intel-gfx.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: Ben Widawsky <ben@bwidawsk.net>
To: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 2/3] drm/i915: close PM interrupt masking races in the rps work func
Date: Sun, 4 Sep 2011 10:08:17 -0700	[thread overview]
Message-ID: <20110904100817.16c6c4cc@bwidawsk.net> (raw)
In-Reply-To: <1315150502-12537-3-git-send-email-daniel.vetter@ffwll.ch>

On Sun,  4 Sep 2011 17:35:01 +0200
Daniel Vetter <daniel.vetter@ffwll.ch> wrote:

> This patch closes the following race:
> 
> We get a PM interrupt A, mask it, set dev_priv->iir = PM_A and kick
> of the work item. Scheduler isn't grumpy, so the work queue takes
> rps_lock, grabs pm_iir = dev_priv->pm_iir and pm_imr = READ(PMIMR).
> Note that pm_imr == pm_iir because we've just masked the interrupt
> we've got.
> 
> Now hw sends out PM interrupt B (not masked), we process it and mask
> it.  Later on the irq handler also clears PMIIR.
> 
> Then the work item proceeds and at the end clears PMIMR. Because
> (local) pm_imr == pm_iir we have
>         pm_imr & ~pm_iir == 0
> so all interrupts are enabled.
> 
> Hardware is still interrupt-happy, and sends out a new PM interrupt B.
> PMIMR doesn't mask B (it does not mask anything), PMIIR is cleared, so
> we get it and hit the WARN in the interrupt handler (because
> dev_priv->pm_iir == PM_B).
> 
> That's why I've moved the
>         WRITE(PMIMR, 0)
> up under the protection of the rps_lock. And write an uncoditional 0
> to PMIMR, because that's what we'll do anyway.
> 
> This races looks much more likely because we can arbitrarily extend
> the window by grabing dev->struct mutex right after the irq handler
> has processed the first PM_B interrupt.
> 
> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
> ---
>  drivers/gpu/drm/i915/i915_irq.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/i915_irq.c
> b/drivers/gpu/drm/i915/i915_irq.c index 2fdd9f9..21ebcbd 100644
> --- a/drivers/gpu/drm/i915/i915_irq.c
> +++ b/drivers/gpu/drm/i915/i915_irq.c
> @@ -383,6 +383,7 @@ static void gen6_pm_rps_work(struct work_struct
> *work) pm_iir = dev_priv->pm_iir;
>  	dev_priv->pm_iir = 0;
>  	pm_imr = I915_READ(GEN6_PMIMR);
> +	I915_WRITE(GEN6_PMIMR, 0);
>  	spin_unlock_irq(&dev_priv->rps_lock);
>  
>  	if (!pm_iir)
> @@ -420,7 +421,6 @@ static void gen6_pm_rps_work(struct work_struct
> *work)
>  	 * an *extremely* unlikely race with gen6_rps_enable() that
> is prevented
>  	 * by holding struct_mutex for the duration of the write.
>  	 */
> -	I915_WRITE(GEN6_PMIMR, pm_imr & ~pm_iir);
>  	mutex_unlock(&dev_priv->dev->struct_mutex);
>  }
>  

How about this:

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 55518e3..3bc1479 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -415,12 +415,7 @@ static void gen6_pm_rps_work(struct work_struct *work)
        gen6_set_rps(dev_priv->dev, new_delay);
        dev_priv->cur_delay = new_delay;
 
-       /*
-        * rps_lock not held here because clearing is non-destructive. There is
-        * an *extremely* unlikely race with gen6_rps_enable() that is prevented
-        * by holding struct_mutex for the duration of the write.
-        */
-       I915_WRITE(GEN6_PMIMR, pm_imr & ~pm_iir);
+       I915_WRITE(GEN6_PMIMR, pm_imr & dev_priv->pm_iir);
        mutex_unlock(&dev_priv->dev->struct_mutex);
 }

Ben

  reply	other threads:[~2011-09-04 17:08 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-09-04  3:24 [PATCH] drm/i915: Fix rps irq warning Ben Widawsky
2011-09-04  9:03 ` Chris Wilson
2011-09-04 15:49   ` Ben Widawsky
2011-09-04 15:34     ` [PATCH 0/3] slaughter rps races some more Daniel Vetter
2011-09-04 15:35       ` [PATCH 1/3] drm/i915: close PM interrupt masking races in the irq handler Daniel Vetter
2011-09-04 17:09         ` Ben Widawsky
2011-09-04 15:35       ` [PATCH 2/3] drm/i915: close PM interrupt masking races in the rps work func Daniel Vetter
2011-09-04 17:08         ` Ben Widawsky [this message]
2011-09-04 19:26           ` Daniel Vetter
2011-09-04 19:56             ` Ben Widawsky
2011-09-04 20:10               ` Daniel Vetter
2011-09-04 21:38                 ` Ben Widawsky
2011-09-05  6:38                   ` Daniel Vetter
2011-09-05  6:51                     ` Ben Widawsky
2011-09-05 12:15                       ` Chris Wilson
2011-09-04 15:35       ` [PATCH 3/3] drm/i915: close rps work vs. rps disable races Daniel Vetter
2011-09-04 17:23         ` Ben Widawsky
2011-09-04 19:17           ` Daniel Vetter
2011-09-04 19:50             ` Ben Widawsky
2011-09-04 19:57               ` Daniel Vetter
2011-09-05  8:15                 ` [PATCH] drm/i915: properly cancel rps_work on module unload Daniel Vetter
2011-09-05 17:27                   ` Ben Widawsky
2011-09-08 12:00 [PATCH 1/3] drm/i915: close PM interrupt masking races in the irq handler Daniel Vetter
2011-09-08 12:00 ` [PATCH 2/3] drm/i915: close PM interrupt masking races in the rps work func Daniel Vetter
2011-09-08 15:19   ` Ben Widawsky

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=20110904100817.16c6c4cc@bwidawsk.net \
    --to=ben@bwidawsk.net \
    --cc=daniel.vetter@ffwll.ch \
    --cc=intel-gfx@lists.freedesktop.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).