Server freezes with latest commit on 22/06/2011

Submitted by Emanuele

Assigned to Default DRI bug account

Link to original bug (#38694)

Description

With this commit my PC freezes sometimes when get on/off standby:
$> xset dpms force standby

I have always had these problems, please refer to:
https://bugzilla.kernel.org/show_bug.cgi?id=spina

but they are occurred rarely, recently. With your commin on 17/06/2011 (commit 122b471f734aa07427b01d4bec35ff1ac28290b5) the problem seemed to be solved. However, with the commit in object it always happens.

I can see a blinking screen but I have to halt the PC with magic key.

Best regards,
Emanuele

added Radeon bugzilla labels

💬 Emanuele said:

Hi,
I have tried some commit in reverse mode.

• Jun 17: same problem.
  
• Jun 16: same problem, but less often.
  
• Jun 13: so far so good.
  

Emanuele

💬 Alex Deucher @agd5f said:

Please attach your xorg log and dmesg output.

📎 Emanuele uploaded an attachment:

Attachment 48905, "Xorg.log":
Xorg.0.log

📎 Emanuele uploaded an attachment:

As I have already said in the other bugzilla, when my I do:
xset dpms force standby
or
echo "low" > /sys/class/drm/card0/device/power_profile
the first time I have this message:
---
NMI: PCI system error (SERR) for reason a1 on CPU 0.
Dazed and confused, but trying to continue
---
but, if I follow you suggestion this disappears:
---
You can disable the PCIE lane changes by removing the
call to radeon_set_pcie_lanes() in rs600_pm_misc() in rs600.c.

Attachment 48906, "dmesg on boot":
dmesg

💬 Emanuele said:

Hi Alex,
I saw your last commit one minute ago. Have you discovered what happens on my PC, yet? Do you need more infos?

Best regards,
Emanuele

💬 Alex Deucher @agd5f said:

Have you changed the kernel you are using or just the ddx? With KMS, there ddx doesn't really do much with respect to modesetting. It just calls into the kernel.

💬 Emanuele said:

Alex,
I'm using the latest vanilla stable kernel (2.6.39.3). Sorry but, what is the ddx?

Emanuele

💬 Alex Deucher @agd5f said:

(In reply to comment 7)

Alex,
I'm using the latest vanilla stable kernel (2.6.39.3). Sorry but, what is the
ddx?

The ddx is the radeon X driver (xf86-video-ati).

💬 Emanuele said:

Thanks for your reply. So, is there a bug in the Kernel? I use to compile the kernel by myself, did I forget something?

But, how is possible that with different commit of ddx there are different situations?

Thanks a lot,
Emanuele

💬 Alex Deucher @agd5f said:

It's hard to say. That's why I was trying to figure out which components you changed (just ddx, ddx and kernel, ddx, kernel, and mesa. etc.) as they could all be to blame potentially. It might be a 3D screen saver that kicks in and hangs the card due to a bug in the 3D driver.

💬 Emanuele said:

I have never used screen saver. My Server X is configured in this way:
xset s 300
xset s blank

When I want to force the standby I use:
xset dpms force standby

I can do all the tests you want,
Emanuele

💬 Emanuele said:

Hi Alex,
now I'm using kernel v3.0.4. I noticed that I have freeze also when I put my current wire into socket, sometimes.
Remember that I receive this message on standby or when I do:
---
echo "low" > /sys/class/drm/card0/device/power_profile
---
NMI: PCI system error (SERR) for reason b1 on CPU 0.
Dazed and confused, but trying to continue
---
and I do this when the current wire is pulled off whereas I do:
---
echo "auto" > /sys/class/drm/card0/device/power_profile
---
when is inserted.

I have noticed also that sometimes I receive that message at boot time:
---[drm] Initialized drm 1.1.0 20060810
[drm] radeon defaulting to kernel modesetting.
[drm] radeon kernel modesetting enabled.
radeon 0000:01:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
radeon 0000:01:00.0: setting latency timer to 64
[drm] initializing kernel modesetting (RV515 0x1002:0x7145 0x1028:0x2003).
[drm] register mmio base: 0xEFDF0000
[drm] register mmio size: 65536
ATOM BIOS: M54P
[drm] Generation 2 PCI interface, using max accessible memory
radeon 0000:01:00.0: VRAM: 256M 0x0000000000000000 - 0x000000000FFFFFFF (128M used)
radeon 0000:01:00.0: GTT: 512M 0x0000000010000000 - 0x000000002FFFFFFF
[drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
[drm] Driver supports precise vblank timestamp query.
radeon 0000:01:00.0: irq 43 for MSI/MSI-X
radeon 0000:01:00.0: radeon: using MSI.
[drm] radeon: irq initialized.
[drm] Detected VRAM RAM=256M, BAR=256M
[drm] RAM width 64bits DDR
[TTM] Zone kernel: Available graphics memory: 1028024 kiB.
[TTM] Initializing pool allocator.
[drm] radeon: 128M of VRAM memory ready
[drm] radeon: 512M of GTT memory ready.
[drm] GART: num cpu pages 131072, num gpu pages 131072
NMI: PCI system error (SERR) for reason b1 on CPU 0.
Dazed and confused, but trying to continue
[drm] radeon: 1 quad pipes, 1 z pipes initialized.
[drm] PCIE GART of 512M enabled (table at 0x00040000).
radeon 0000:01:00.0: WB enabled
[drm] Loading R500 Microcode
[drm] radeon: ring at 0x0000000010001000
[drm] ring test succeeded in 10 usecs
[drm] radeon: ib pool ready.
[drm] ib test succeeded in 0 usecs
[drm] Radeon Display Connectors
[drm] Connector 0:
[drm] VGA
[drm] DDC: 0x7e40 0x7e40 0x7e44 0x7e44 0x7e48 0x7e48 0x7e4c 0x7e4c
[drm] Encoders:
[drm] CRT1: INTERNAL_KLDSCP_DAC1
[drm] Connector 1:
[drm] LVDS
[drm] DDC: 0x7e30 0x7e30 0x7e34 0x7e34 0x7e38 0x7e38 0x7e3c 0x7e3c
[drm] Encoders:
[drm] LCD1: INTERNAL_LVTM1
[drm] Connector 2:
[drm] S-video
[drm] Encoders:
[drm] TV1: INTERNAL_KLDSCP_DAC2
[drm] Radeon display connector VGA-1: No monitor connected or invalid EDID
[drm] Radeon display connector LVDS-1: Found valid EDID
[drm] radeon: power management initialized
[drm] fb mappable at 0xD00C0000
[drm] vram apper at 0xD0000000
[drm] size 4096000
[drm] fb depth is 24
[drm] pitch is 5120
fbcon: radeondrmfb (fb0) is primary device
Console: switching to colour frame buffer device 160x50
fb0: radeondrmfb frame buffer device
drm: registered panic notifier
[drm] Initialized radeon 2.10.0 20080528 for 0000:01:00.0 on minor 0
---

I hope that this can help you to investigate,
Emanuele

💬 Michel Dänzer @daenzer said:

(In reply to comment 4)

but, if I follow you suggestion this disappears:
---
You can disable the PCIE lane changes by removing the
call to radeon_set_pcie_lanes() in rs600_pm_misc() in rs600.c.

Did removing that call only cause the NMI message to disappear, or also the freezes? Is it still the case with a current kernel?

💬 Michel Dänzer @daenzer said:

BTW, please refer to Git commits by their commit IDs rather than by date.

💬 Emanuele said:

Hi Michel,
(In reply to comment 13)

(In reply to comment 4)

but, if I follow you suggestion this disappears:
---
You can disable the PCIE lane changes by removing the
call to radeon_set_pcie_lanes() in rs600_pm_misc() in rs600.c.

Did removing that call only cause the NMI message to disappear, or also the
freezes?

we'll see. I disabled that call and the NMI message has been desappeared. I also tryed 'xset dpms force standby' some times and so far so good.

Is it still the case with a current kernel?
---
$> uname -r
3.0.4
---

With 3.0.X kernel freezes are different. Now I can move the mouse or I can use keyboard, sometimes. In these cases, I'm not obliged to shutdown PC but I can reboot it: I have to press the halt button on my case and I can force reboot with CTRl+ALT+DEL after some seconds (I think when X server is killed, but I can't see nothing).

Thank for all and best regards,
Emanuele

💬 Emanuele said:

(In reply to comment 13)

Did removing that call only cause the NMI message to disappear, or also the
freezes? Is it still the case with a current kernel?

Michel,
after a week we can say: "yes, also freezes have been desappear". Now, all is good!!!

But, when I update the kernel with new increments, patch obvously fails for this file.

Emanuele

closed