qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Anthony Liguori <aliguori@linux.vnet.ibm.com>
To: Jamie Lokier <jamie@shareable.org>
Cc: Gautham R Shenoy <ego@in.ibm.com>,
	Qemu-development List <qemu-devel@nongnu.org>,
	"Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>,
	Corentin Chary <corentin.chary@gmail.com>,
	Paolo Bonzini <pbonzini@redhat.com>, Avi Kivity <avi@redhat.com>
Subject: Re: [Qemu-devel] Re: [PATCH V4 2/3] qemu: Generic task offloading framework: threadlets
Date: Wed, 16 Jun 2010 09:27:12 -0500	[thread overview]
Message-ID: <4C18DF40.2050706@linux.vnet.ibm.com> (raw)
In-Reply-To: <20100616142236.GA20052@shareable.org>

On 06/16/2010 09:22 AM, Jamie Lokier wrote:
> Paolo Bonzini wrote:
>    
>> These should be (at least for now) block-obj-$(CONFIG_POSIX).
>>
>>      
>>> +        while (QTAILQ_EMPTY(&(queue->request_list))&&
>>> +               (ret != ETIMEDOUT)) {
>>> +            ret = qemu_cond_timedwait(&(queue->cond),
>>> +					&(queue->lock), 10*100000);
>>> +        }
>>>        
>> Using qemu_cond_timedwait is a hack for not properly broadcasting the
>> condvar in flush_threadlet_queue.
>>      
> Are you sure?  It looks like it also expires idle threads after a
> fixed amount of idle time.
>    

Yup, that's the intention of the code.

We signal instead of broadcast because broadcasting causes all threads 
to wake up which hurts performance.  AFAICT, there is no correctness 
issue in using signal vs. broadcast.

Regards,

Anthony Liguori

> -- Jamie
>    

  reply	other threads:[~2010-06-16 14:27 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-06-16 11:56 [Qemu-devel] [PATCH V4 0/3] qemu: Threadlets: A generic task offloading framework Gautham R Shenoy
2010-06-16 11:56 ` [Qemu-devel] [PATCH V4 1/3] qemu: Add qemu-barrier support to qemu-thread framework Gautham R Shenoy
2010-06-16 14:40   ` [Qemu-devel] " Anthony Liguori
2010-06-16 14:42     ` Paolo Bonzini
2010-06-16 11:56 ` [Qemu-devel] [PATCH V4 2/3] qemu: Generic task offloading framework: threadlets Gautham R Shenoy
2010-06-16 12:34   ` [Qemu-devel] " Paolo Bonzini
2010-06-16 14:22     ` Jamie Lokier
2010-06-16 14:27       ` Anthony Liguori [this message]
2010-06-16 14:29       ` Paolo Bonzini
2010-06-16 14:38         ` Anthony Liguori
2010-06-16 14:52           ` Paolo Bonzini
2010-06-16 15:20             ` Anthony Liguori
2010-06-16 15:47               ` Corentin Chary
2010-06-16 15:52                 ` Anthony Liguori
2010-06-16 16:06                   ` Corentin Chary
2010-06-17  9:16                     ` Gautham R Shenoy
2010-06-17  9:12               ` Gautham R Shenoy
2010-06-16 14:58           ` Jamie Lokier
2010-06-16 15:07             ` Jamie Lokier
2010-06-16 16:45               ` Paolo Bonzini
2010-06-17  8:53     ` Gautham R Shenoy
2010-06-17 10:09       ` Paolo Bonzini
2010-06-17 18:05         ` Anthony Liguori
2010-06-18  7:52           ` Paolo Bonzini
2010-06-16 11:57 ` [Qemu-devel] [PATCH V4 3/3] qemu: Convert AIO code to use threadlets Gautham R Shenoy
2010-06-16 13:09 ` [Qemu-devel] [PATCH V4 0/3] qemu: Threadlets: A generic task offloading framework Anthony Liguori
2010-06-16 13:18   ` Paolo Bonzini

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=4C18DF40.2050706@linux.vnet.ibm.com \
    --to=aliguori@linux.vnet.ibm.com \
    --cc=aneesh.kumar@linux.vnet.ibm.com \
    --cc=avi@redhat.com \
    --cc=corentin.chary@gmail.com \
    --cc=ego@in.ibm.com \
    --cc=jamie@shareable.org \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.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 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).