Bug 38622 - [radeon cayman regresion] artefacts on screen
Summary: [radeon cayman regresion] artefacts on screen
Status: RESOLVED CODE_FIX
Alias: None
Product: Drivers
Classification: Unclassified
Component: Video(DRI - non Intel) (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: drivers_video-dri
URL:
Keywords:
Depends on:
Blocks: 36912
  Show dependency tree
 
Reported: 2011-07-01 09:35 UTC by Marek Hobler, 'neutrinus'
Modified: 2012-08-24 15:06 UTC (History)
5 users (show)

See Also:
Kernel Version: 3.0.0-rc3
Tree: Mainline
Regression: Yes


Attachments
dmesg_3.0.0-rc2-amd64 (62.68 KB, text/plain)
2011-07-01 10:52 UTC, Marek Hobler, 'neutrinus'
Details
dmesg_3.0.0-rc5-amd64 (58.23 KB, text/plain)
2011-07-01 10:53 UTC, Marek Hobler, 'neutrinus'
Details
bisect log (3.92 KB, text/plain)
2011-07-01 19:18 UTC, Marek Hobler, 'neutrinus'
Details
dmesg when not working and skips were applied (83.17 KB, text/plain)
2011-07-04 21:44 UTC, Marek Hobler, 'neutrinus'
Details
Xorg.log when not working and skips were applied (60.36 KB, text/plain)
2011-07-04 21:46 UTC, Marek Hobler, 'neutrinus'
Details

Description Marek Hobler, 'neutrinus' 2011-07-01 09:35:40 UTC
When running kernel versions => 3.0.0-rc3 there are a lot of artifacts on screen when runing xorg. 

Kernel 3.0.0-rc2 and olders were producing clean image, right now there are artifacts on screen, making any fonts/graphics unreadable. 3.0.0-rc[3-5] are affected. It fully reproducible: just run kde :) Its happends everytime when I switch 3.0.0-rc2 kernel to something newer.

I'm using debian AMD64 sid+experimental, kernel images from experimental. I'm using xorg "radeon" driver. Mesa: 7.11~0-2.

There are no Ops, no strange logs in dmesg...

CPU: Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz
GPU: AMD Radeon HD 6970 with six fullhd monitors (4x DP and 2x DVI)

Modules:
hmac 12923 0 - Live 0xffffffffa054c000
nls_utf8 12456 0 - Live 0xffffffffa0509000
cifs 242524 0 - Live 0xffffffffa050f000
parport_pc 22395 0 - Live 0xffffffffa0461000
ppdev 12763 0 - Live 0xffffffffa0436000
lp 17270 0 - Live 0xffffffffa0446000
parport 31929 3 parport_pc,ppdev,lp, Live 0xffffffffa0500000
fuse 62334 1 - Live 0xffffffffa04aa000
nfsd 260601 2 - Live 0xffffffffa04bf000
nfs 259361 0 - Live 0xffffffffa0469000
lockd 71503 2 nfsd,nfs, Live 0xffffffffa044e000
fscache 36663 2 cifs,nfs, Live 0xffffffffa043c000
auth_rpcgss 37204 2 nfsd,nfs, Live 0xffffffffa042b000
nfs_acl 12511 2 nfsd,nfs, Live 0xffffffffa03ec000
sunrpc 172969 6 nfsd,nfs,lockd,auth_rpcgss,nfs_acl, Live 0xffffffffa03ff000
max6650 12900 0 - Live 0xffffffffa03fa000
coretemp 12939 0 - Live 0xffffffffa03f5000
hwmon_vid 12388 0 - Live 0xffffffffa03e0000
loop 22711 0 - Live 0xffffffffa03e5000
sha256_generic 16797 2 - Live 0xffffffffa0387000
aesni_intel 50644 24 - Live 0xffffffffa03d2000
cryptd 14549 9 aesni_intel, Live 0xffffffffa00db000
aes_x86_64 16796 1 aesni_intel, Live 0xffffffffa034d000
aes_generic 33026 2 aesni_intel,aes_x86_64, Live 0xffffffffa0343000
cbc 12755 0 - Live 0xffffffffa017f000
dm_crypt 22611 1 - Live 0xffffffffa0380000
dm_mod 63164 3 dm_crypt, Live 0xffffffffa0332000
snd_hda_codec_hdmi 26321 1 - Live 0xffffffffa0289000
snd_hda_codec_realtek 278027 1 - Live 0xffffffffa038d000
snd_usb_audio 84855 1 - Live 0xffffffffa036a000
snd_usbmidi_lib 23475 1 snd_usb_audio, Live 0xffffffffa0363000
uvcvideo 57791 0 - Live 0xffffffffa0353000
snd_seq_midi 12848 0 - Live 0xffffffffa00ef000
videodev 66713 1 uvcvideo, Live 0xffffffffa0240000
snd_seq_midi_event 13316 1 snd_seq_midi, Live 0xffffffffa01f8000
media 18184 2 uvcvideo,videodev, Live 0xffffffffa023a000
v4l2_compat_ioctl32 16567 1 videodev, Live 0xffffffffa0234000
snd_rawmidi 23297 2 snd_usbmidi_lib,snd_seq_midi, Live 0xffffffffa01c4000
joydev 17262 0 - Live 0xffffffffa0185000
radeon 648169 3 - Live 0xffffffffa0292000
ttm 52979 1 radeon, Live 0xffffffffa027b000
drm_kms_helper 27216 1 radeon, Live 0xffffffffa01f0000
drm 163231 5 radeon,ttm,drm_kms_helper, Live 0xffffffffa0252000
snd_hda_intel 26140 5 - Live 0xffffffffa01d0000
snd_hda_codec 72699 3 snd_hda_codec_hdmi,snd_hda_codec_realtek,snd_hda_intel, Live 0xffffffffa01ac000
evdev 17558 13 - Live 0xffffffffa0179000
i2c_algo_bit 12850 1 radeon, Live 0xffffffffa01a3000
snd_hwdep 13186 2 snd_usb_audio,snd_hda_codec, Live 0xffffffffa018d000
snd_pcm 68104 5 snd_hda_codec_hdmi,snd_usb_audio,snd_hda_intel,snd_hda_codec, Live 0xffffffffa0222000
snd_seq 45208 2 snd_seq_midi,snd_seq_midi_event, Live 0xffffffffa0215000
snd_timer 22581 3 snd_pcm,snd_seq, Live 0xffffffffa0199000
snd_seq_device 13137 3 snd_seq_midi,snd_rawmidi,snd_seq, Live 0xffffffffa0210000
i2c_i801 16870 0 - Live 0xffffffffa01ea000
snd 52823 22 snd_hda_codec_hdmi,snd_hda_codec_realtek,snd_usb_audio,snd_usbmidi_lib,snd_rawmidi,snd_hda_intel,snd_hda_codec,snd_hwdep,snd_pcm,snd_seq,snd_timer,snd_seq_device, Live 0xffffffffa0202000
i2c_core 23909 7 max6650,videodev,radeon,drm_kms_helper,drm,i2c_algo_bit,i2c_i801, Live 0xffffffffa01e3000
pcspkr 12579 0 - Live 0xffffffffa01fd000
serio_raw 12846 0 - Live 0xffffffffa01bf000
soundcore 13152 1 snd, Live 0xffffffffa01de000
snd_page_alloc 13043 2 snd_hda_intel,snd_pcm, Live 0xffffffffa01d9000
power_supply 13475 1 radeon, Live 0xffffffffa01cb000
processor 27942 8 - Live 0xffffffffa0171000
button 12930 0 - Live 0xffffffffa0169000
thermal_sys 17949 1 processor, Live 0xffffffffa0193000
ext4 318236 4 - Live 0xffffffffa011a000
mbcache 13066 1 ext4, Live 0xffffffffa003d000
jbd2 62709 1 ext4, Live 0xffffffffa0109000
crc16 12343 1 ext4, Live 0xffffffffa0060000
sg 25985 0 - Live 0xffffffffa00e7000
sr_mod 21811 0 - Live 0xffffffffa00e0000
cdrom 35093 1 sr_mod, Live 0xffffffffa00ff000
sd_mod 36259 7 - Live 0xffffffffa0096000
crc_t10dif 12348 1 sd_mod, Live 0xffffffffa002d000
usbhid 40516 0 - Live 0xffffffffa00f4000
hid 73172 1 usbhid, Live 0xffffffffa00c8000
ahci 25089 4 - Live 0xffffffffa008e000
libahci 22767 1 ahci, Live 0xffffffffa0026000
libata 149043 2 ahci,libahci, Live 0xffffffffa00a2000
scsi_mod 162442 4 sg,sr_mod,sd_mod,libata, Live 0xffffffffa0065000
xhci_hcd 68199 0 - Live 0xffffffffa004e000
r8169 42411 0 - Live 0xffffffffa0042000
mii 12675 1 r8169, Live 0xffffffffa0021000
ehci_hcd 40090 0 - Live 0xffffffffa0032000
usbcore 128338 7 snd_usb_audio,snd_usbmidi_lib,uvcvideo,usbhid,xhci_hcd,ehci_hcd, Live 0xffffffffa0000000


lspci -vvv (part of)
01:00.0 VGA compatible controller: ATI Technologies Inc Cayman XT [AMD Radeon HD 6900 Series] (prog-if 00 [VGA controller])
        Subsystem: ASUSTeK Computer Inc. Device 03bc
        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, Cache Line Size: 4 bytes
        Interrupt: pin A routed to IRQ 52
        Region 0: Memory at e0000000 (64-bit, prefetchable) [size=256M]
        Region 2: Memory at fbec0000 (64-bit, non-prefetchable) [size=128K]
        Region 4: I/O ports at ee00 [size=256]
        [virtual] Expansion ROM at fbe00000 [disabled] [size=128K]
        Capabilities: <access denied>
        Kernel driver in use: radeon




If there is any more information needed I would be more than happy to provide it.
Comment 1 Michel Dänzer 2011-07-01 10:07:50 UTC
(In reply to comment #0)
> When running kernel versions => 3.0.0-rc3 there are a lot of artifacts on
> screen when runing xorg. 

Are the artifacts captured in a screenshot, or only in a photograph taken from the screen? It might be helpful if you could attach a picture showing the artifacts either way.

> Kernel 3.0.0-rc2 and olders were producing clean image, [...]

Can you bisect?
Comment 2 Marek Hobler, 'neutrinus' 2011-07-01 10:40:02 UTC
(In reply to comment #1)
> Are the artifacts captured in a screenshot, or only in a photograph taken
> from
> the screen? It might be helpful if you could attach a picture showing the
> artifacts either way.
both on screenshoot (using ksnapshot) and photograph

screenshoot: (~10MB)
http://neutrinus.com/tmp/kaczka.png

(this is worst case - normally its not like whole windows garbaged)

> Can you bisect?

yes :) (but it may take some time)
Comment 3 Marek Hobler, 'neutrinus' 2011-07-01 10:52:57 UTC
Created attachment 64212 [details]
dmesg_3.0.0-rc2-amd64
Comment 4 Marek Hobler, 'neutrinus' 2011-07-01 10:53:26 UTC
Created attachment 64222 [details]
dmesg_3.0.0-rc5-amd64
Comment 5 Marek Hobler, 'neutrinus' 2011-07-01 19:18:29 UTC
Created attachment 64412 [details]
bisect log

skips were entered when xserver didn't even start properly - there was no way to test if there are artefacts

Hope that it will help, if there is anything more that I can provide - just ask for it.
Comment 6 Marek Hobler, 'neutrinus' 2011-07-03 21:20:45 UTC
I have just tested current master Linus tree - it still broken (so no hope that it was already fixed in rc5+)

Would it help anything if I record video how the artifact change when using system?
Comment 7 Michel Dänzer 2011-07-04 17:54:32 UTC
(In reply to comment #5)
> skips were entered when xserver didn't even start properly - there was no way
> to test if there are artefacts

Can you provide Xorg.0.log and dmesg for such a failure?

Your log narrows things down somewhat, but I'm afraid there's still a couple of potentially relevant changes in there.


(In reply to comment #6)
> Would it help anything if I record video how the artifact change when using
> system?

Possibly.
Comment 8 Marek Hobler, 'neutrinus' 2011-07-04 21:44:40 UTC
Created attachment 64622 [details]
dmesg when not working and skips were applied
Comment 9 Marek Hobler, 'neutrinus' 2011-07-04 21:46:07 UTC
Created attachment 64632 [details]
Xorg.log when not working and skips were applied
Comment 10 Marek Hobler, 'neutrinus' 2011-07-04 21:48:16 UTC
(In reply to comment #7)
> (In reply to comment #5)
> > skips were entered when xserver didn't even start properly - there was no
> way
> > to test if there are artefacts
> 
> Can you provide Xorg.0.log and dmesg for such a failure?
git checkout 2f2f96d1ea72e6602ae440ac2867a6004edb37a1

all monitors went black, after few seconds backlight turns off, then on and again few seconds of nothing (video: http://www.youtube.com/watch?v=_tqZGrNexqo).

dmesg and xorg.log already attached
Comment 11 Marek Hobler, 'neutrinus' 2011-07-04 22:32:23 UTC
(In reply to comment #7)
> (In reply to comment #6)
> > Would it help anything if I record video how the artifact change when using
> > system?
> 
> Possibly.
so, there is video of artifacts when runing debian kernel 3.0.0-rc5:
http://www.youtube.com/watch?v=xYjZDSdvz4s

(sorry for poor quality - its recorded using cellphone)
Comment 12 Michel Dänzer 2011-08-29 17:33:42 UTC
This could be the same as https://bugs.freedesktop.org/show_bug.cgi?id=40221 .
Comment 13 Marek Hobler, 'neutrinus' 2011-09-02 21:17:04 UTC
I just checked kernel: vmlinuz-3.1.0-rc4-amd64 (from debian package) and problem is still there...

Michel: yes, it looks similar
Comment 14 Dave Airlie 2012-02-12 16:50:31 UTC
should be fixed in linux v3.2-rc1.

Note You need to log in before you can comment on or make changes to this bug.