Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
i915 weird problems
View unanswered posts
View posts from last 24 hours

 
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
zhushazang
Tux's lil' helper
Tux's lil' helper


Joined: 01 Dec 2005
Posts: 81

PostPosted: Thu Apr 07, 2011 2:24 pm    Post subject: i915 weird problems Reply with quote

Hello Brothers.

Sometimes using mplayer or mplayer2 in random video files (i mean, avi, wmv, mov, rmvb, etc) my system gonna be unstable.

In such case colors are changes. For example, in yakuake i use black terminal with green fonts. Fonts change to blue. In thunderbird, color fonts change from black to cyan, and so on.

When this shit happens sometimes i can solve the problem restarting X, sometimes only restarting the system "like a Windows® NT".

Looking in dmesg output before restart i see this:

intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 49570, limit 35000
[drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -11 (awaiting
2518546 at 2518545, next 2518551)
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 49566, limit 35000
[drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[drm:init_ring_common] *ERROR* failed to set render ring head to zero ctl
00000000 head 6ca04014 tail 00000000 start 00001000
[drm:init_ring_common] *ERROR* render ring initialization failed ctl 0001f003
head 6ca04014 tail 00000000 start 00001000
[drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[drm:i915_do_wait_request] *ERROR* i915_do_wait_request returns -11 (awaiting
2518555 at 2518545, next 2518556)
[drm:i915_reset] *ERROR* GPU hanging too fast, declaring wedged!
[drm:i915_reset] *ERROR* Failed to reset chip.
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 49729, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 49170, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 38103, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 46063, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 43878, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 44701, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 42681, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 42419, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 41974, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 44362, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 41870, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 41822, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 42129, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 41356, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 41115, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 42033, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 41155, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 42162, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 41109, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 42002, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 41329, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 42565, limit 35000
intel ips 0000:00:1f.6: MCP limit exceeded: Avg power 41526, limit 35000

The system are:

2.6.38-gentoo-r1-asgard #4 SMP Wed Apr 6 09:48:06 BRT 2011 x86_64 Intel(R) Core(TM) i5 CPU M 460 @ 2.53GHz GenuineIntel GNU/Linux


00:02.0 VGA compatible controller: Intel Corporation Core Processor Integrated Graphics Controller (rev 18) (prog-if 00 [VGA controller])
Subsystem: Lenovo Device 3920
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
Latency: 0
Interrupt: pin A routed to IRQ 42
Region 0: Memory at f0000000 (64-bit, non-prefetchable) [size=4M]
Region 2: Memory at d0000000 (64-bit, prefetchable) [size=256M]
Region 4: I/O ports at 1800 [size=8]
Expansion ROM at <unassigned> [disabled]
Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit-
Address: fee0a00c Data: 4171
Capabilities: [d0] Power Management version 2
Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
Capabilities: [a4] PCI Advanced Features
AFCap: TP+ FLR+
AFCtrl: FLR-
AFStatus: TP-
Kernel driver in use: i915
Kernel modules: i915

media-video/mplayer (1.0_rc4_p20110322 / 2.0 )
x11-drivers/xf86-video-intel (2.14.0)
x11-base/xorg-server (1.10.0.901)

Maybe it is only abug, someone know the mailing list addres to report this things to intel driver's developers?

Thanks for now.
Back to top
View user's profile Send private message
BradN
Advocate
Advocate


Joined: 19 Apr 2002
Posts: 2391
Location: Wisconsin (USA)

PostPosted: Thu Apr 07, 2011 10:00 pm    Post subject: Reply with quote

I had a lot of problems like this with i845 on kernel 2.6.36 and earlier where the GPU hang messages would appear but the system would keep running (without any acceleration including xv) until backlight events which would hang the whole system. The GPU hangs would occur mainly during xv operations.

With 2.6.37 and newest intel drivers, the problem went away, leaving only a minor bug related to bitmap style font rendering where sometimes a scanline of a random character would not be rendered, making for example an e look more like a c with the middle line missing. But it was irritating because it would do this for all instances of that character.

I will try updating to 2.6.38 and see if more problems come back.

I can say that the developers have been having tons of problems with various intel graphics related to the KMS change and it seems these chips are somewhat poorly designed and have a lot of dependencies on operation ordering (messed up by memory caching) and perhaps timing that make it hard to write a working driver.
Back to top
View user's profile Send private message
BradN
Advocate
Advocate


Joined: 19 Apr 2002
Posts: 2391
Location: Wisconsin (USA)

PostPosted: Thu Apr 07, 2011 11:51 pm    Post subject: Reply with quote

Update: I've upgraded to 2.6.38 and I'm noticing severe video corruption problems, no GPU hangs yet but it wouldn't surprise me if that occurs as well.

I recommend trying 2.6.37 in combination with xf86-video-intel 2.14.0 - this has given me the best results so far.
Back to top
View user's profile Send private message
zhushazang
Tux's lil' helper
Tux's lil' helper


Joined: 01 Dec 2005
Posts: 81

PostPosted: Sun May 08, 2011 10:39 pm    Post subject: Reply with quote

Intel vga and linux is very complicated and poor supported.
Back to top
View user's profile Send private message
doomass
n00b
n00b


Joined: 09 May 2011
Posts: 3

PostPosted: Mon May 09, 2011 5:17 am    Post subject: Reply with quote

I was having similar problems with my I915 video. Enabling DRM_I915_KMS in the kernel seemed to fix it.


│ Symbol: DRM_I915_KMS [=y] │
│ Type : boolean │
│ Prompt: Enable modesetting on intel by default │
│ Defined at drivers/gpu/drm/Kconfig:126 │
│ Depends on: <choice> && DRM_I915 [=y] │
│ Location: │
│ -> Device Drivers │
│ -> Graphics support │
│ -> Direct Rendering Manager (XFree86 4.1.0 and higher DRI support) (DRM [=y]) │
│ -> Intel 830M, 845G, 852GM, 855GM, 865G (<choice> [=y]) │
│ -> i915 driver (DRM_I915 [=y])
Back to top
View user's profile Send private message
BradN
Advocate
Advocate


Joined: 19 Apr 2002
Posts: 2391
Location: Wisconsin (USA)

PostPosted: Mon May 09, 2011 10:42 pm    Post subject: Reply with quote

It seems everything about the KMS switchover has been problematic for Intel drivers... I have a feeling things should have been done in a testing tree until better stability was achieved. I mean, the old drivers were working pretty well for most people until this knocked everything around.

But at least this way should force the new KMS system to stabilize faster because people are actually using it (or trying to...), so maybe some months from now we'll be happy it's over with, and have better stability than before due to the kernel enforcing valid device interactions. If it's working properly with intel drivers, there might be less of a barrier for nvidia, etc, using a KMS based system once it's shown to be useful.

I must say, I'm pretty impressed with how far Nouveau has come, at least for an old geforce 2 card I tried it with. It's actually usable now, if not totally bug free. And it's got nearly instant switching between X and linux framebuffer terminals due to the kernel controlling the low level video operations. At least it's keeping our dual pentium 3 system working - the nvidia drivers were less stable the last time we updated them, and a locking up machine isn't very useful for real work.
Back to top
View user's profile Send private message
generatorglukoff
n00b
n00b


Joined: 08 Aug 2008
Posts: 18

PostPosted: Sun Jul 10, 2011 10:41 pm    Post subject: Reply with quote

I've just had this bug on 2.6.39.2. It happens seldom and last one was not attached to anything GPU related (no suspend-resume, no 3d graphics/games, only web surfing).

I'm trying to post it into freedesktop bugzilla, but, wtf, it's full of reports for this bug and I don't even know where to post.
_________________
Gentoo Router Overlay: http://bitbucket.org/generatorglukoff/gentoo-router-overlay
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum