Skip to contentSkip to conversation

X hang with quirk VT switches

DISCUSSION Takashi Iwai 2 participants 6 messages Dec 3, 2014 – Dec 4, 2014

lore
Share & tools
lore
#0 lore raw

Hi,

while checking the reported bug about VT switch hang on openSUSE 13.2, I also could reproduce a similar issue as reported: namely, X hangs when repeatedly switching VT quickly.

For example, running the following on KDE results in the stall of X.

	% for i in $(seq 1 100); do chvt 1; chvt 7; done

Looking at the sysrq-t output, it stalls at drm_read(). And after putting some debug prints at event handling codes, it shows like:

drm_queue_vblank_event event_space=4064
send_vblank_event event_space=4064
drm_poll ENTER event_space=4064
drm_poll mask=0x41 event_space=4064
drm_poll ENTER event_space=4064
drm_poll mask=0x41 event_space=4064
drm_read ENTER event_space=4064
drm_read total=32 event_space=4096
drm_poll ENTER event_space=4096
drm_poll mask=0x0 event_space=4096
drm_read ENTER event_space=4096
drm_read ENTER event_space=4096
drm_read ENTER event_space=4096

So, after a vblank event, two poll calls succeeded, followed by one drm_read(). After that, there were one poll call without event, followed by three(!) drm_read() calls. The last three drm_read() never exited, thus X stalled. So, this looks like a race or a refcount issue somewhere.

Note that the problem disappears when passing drm.debug=0x0e. And it doesn't seem to happen with a 2D desktop, either. The race window is supposed to be fairly small.

The problem is reproduced reliably with xf86-video-intel 2.99.916 and the latest Linus tree on a HP laptop with IVB. Also seen on HSW and other chips, too.

Does this ring a bell to anyone?

thanks,

Takashi
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#1 lore raw
Quoted 27 lines · Takashi Iwai

Hi,

while checking the reported bug about VT switch hang on openSUSE 13.2, I also could reproduce a similar issue as reported: namely, X hangs when repeatedly switching VT quickly.

For example, running the following on KDE results in the stall of X.

	% for i in $(seq 1 100); do chvt 1; chvt 7; done

Looking at the sysrq-t output, it stalls at drm_read(). And after putting some debug prints at event handling codes, it shows like:

drm_queue_vblank_event event_space=4064
send_vblank_event event_space=4064
drm_poll ENTER event_space=4064
drm_poll mask=0x41 event_space=4064
drm_poll ENTER event_space=4064
drm_poll mask=0x41 event_space=4064
drm_read ENTER event_space=4064
drm_read total=32 event_space=4096
drm_poll ENTER event_space=4096
drm_poll mask=0x0 event_space=4096
drm_read ENTER event_space=4096
drm_read ENTER event_space=4096
drm_read ENTER event_space=4096

Quoted 5 lines

So, after a vblank event, two poll calls succeeded, followed by one drm_read(). After that, there were one poll call without event, followed by three(!) drm_read() calls. The last three drm_read() never exited, thus X stalled. So, this looks like a race or a refcount issue somewhere.

The key question is how did you get 3 calls to drm_read that each didn't return? The only place where we call drm_read without first doing a poll is in the WakeupHandler with the drm fd flagged for reads. This is broken in ZaphodHeads as the drm fd is not O_NONBLOCK without

commit bd008e5b2953186fc0c6633a885ade95e7043800

Author: Chris Wilson <[email protected]>
Date:   Tue Oct 7 14:13:51 2014 +0100

drm: Implement O_NONBLOCK support on /dev/dri/cardN

I assume that isn't the case as I expect you would have mentioned using ZaphodHeads.

Maarten found a very similar issue using radeon:

https://bugs.freedesktop.org/show_bug.cgi?id=77885

where the supposition was that the DRM_MASTER was not being relinquished.
-Chris

Signature · 5 lines
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#2 lore raw

At Wed, 3 Dec 2014 18:31:45 +0000,

Quoted 29 lines · Chris Wilson
Quoted 27 lines · Takashi Iwai

Hi,

while checking the reported bug about VT switch hang on openSUSE 13.2, I also could reproduce a similar issue as reported: namely, X hangs when repeatedly switching VT quickly.

For example, running the following on KDE results in the stall of X.

	% for i in $(seq 1 100); do chvt 1; chvt 7; done

Looking at the sysrq-t output, it stalls at drm_read(). And after putting some debug prints at event handling codes, it shows like:

drm_queue_vblank_event event_space=4064
send_vblank_event event_space=4064
drm_poll ENTER event_space=4064
drm_poll mask=0x41 event_space=4064
drm_poll ENTER event_space=4064
drm_poll mask=0x41 event_space=4064
drm_read ENTER event_space=4064
drm_read total=32 event_space=4096
drm_poll ENTER event_space=4096
drm_poll mask=0x0 event_space=4096
drm_read ENTER event_space=4096
drm_read ENTER event_space=4096
drm_read ENTER event_space=4096

Quoted 8 lines
Quoted 5 lines

So, after a vblank event, two poll calls succeeded, followed by one drm_read(). After that, there were one poll call without event, followed by three(!) drm_read() calls. The last three drm_read() never exited, thus X stalled. So, this looks like a race or a refcount issue somewhere.

The key question is how did you get 3 calls to drm_read that each didn't return?

I asked this to myself, too...

Quoted 4 lines

The only place where we call drm_read without first doing a poll is in the WakeupHandler with the drm fd flagged for reads. This is broken in ZaphodHeads as the drm fd is not O_NONBLOCK without

Quoted 8 lines

commit bd008e5b2953186fc0c6633a885ade95e7043800

Author: Chris Wilson <[email protected]>
Date:   Tue Oct 7 14:13:51 2014 +0100

drm: Implement O_NONBLOCK support on /dev/dri/cardN

I assume that isn't the case as I expect you would have mentioned using ZaphodHeads.

Right, no extra option is used. And, eDP output only.

Quoted 6 lines

Maarten found a very similar issue using radeon:

https://bugs.freedesktop.org/show_bug.cgi?id=77885

where the supposition was that the DRM_MASTER was not being relinquished.

Interesting. In the original bug report we've got, the symptom happened at switching two X sessions, as the bfo entry above.

	http://bugzilla.opensuse.org/show_bug.cgi?id=904097

The reporter mentioned there that his issue was fixed with 3.17.x kernel, so it's likely a different problem from what I've seen. The bug couldn't be reproduced on my machine, so I tried the simple VT quick switches, then hits a new bug now.

In anyway, I'll poke a bit more deeply later (maybe on Friday).

But, before that, I'd like to ask: can anyone reproduce this problem at all...?

thanks,

Takashi
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#3 lore raw
In reply to Chris Wilson

At Wed, 3 Dec 2014 18:31:45 +0000,

Quoted 40 lines · Chris Wilson
Quoted 27 lines · Takashi Iwai

Hi,

while checking the reported bug about VT switch hang on openSUSE 13.2, I also could reproduce a similar issue as reported: namely, X hangs when repeatedly switching VT quickly.

For example, running the following on KDE results in the stall of X.

	% for i in $(seq 1 100); do chvt 1; chvt 7; done

Looking at the sysrq-t output, it stalls at drm_read(). And after putting some debug prints at event handling codes, it shows like:

drm_queue_vblank_event event_space=4064
send_vblank_event event_space=4064
drm_poll ENTER event_space=4064
drm_poll mask=0x41 event_space=4064
drm_poll ENTER event_space=4064
drm_poll mask=0x41 event_space=4064
drm_read ENTER event_space=4064
drm_read total=32 event_space=4096
drm_poll ENTER event_space=4096
drm_poll mask=0x0 event_space=4096
drm_read ENTER event_space=4096
drm_read ENTER event_space=4096
drm_read ENTER event_space=4096

Quoted 5 lines

So, after a vblank event, two poll calls succeeded, followed by one drm_read(). After that, there were one poll call without event, followed by three(!) drm_read() calls. The last three drm_read() never exited, thus X stalled. So, this looks like a race or a refcount issue somewhere.

The key question is how did you get 3 calls to drm_read that each didn't return? The only place where we call drm_read without first doing a poll is in the WakeupHandler with the drm fd flagged for reads. This is broken in ZaphodHeads as the drm fd is not O_NONBLOCK without

Quoted 8 lines

commit bd008e5b2953186fc0c6633a885ade95e7043800

Author: Chris Wilson <[email protected]>
Date:   Tue Oct 7 14:13:51 2014 +0100

drm: Implement O_NONBLOCK support on /dev/dri/cardN

I assume that isn't the case as I expect you would have mentioned using ZaphodHeads.

I took a look back at drm_read() code again, and I found that the function doesn't care about O_NONBLOCK at all. (And there is a memory leak, too.)

So I added the support for O_NONBLOCK, and the problem seems resolved.

Although this is no right "fix" (the caller side should be fixed), it would be good to have in anyway. I'm going to send patches for review to dri-devel ML, as it's no i915 specific.

thanks,

Takashi
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#4 lore raw
Quoted 55 lines · Takashi Iwai

At Wed, 3 Dec 2014 18:31:45 +0000,

Quoted 40 lines · Chris Wilson
Quoted 27 lines · Takashi Iwai

Hi,

while checking the reported bug about VT switch hang on openSUSE 13.2, I also could reproduce a similar issue as reported: namely, X hangs when repeatedly switching VT quickly.

For example, running the following on KDE results in the stall of X.

	% for i in $(seq 1 100); do chvt 1; chvt 7; done

Looking at the sysrq-t output, it stalls at drm_read(). And after putting some debug prints at event handling codes, it shows like:

drm_queue_vblank_event event_space=4064
send_vblank_event event_space=4064
drm_poll ENTER event_space=4064
drm_poll mask=0x41 event_space=4064
drm_poll ENTER event_space=4064
drm_poll mask=0x41 event_space=4064
drm_read ENTER event_space=4064
drm_read total=32 event_space=4096
drm_poll ENTER event_space=4096
drm_poll mask=0x0 event_space=4096
drm_read ENTER event_space=4096
drm_read ENTER event_space=4096
drm_read ENTER event_space=4096

Quoted 5 lines

So, after a vblank event, two poll calls succeeded, followed by one drm_read(). After that, there were one poll call without event, followed by three(!) drm_read() calls. The last three drm_read() never exited, thus X stalled. So, this looks like a race or a refcount issue somewhere.

The key question is how did you get 3 calls to drm_read that each didn't return? The only place where we call drm_read without first doing a poll is in the WakeupHandler with the drm fd flagged for reads. This is broken in ZaphodHeads as the drm fd is not O_NONBLOCK without

Quoted 8 lines

commit bd008e5b2953186fc0c6633a885ade95e7043800

Author: Chris Wilson <[email protected]>
Date:   Tue Oct 7 14:13:51 2014 +0100

drm: Implement O_NONBLOCK support on /dev/dri/cardN

I assume that isn't the case as I expect you would have mentioned using ZaphodHeads.

I took a look back at drm_read() code again, and I found that the function doesn't care about O_NONBLOCK at all. (And there is a memory leak, too.)

Quoted 6 lines

So I added the support for O_NONBLOCK, and the problem seems resolved.

Although this is no right "fix" (the caller side should be fixed), it would be good to have in anyway. I'm going to send patches for review to dri-devel ML, as it's no i915 specific.

I disagree. drm has claimed to support O_NONBLOCK since its inception, but the implementation was buggy. However, I don't think there is a case in non-ZaphodHeads where we use read() without first select/poll reporting that there is something to use (and the problem with ZaphodHeads is that we have two screens that share the same drm fd without clearing the select read flags... hmm)
-Chris

Signature · 5 lines
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#5 lore raw

At Thu, 4 Dec 2014 11:21:47 +0000,

Quoted 60 lines · Chris Wilson
Quoted 58 lines · Takashi Iwai

At Wed, 3 Dec 2014 18:31:45 +0000,

Quoted 40 lines · Chris Wilson
Quoted 27 lines · Takashi Iwai

Hi,

while checking the reported bug about VT switch hang on openSUSE 13.2, I also could reproduce a similar issue as reported: namely, X hangs when repeatedly switching VT quickly.

For example, running the following on KDE results in the stall of X.

	% for i in $(seq 1 100); do chvt 1; chvt 7; done

Looking at the sysrq-t output, it stalls at drm_read(). And after putting some debug prints at event handling codes, it shows like:

drm_queue_vblank_event event_space=4064
send_vblank_event event_space=4064
drm_poll ENTER event_space=4064
drm_poll mask=0x41 event_space=4064
drm_poll ENTER event_space=4064
drm_poll mask=0x41 event_space=4064
drm_read ENTER event_space=4064
drm_read total=32 event_space=4096
drm_poll ENTER event_space=4096
drm_poll mask=0x0 event_space=4096
drm_read ENTER event_space=4096
drm_read ENTER event_space=4096
drm_read ENTER event_space=4096

Quoted 5 lines

So, after a vblank event, two poll calls succeeded, followed by one drm_read(). After that, there were one poll call without event, followed by three(!) drm_read() calls. The last three drm_read() never exited, thus X stalled. So, this looks like a race or a refcount issue somewhere.

The key question is how did you get 3 calls to drm_read that each didn't return? The only place where we call drm_read without first doing a poll is in the WakeupHandler with the drm fd flagged for reads. This is broken in ZaphodHeads as the drm fd is not O_NONBLOCK without

Quoted 8 lines

commit bd008e5b2953186fc0c6633a885ade95e7043800

Author: Chris Wilson <[email protected]>
Date:   Tue Oct 7 14:13:51 2014 +0100

drm: Implement O_NONBLOCK support on /dev/dri/cardN

I assume that isn't the case as I expect you would have mentioned using ZaphodHeads.

I took a look back at drm_read() code again, and I found that the function doesn't care about O_NONBLOCK at all. (And there is a memory leak, too.)

So I added the support for O_NONBLOCK, and the problem seems resolved.

Quoted 6 lines
Quoted 3 lines

Although this is no right "fix" (the caller side should be fixed), it would be good to have in anyway. I'm going to send patches for review to dri-devel ML, as it's no i915 specific.

I disagree. drm has claimed to support O_NONBLOCK since its inception, but the implementation was buggy.

The nonblock read is obviously buggy. If the current implementation is intentional, then the nonblock flag is somehow misused...

Quoted 5 lines

However, I don't think there is a case
in non-ZaphodHeads where we use read() without first select/poll reporting that there is something to use (and the problem with ZaphodHeads is that we have two screens that share the same drm fd without clearing the select read flags... hmm)

In my case, I'm using a single screen, so this can't be. And, my rough guess is that this isn't about the lack of poll but rather some race between poll/read or two reads. That explains why my patch worked.

In anyway I'd need to trap X stall and diagnose, but I have to leave
my machine now. Will check it tomorrow.

Meanwhile, it's interesting to see whether this covers Maarten's case, too...

thanks,

Takashi
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx