Skip to contentSkip to conversation

FBC underruns on GMA4500?

DISCUSSION Alexey Fisher 2 participants 4 messages Aug 7, 2010 – Aug 9, 2010

lore
Share & tools
lore
#0 lore raw

Hallo all,

i have some times on my primer desktop with Intel DG45ID board some sort of flickering, it schort (1 second) and only half of the display. Same sort of flickering i had with Bug 24314 - [945GME FBC] underruns followed by freeze after resume with FBC enabled. The difference is - it will not freeze.

If i understand it correctly , it is not normal. But i can't report it, there is no messages in dmesg or xorg.log

Any thoughts about this?
Regards,
Alexey

#1 lore raw
Quoted 10 lines · On Sat, 07 Aug 2010 11:46:19 +0200 Alexey Fisher <[email protected]>…

Hallo all,

i have some times on my primer desktop with Intel DG45ID board some sort of flickering, it schort (1 second) and only half of the display. Same sort of flickering i had with Bug 24314 - [945GME FBC] underruns followed by freeze after resume with FBC enabled. The difference is - it will not freeze.

If i understand it correctly , it is not normal. But i can't report it, there is no messages in dmesg or xorg.log

Sounds like potential FIFO underruns, maybe your memory configuration doesn't match the latency we expect in intel_display.c. You could try increasing the latency constant in that file to see if it makes the flicker go away.

If that helps it's just another example of why we need better latency detection in the driver; unfortunately we don't know how to do that yet...

Signature · 1 line
Jesse Barnes, Intel Open Source Technology Center
#2 lore raw
Quoted 11 lines · Jesse Barnes
Quoted 8 lines · On Sat, 07 Aug 2010 11:46:19 +0200 Alexey Fisher <[email protected]>…

Hallo all,

i have some times on my primer desktop with Intel DG45ID board some sort of flickering, it schort (1 second) and only half of the display. Same sort of flickering i had with Bug 24314 - [945GME FBC] underruns followed by freeze after resume with FBC enabled. The difference is - it will not freeze.

Quoted 7 lines
Quoted 2 lines

If i understand it correctly , it is not normal. But i can't report it, there is no messages in dmesg or xorg.log

Sounds like potential FIFO underruns, maybe your memory configuration doesn't match the latency we expect in intel_display.c. You could try increasing the latency constant in that file to see if it makes the flicker go away.

Can you please point me what file to change? And haw can i check what latency is used?

Quoted 3 lines

If that helps it's just another example of why we need better latency detection in the driver; unfortunately we don't know how to do that yet...

Please take a look to this bug:
https://bugs.freedesktop.org/show_bug.cgi?id=29454
may be FIFO underrun and hangs have same source.

Regards,
Alexey

#3 lore raw
Quoted 15 lines · On Mon, 09 Aug 2010 19:20:34 +0200 Alexey Fisher <[email protected]>…
Quoted 14 lines · Jesse Barnes
Quoted 10 lines · On Sat, 07 Aug 2010 11:46:19 +0200 Alexey Fisher <[email protected]>…

Hallo all,

i have some times on my primer desktop with Intel DG45ID board some sort of flickering, it schort (1 second) and only half of the display. Same sort of flickering i had with Bug 24314 - [945GME FBC] underruns followed by freeze after resume with FBC enabled. The difference is - it will not freeze.

If i understand it correctly , it is not normal. But i can't report it, there is no messages in dmesg or xorg.log

Quoted 7 lines
Quoted 4 lines

Sounds like potential FIFO underruns, maybe your memory configuration doesn't match the latency we expect in intel_display.c. You could try increasing the latency constant in that file to see if it makes the flicker go away.

Can you please point me what file to change? And haw can i check what latency is used?

drivers/gpu/drm/i915/intel_display.c, modify (increase) the latency_ns variable to say 10000.

Quoted 7 lines
Quoted 3 lines

If that helps it's just another example of why we need better latency detection in the driver; unfortunately we don't know how to do that yet...

Please take a look to this bug:
https://bugs.freedesktop.org/show_bug.cgi?id=29454
may be FIFO underrun and hangs have same source.

Yeah repeated underruns can lead to chip hangs.

Signature · 1 line
Jesse Barnes, Intel Open Source Technology Center