kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Avi Kivity <avi@redhat.com>
To: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: ehrhardt@linux.vnet.ibm.com, kvm@vger.kernel.org,
	Carsten Otte <cotte@de.ibm.com>
Subject: Re: [PATCH 2/6] kvm-s390: use hrtimer for clock wakeup from idle
Date: Thu, 07 May 2009 13:19:32 +0300	[thread overview]
Message-ID: <4A02B5B4.6050701@redhat.com> (raw)
In-Reply-To: <200905061436.13071.borntraeger@de.ibm.com>

Christian Borntraeger wrote:
>> Why can't this be done from the timer context (after adjusting the locks)?
>>     
>
> It can be done, in fact I did that in my first version. The thing is, we would need to change all local_int.locks to spinlock_irqs, since standard timers are softirqs and hrtimers and hardirqs. Enabling and disabling irqs is a relatively expensive operating on s390 (but locks via compare and swap are quite cheap). Since we take this specific lock in lots of places this lock is on some hot pathes. The idle wakeup on the other hand is not that critical.
>   

Makes sense.

Back when irqsave/irqrestore were expensive on x86 (P4 days, I think it 
was ~100 cycles) there was talk of using a software irq disable flag 
instead of using the hardware support.  So

- local_irq_disable() sets a flag

- interrupt handlers check the flag, if set they schedule the interrupt 
handler and return immediately

- local_irq_restore() clears the flag and fires and scheduled handlers

Now these operations are pretty cheap on x86, but maybe that can apply 
to s390.

-- 
error compiling committee.c: too many arguments to function


  reply	other threads:[~2009-05-07 10:19 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-05-05 14:39 [PATCH 0/6] kvm-s390: collection of kvm-s390 fixes ehrhardt
2009-05-05 14:39 ` [PATCH 1/6] kvm-s390: Fix memory slot versus run ehrhardt
2009-05-06 12:01   ` Avi Kivity
2009-05-11 13:00     ` Christian Ehrhardt
2009-05-11 13:15       ` Avi Kivity
2009-05-11 13:46         ` Christian Ehrhardt
2009-05-11 14:02           ` Avi Kivity
2009-05-11 14:42             ` Christian Ehrhardt
2009-05-11 15:01               ` Avi Kivity
2009-05-12  9:15                 ` Christian Ehrhardt
2009-05-12 11:35                   ` Avi Kivity
2009-05-12 13:33                     ` Christian Ehrhardt
2009-05-17 22:31                       ` Avi Kivity
2009-05-20 12:05                         ` Christian Ehrhardt
2009-05-05 14:39 ` [PATCH 2/6] kvm-s390: use hrtimer for clock wakeup from idle ehrhardt
2009-05-06 12:10   ` Avi Kivity
2009-05-06 12:36     ` Christian Borntraeger
2009-05-07 10:19       ` Avi Kivity [this message]
2009-05-07 10:34         ` Christian Borntraeger
2009-05-20 15:48         ` Hollis Blanchard
2009-05-05 14:39 ` [PATCH 3/6] kvm-s390: optimize float int lock: spin_lock_bh --> spin_lock ehrhardt
2009-05-05 14:39 ` [PATCH 4/6] kvm-s390: Unlink vcpu on destroy ehrhardt
2009-05-06 12:11   ` Avi Kivity
2009-05-11 13:00     ` Christian Ehrhardt
2009-05-05 14:39 ` [PATCH 5/6] kvm-s390: Sanity check on validity intercept ehrhardt
2009-05-05 14:39 ` [PATCH 6/6] kvm-s390: Verify memory in kvm run ehrhardt

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=4A02B5B4.6050701@redhat.com \
    --to=avi@redhat.com \
    --cc=borntraeger@de.ibm.com \
    --cc=cotte@de.ibm.com \
    --cc=ehrhardt@linux.vnet.ibm.com \
    --cc=kvm@vger.kernel.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).