbell
2015-04-14 16:37     ` David Vrabel
2015-04-15  8:34       ` Ian Campbell
2015-04-16 11:13   ` Tim Deegan
2015-04-16 15:28   ` Jan Beulich
2015-04-17 12:32     ` Ian Campbell
2015-04-17 12:33       ` Ian Campbell
2015-04-17 12:34       ` David Vrabel
2015-04-17 13:09         ` Ian Campbell
2015-04-17 13:13           ` David Vrabel
2015-04-10 14:19 ` [PATCHv2 4/6] x86: provide xadd() David Vrabel
2015-04-16 11:25   ` Tim Deegan
2015-04-16 11:38     ` David Vrabel
2015-04-16 11:41       ` Tim Deegan
2015-04-16 11:49     ` Jan Beulich
2015-04-16 12:49       ` Tim Deegan
2015-04-16 12:55         ` Ian Campbell
2015-04-16 13:19           ` Jan Beulich
2015-04-16 14:07             ` Ian Campbell
2015-04-16 14:09               ` Ian Campbell
2015-04-16 12:00     ` Ian Campbell
2015-04-10 14:19 ` [PATCHv2 5/6] xen: use ticket locks for spin locks David Vrabel
2015-04-10 15:29   ` Andrew Cooper
2015-04-10 16:44   ` Boris Ostrovsky
2015-04-10 17:29     ` David Vrabel [this message]
2015-04-10 17:58       ` Boris Ostrovsky
2015-04-16 12:03   ` Tim Deegan
2015-04-16 12:46     ` David Vrabel
2015-04-16 12:50       ` Tim Deegan
2015-04-16 13:32     ` Jan Beulich
2015-04-21  9:59     ` David Vrabel
2015-04-10 14:19 ` [PATCHv2 6/6] x86, arm: remove asm/spinlock.h from all architectures David Vrabel
2015-04-16 12:51   ` Tim Deegan
2015-04-16 15:29   ` Jan Beulich

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=5528086D.7090407@cantab.net \
    --to=dvrabel@cantab.net \
    --cc=boris.ostrovsky@oracle.com \
    --cc=david.vrabel@citrix.com \
    --cc=ian.campbell@citrix.com \
    --cc=jbeulich@suse.com \
    --cc=keir@xen.org \
    --cc=stefano.stabellini@eu.citrix.com \
    --cc=tim@xen.org \
    --cc=xen-devel@lists.xenproject.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.