Re: [PROBLEM] i915/drm: Insufficient FIFO for plane

Previous thread: [GIT PULL] SLAB fixes for 2.6.36-rc2 by Pekka Enberg on Sunday, August 22, 2010 - 8:17 am. (1 message)

Next thread: [PATCH] vt: Fix console corruption on driver hand-over. by Francisco Jerez on Sunday, August 22, 2010 - 8:37 am. (2 messages)
From: Pekka Enberg
Date: Sunday, August 22, 2010 - 8:26 am

Hi,

I'm seeing tons of these error messages with current Linus' git:

[   95.941618] [drm:intel_calculate_wm] *ERROR* Insufficient FIFO for 
plane, expect flickering: entries required = 36, available = 31.

Contrary to the error message, I don't see any flickering. I've attached 
my dmesg and config.

 			Pekka
From: Maciej Rutecki
Date: Thursday, August 26, 2010 - 11:27 am

From: Pekka Enberg
Date: Sunday, August 29, 2010 - 4:32 am

On Thu, Aug 26, 2010 at 9:27 PM, Maciej Rutecki

The "insufficient FIFO for plane" problem seems to have gone away with
latest Linus' git. I suppose commit
9559fcdbff4f93d29af04478bbc48294519424f5 ("drm/i915: fix vblank wait
test condition") fixed it although I don't see any mention of the
error in the changelog?

                         Pekka
--

From: Sitsofe Wheeler
Date: Sunday, August 29, 2010 - 6:27 am

I found that commit reduced the number of "insufficient FIFO for plane"
messages in certain circumstances too (when the system is stable I only
see one at boot). However when my system is in "unstable mode" (see
http://marc.info/?l=linux-kernel&m=128272404206696&w=2 ), I can still
generate that message using the modetest (at the console) or xrandr (in
X) tools or by doing VT switches. When this is the case, the message
generally arrives in batches of two.

-- 
Sitsofe | http://sucs.org/~sits/
--

Previous thread: [GIT PULL] SLAB fixes for 2.6.36-rc2 by Pekka Enberg on Sunday, August 22, 2010 - 8:17 am. (1 message)

Next thread: [PATCH] vt: Fix console corruption on driver hand-over. by Francisco Jerez on Sunday, August 22, 2010 - 8:37 am. (2 messages)