netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Zachary Amsden <zach@vmware.com>
To: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: xen-devel@lists.xensource.com, akpm@linux-foundation.org,
	virtualization@lists.osdl.org, netdev@vger.kernel.org,
	Rusty Russell <rusty@rustcorp.com.au>,
	linux-kernel@vger.kernel.org, chrisw@sous-sol.org,
	Andi Kleen <ak@muc.de>,
	"Eric W. Biederman" <ebiederm@xmission.com>,
	anthony@codemonkey.ws, mingo@elte.hu,
	Linus Torvalds <torvalds@linux-foundation.org>,
	David Miller <davem@davemloft.net>
Subject: Re: [patch 13/26] Xen-paravirt_ops: Consistently wrap paravirt ops callsites to make them patchable
Date: Tue, 20 Mar 2007 14:09:13 -0800	[thread overview]
Message-ID: <46005B89.5070301@vmware.com> (raw)
In-Reply-To: <46000C7E.4070001@goop.org>

Jeremy Fitzhardinge wrote:

>> Yeah, disable interrupts, and set a flag that the fake "sti" can test, and 
>> just return without doing anything.
>>
>> (You may or may not also need to do extra work to Ack the hardware 
>> interrupt etc, which may be irq-controller specific. Once the CPU has 
>> accepted the interrupt, you may not be able to just leave it dangling)
>>   
>>     
>
> So it would be something like:
>
>     pda.intr_mask = 1;		/* disable interrupts */
>     ...
>     pda.intr_mask = 0;		/* enable interrupts */
>     if (xchg(&pda.intr_pending, 0))	/* check pending */
>   

Well, can't do xchg, since it implies #LOCK, and you'll lose more than 
you gain on the processors where it matters.  Cmpxchg is fine, but 
processor dependent.

Or, just make the interrupt handlers use software resend for IRQs when 
pda.intr_mask is set to zero.  Now, local_irq_save / restore are very 
pretty:

int local_irq_save(void)
{
   int tmp = pda.intr_mask;
   pda.intr_mask = 0;
   /*
    * note there is a window here where local IRQs notice intr_mask == 0
    * in that case, they will attempt to resend the IRQ via a tasklet,
    * and will succeed, albeit through a slightly longer path
    */
   local_bh_disable();
   return tmp;
}

void local_irq_restore(int enabled)
{
    pda.intr_mask = enabled;
    /*
     * note there is a window here where softirqs are not processed by
     * the interrupt handler, but that is not a problem, since it will
     * get done here in the outer enable of any nested pair.
     */
    if (enabled)
        local_bh_enable();
}

I think Ingo's suggestion of using the hardirq tracing is another way 
that could work, but it seems to be too heavyweight and tied too much to 
the lockdep verification code - plus it inserts additional 
raw_irq_disable in places that seem counter to the goal of getting rid 
of them in the first place.  Perhaps I'm misunderstanding the code, though.

Zach

  reply	other threads:[~2007-03-20 22:09 UTC|newest]

Thread overview: 93+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20070301232443.195603797@goop.org>
2007-03-01 23:25 ` [patch 26/26] Xen-paravirt_ops: Add the Xen virtual network device driver Jeremy Fitzhardinge
2007-03-02  0:42   ` Stephen Hemminger
2007-03-02  0:56     ` Jeremy Fitzhardinge
2007-03-02  1:30       ` [RFC] Arp announce (for Xen) Stephen Hemminger
2007-03-02  8:09         ` Pekka Savola
2007-03-02 18:29           ` Ben Greear
2007-03-02 19:59             ` Stephen Hemminger
2007-03-02  8:46         ` Keir Fraser
2007-03-02 12:54         ` Andi Kleen
2007-03-02 14:08           ` jamal
2007-03-02 18:08         ` Chris Wright
2007-03-06  4:35         ` David Miller
2007-03-06 18:51           ` [RFC] ARP notify option Stephen Hemminger
2007-03-06 19:04             ` Jeremy Fitzhardinge
2007-03-06 19:07             ` Chris Wright
2007-03-06 21:18             ` Chris Friesen
2007-03-06 22:52               ` Stephen Hemminger
2007-03-07  6:42               ` Pekka Savola
2007-03-07 17:00                 ` Stephen Hemminger
2007-03-02  1:21     ` [patch 26/26] Xen-paravirt_ops: Add the Xen virtual network device driver Christoph Hellwig
2007-03-02  1:26       ` Chris Wright
     [not found] ` <20070301232527.956565107@goop.org>
     [not found]   ` <20070316092445.GM23174@elte.hu>
     [not found]     ` <20070316.023331.59468179.davem@davemloft.net>
2007-03-16 20:38       ` [patch 13/26] Xen-paravirt_ops: Consistently wrap paravirt ops callsites to make them patchable Jeremy Fitzhardinge
2007-03-17 10:33         ` Rusty Russell
2007-03-18  7:33           ` David Miller
2007-03-18  7:59             ` Jeremy Fitzhardinge
2007-03-18 12:08             ` Andi Kleen
2007-03-18 15:58               ` Jeremy Fitzhardinge
2007-03-18 17:04                 ` Andi Kleen
2007-03-18 17:29                   ` Jeremy Fitzhardinge
2007-03-18 19:30                     ` Andi Kleen
2007-03-18 23:46                       ` Jeremy Fitzhardinge
2007-03-19 10:57                         ` Andi Kleen
2007-03-19 17:58                           ` Jeremy Fitzhardinge
2007-03-19 19:08                           ` David Miller
2007-03-19 20:59                             ` Andi Kleen
2007-03-20  3:18                               ` Linus Torvalds
2007-03-20  3:47                                 ` David Miller
2007-03-20  4:19                                   ` Eric W. Biederman
2007-03-20 13:28                                     ` Andi Kleen
2007-03-20 16:25                                       ` Eric W. Biederman
2007-03-20 17:42                                         ` Andi Kleen
2007-03-20 16:52                                           ` Linus Torvalds
2007-03-20 18:03                                             ` Andi Kleen
2007-03-20 17:27                                               ` Linus Torvalds
2007-03-20 19:21                                                 ` Andi Kleen
2007-03-20 18:49                                                   ` Linus Torvalds
2007-03-20 20:23                                                     ` Andi Kleen
2007-03-20 21:39                                                       ` Alan Cox
2007-03-20 21:49                                                         ` [Xen-devel] " Andi Kleen
2007-03-20 23:51                                                           ` Linus Torvalds
2007-03-20 23:43                                                       ` Linus Torvalds
2007-03-21  6:08                                                       ` Andrew Morton
2007-03-20 16:12                                     ` Chuck Ebbert
2007-03-20  1:23                         ` Zachary Amsden
2007-03-20  1:45                           ` Jeremy Fitzhardinge
2007-03-19  2:47               ` Rusty Russell
2007-03-19 18:25                 ` Eric W. Biederman
2007-03-19 18:38                   ` Linus Torvalds
2007-03-19 18:44                     ` Linus Torvalds
2007-03-19 19:33                     ` Jeremy Fitzhardinge
2007-03-20  0:01                     ` Rusty Russell
2007-03-20  2:00                       ` Zachary Amsden
2007-03-20  4:20                         ` Rusty Russell
2007-03-20  5:54                         ` Jeremy Fitzhardinge
2007-03-20 11:33                           ` Andreas Kleen
2007-03-20 15:09                           ` Linus Torvalds
2007-03-20 15:58                             ` Eric W. Biederman
2007-03-20 16:06                               ` Linus Torvalds
2007-03-20 16:31                                 ` Jeremy Fitzhardinge
2007-03-20 22:09                                   ` Zachary Amsden [this message]
2007-03-21  0:24                                     ` Linus Torvalds
2007-03-21  2:53                                       ` Zachary Amsden
2007-03-21  2:15                                         ` Linus Torvalds
2007-03-21  3:43                                           ` Zachary Amsden
2007-03-20 22:43                                   ` Matt Mackall
2007-03-20 23:08                                     ` Zachary Amsden
2007-03-20 23:33                                       ` Jeremy Fitzhardinge
2007-03-21  1:14                                         ` Zachary Amsden
2007-03-20 23:41                                       ` Matt Mackall
2007-03-21  0:20                                     ` Jeremy Fitzhardinge
2007-03-20 19:28                                 ` Andi Kleen
2007-03-20 19:54                                   ` Zachary Amsden
2007-03-20 20:02                                     ` Andi Kleen
2007-03-20 16:26                               ` Jeremy Fitzhardinge
2007-03-20 22:41                               ` Rusty Russell
2007-03-20 17:00                             ` Ingo Molnar
2007-03-21  0:03                             ` Paul Mackerras
2007-04-12 23:16                               ` David Miller
2007-03-19 18:41                   ` Chris Wright
2007-03-19 19:10                   ` Jeremy Fitzhardinge
2007-03-19 19:46                     ` David Miller
2007-03-19 20:06                       ` Jeremy Fitzhardinge
2007-03-19 23:42                     ` Andi Kleen

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=46005B89.5070301@vmware.com \
    --to=zach@vmware.com \
    --cc=ak@muc.de \
    --cc=akpm@linux-foundation.org \
    --cc=anthony@codemonkey.ws \
    --cc=chrisw@sous-sol.org \
    --cc=davem@davemloft.net \
    --cc=ebiederm@xmission.com \
    --cc=jeremy@goop.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=netdev@vger.kernel.org \
    --cc=rusty@rustcorp.com.au \
    --cc=torvalds@linux-foundation.org \
    --cc=virtualization@lists.osdl.org \
    --cc=xen-devel@lists.xensource.com \
    /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).