Skip to contentSkip to conversation

Linux 3.16-rc1: drm:intel_dp_start_link_train *ERROR* too many full retries, give up

DISCUSSION Oliver Hartkopp 3 participants 6 messages Jun 16, 2014 – Jun 30, 2014

lore
Share & tools
lore
#0 lore raw

Hey all,

with my Debian unstable the upgrade from 3.15 to 3.16-rc1 failed with this fancy [drm:intel_dp_start_link_train *ERROR* too many full retries, give up] message loop.

lightdm/XFCE4 doesn't start anymore (it tries several times but fails always).

Hm - I just found this stuff in my stock Debian Kernel too:
3.14-1-686-pae #1 SMP Debian 3.14.5-1 (2014-06-05) i686 GNU/Linux

[    6.461232] [drm:intel_dp_start_link_train] *ERROR* too many full retries, give up
[    6.873772] dell_wmi: Received unknown WMI event (0x11)
[    6.884336] Console: switching to colour frame buffer device 240x67
[    6.948511] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device
[    6.948513] i915 0000:00:02.0: registered panic notifier
[    7.430429] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0

But at least it starts with the 3.14 (any the latest 3.15).

I have a Intel(R) Core(TM) i7 CPU M 640 @ 2.80GHz with Intel HD graphics.

Any idea?

Regards,
Oliver

#1 lore raw
Quoted 4 lines · Oliver Hartkopp

Hey all,

with my Debian unstable the upgrade from 3.15 to 3.16-rc1 failed with this fancy [drm:intel_dp_start_link_train *ERROR* too many full retries, give up] message loop.

This message in dmesg and

Quoted 1 line

lightdm/XFCE4 doesn't start anymore (it tries several times but fails always).

xfce not starting shouldn't be related. Are you sure xfce isn't starting and that it's not just a black screen, but otherwise working X?

Please attach complete dmesg with drm.debug=0xe added to your kernel cmdline and Xorg.log.

Thanks, Daniel

Quoted 20 lines

Hm - I just found this stuff in my stock Debian Kernel too:
3.14-1-686-pae #1 SMP Debian 3.14.5-1 (2014-06-05) i686 GNU/Linux

[    6.461232] [drm:intel_dp_start_link_train] *ERROR* too many full retries, give up
[    6.873772] dell_wmi: Received unknown WMI event (0x11)
[    6.884336] Console: switching to colour frame buffer device 240x67
[    6.948511] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device
[    6.948513] i915 0000:00:02.0: registered panic notifier
[    7.430429] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0

But at least it starts with the 3.14 (any the latest 3.15).

I have a Intel(R) Core(TM) i7 CPU M 640 @ 2.80GHz with Intel HD graphics.

Any idea?

Regards,
Oliver

Signature · 3 lines
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
#2 lore raw

Hello Daniel,

thanks for your fast reply.

I started the system with drm.debug=0xe with the same effect - therefore I was not able to log in %-(

Btw. when starting into single user mode, I was able to say 'xinit' which produced the borderless xterm at the top left position. So obviously the display is working. Running lightdm from the console in single user mode does not work (dbus missing?). Starting lightdm a second time kills the console.

I think I need to dig into the lightdm and startup stuff. So this might not be an i915 issue. Sorry for the noise.

I'll send an update if I discover any new info ...

Thanks for your support so far,
Oliver

Quoted 43 lines · Daniel Vetter
Quoted 4 lines · Oliver Hartkopp

Hey all,

with my Debian unstable the upgrade from 3.15 to 3.16-rc1 failed with this fancy [drm:intel_dp_start_link_train *ERROR* too many full retries, give up] message loop.

This message in dmesg and

Quoted 1 line

lightdm/XFCE4 doesn't start anymore (it tries several times but fails always).

xfce not starting shouldn't be related. Are you sure xfce isn't starting and that it's not just a black screen, but otherwise working X?

Please attach complete dmesg with drm.debug=0xe added to your kernel cmdline and Xorg.log.

Thanks, Daniel

Quoted 20 lines

Hm - I just found this stuff in my stock Debian Kernel too:
3.14-1-686-pae #1 SMP Debian 3.14.5-1 (2014-06-05) i686 GNU/Linux

[    6.461232] [drm:intel_dp_start_link_train] *ERROR* too many full retries, give up
[    6.873772] dell_wmi: Received unknown WMI event (0x11)
[    6.884336] Console: switching to colour frame buffer device 240x67
[    6.948511] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device
[    6.948513] i915 0000:00:02.0: registered panic notifier
[    7.430429] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0

But at least it starts with the 3.14 (any the latest 3.15).

I have a Intel(R) Core(TM) i7 CPU M 640 @ 2.80GHz with Intel HD graphics.

Any idea?

Regards,
Oliver

#3 lore raw

Hello Daniel,

just to close my request:

The latest git pull from Linus' tree fixed my issue.

vmlinuz-3.16.0-rc2-00211-gd7933ab (26.6. 21:19) was bad vmlinuz-3.16.0-rc2-00319-g3e7b256 (28.6. 09:23) works again

There are two pulls with 'drm-fixes' and 'x86/urgent' in this area. Don't know what it was in detail - but it's fixed now (lightdm/xfce4 runs).

Thanks,
Oliver

Quoted 64 lines · Oliver Hartkopp

Hello Daniel,

thanks for your fast reply.

I started the system with drm.debug=0xe with the same effect - therefore I was not able to log in %-(

Btw. when starting into single user mode, I was able to say 'xinit' which produced the borderless xterm at the top left position. So obviously the display is working. Running lightdm from the console in single user mode does not work (dbus missing?). Starting lightdm a second time kills the console.

I think I need to dig into the lightdm and startup stuff. So this might not be an i915 issue. Sorry for the noise.

I'll send an update if I discover any new info ...

Thanks for your support so far,
Oliver

Quoted 43 lines · Daniel Vetter
Quoted 4 lines · Oliver Hartkopp

Hey all,

with my Debian unstable the upgrade from 3.15 to 3.16-rc1 failed with this fancy [drm:intel_dp_start_link_train *ERROR* too many full retries, give up] message loop.

This message in dmesg and

Quoted 1 line

lightdm/XFCE4 doesn't start anymore (it tries several times but fails always).

xfce not starting shouldn't be related. Are you sure xfce isn't starting and that it's not just a black screen, but otherwise working X?

Please attach complete dmesg with drm.debug=0xe added to your kernel cmdline and Xorg.log.

Thanks, Daniel

Quoted 20 lines

Hm - I just found this stuff in my stock Debian Kernel too:
3.14-1-686-pae #1 SMP Debian 3.14.5-1 (2014-06-05) i686 GNU/Linux

[    6.461232] [drm:intel_dp_start_link_train] *ERROR* too many full retries, give up
[    6.873772] dell_wmi: Received unknown WMI event (0x11)
[    6.884336] Console: switching to colour frame buffer device 240x67
[    6.948511] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device
[    6.948513] i915 0000:00:02.0: registered panic notifier
[    7.430429] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0

But at least it starts with the 3.14 (any the latest 3.15).

I have a Intel(R) Core(TM) i7 CPU M 640 @ 2.80GHz with Intel HD graphics.

Any idea?

Regards,
Oliver

#4 lore raw
Quoted 11 lines · Oliver Hartkopp

Hello Daniel,

just to close my request:

The latest git pull from Linus' tree fixed my issue.

vmlinuz-3.16.0-rc2-00211-gd7933ab (26.6. 21:19) was bad vmlinuz-3.16.0-rc2-00319-g3e7b256 (28.6. 09:23) works again

There are two pulls with 'drm-fixes' and 'x86/urgent' in this area. Don't know what it was in detail - but it's fixed now (lightdm/xfce4 runs).

Thanks for reporting back, I'm glad it works for you now.

However, for closure, I would appreciate it if you could do a reverse bisect to find out which commit fixed things for you.

Thanks,
Jani.

Quoted 74 lines

Thanks,
Oliver

Quoted 15 lines · Oliver Hartkopp

Hello Daniel,

thanks for your fast reply.

I started the system with drm.debug=0xe with the same effect - therefore I was not able to log in %-(

Btw. when starting into single user mode, I was able to say 'xinit' which produced the borderless xterm at the top left position. So obviously the display is working. Running lightdm from the console in single user mode does not work (dbus missing?). Starting lightdm a second time kills the console.

I think I need to dig into the lightdm and startup stuff. So this might not be an i915 issue. Sorry for the noise.

Quoted 49 lines

I'll send an update if I discover any new info ...

Thanks for your support so far,
Oliver

Quoted 43 lines · Daniel Vetter
Quoted 4 lines · Oliver Hartkopp

Hey all,

with my Debian unstable the upgrade from 3.15 to 3.16-rc1 failed with this fancy [drm:intel_dp_start_link_train *ERROR* too many full retries, give up] message loop.

This message in dmesg and

Quoted 1 line

lightdm/XFCE4 doesn't start anymore (it tries several times but fails always).

xfce not starting shouldn't be related. Are you sure xfce isn't starting and that it's not just a black screen, but otherwise working X?

Please attach complete dmesg with drm.debug=0xe added to your kernel cmdline and Xorg.log.

Thanks, Daniel

Quoted 20 lines

Hm - I just found this stuff in my stock Debian Kernel too:
3.14-1-686-pae #1 SMP Debian 3.14.5-1 (2014-06-05) i686 GNU/Linux

[    6.461232] [drm:intel_dp_start_link_train] *ERROR* too many full retries, give up
[    6.873772] dell_wmi: Received unknown WMI event (0x11)
[    6.884336] Console: switching to colour frame buffer device 240x67
[    6.948511] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device
[    6.948513] i915 0000:00:02.0: registered panic notifier
[    7.430429] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0

But at least it starts with the 3.14 (any the latest 3.15).

I have a Intel(R) Core(TM) i7 CPU M 640 @ 2.80GHz with Intel HD graphics.

Any idea?

Regards,
Oliver

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Signature · 1 line
Jani Nikula, Intel Open Source Technology Center
#5 lore raw

Hello Jani,

Quoted 7 lines · Jani Nikula
Quoted 5 lines · Oliver Hartkopp

The latest git pull from Linus' tree fixed my issue.

vmlinuz-3.16.0-rc2-00211-gd7933ab (26.6. 21:19) was bad vmlinuz-3.16.0-rc2-00319-g3e7b256 (28.6. 09:23) works again

Quoted 7 lines
Quoted 2 lines

There are two pulls with 'drm-fixes' and 'x86/urgent' in this area. Don't know what it was in detail - but it's fixed now (lightdm/xfce4 runs).

Thanks for reporting back, I'm glad it works for you now.

However, for closure, I would appreciate it if you could do a reverse bisect to find out which commit fixed things for you.

I checked the 'x86/urgent' patches which talk about some major glibc breakage that have been fixed in that pull request. So I just reverted the patches form the 'drm-fixes' to make sure that my problem was not introduced by the i915/drm stuff.

It finally made do difference whether the last 'drm-fixes' where applied or not. It obviously was the glibc issue for x32 which killed my lightdm/xfce4 startup.

Best regards,
Oliver