http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 Takashi Iwai changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |patrik.jakobsson@suse.com, | |tiwai@suse.com, | |tzimmermann@suse.com -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c3 --- Comment #3 from Patrik Jakobsson --- One person claims a similar problem disappeared between v5.14.6 and v5.15.7. Possible fixes are: 9af4bf2171c1 drm/i915/dp: return proper DPRX link training result c8dead5751b8 drm/i915/dp: Use max params for panels < eDP 1.4 Both were backported to the 15.4 kernel in mid/late September. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c5 --- Comment #5 from Andreas --- Created attachment 854474 --> http://bugzilla.opensuse.org/attachment.cgi?id=854474&action=edit hwinfo -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c7 --- Comment #7 from Takashi Iwai --- (In reply to Patrik Jakobsson from comment #2)

Possible upstream bug report: https://gitlab.freedesktop.org/drm/intel/-/issues/4458 (In reply to Patrik Jakobsson from comment #3) One person claims a similar problem disappeared between v5.14.6 and v5.15.7.

As far as I checked, both of the above are about Thinkpad P15 Gen 1, and it's also with Cometlake-H. So this could be a problem with that chip.

Possible fixes are: 9af4bf2171c1 drm/i915/dp: return proper DPRX link training result c8dead5751b8 drm/i915/dp: Use max params for panels < eDP 1.4

Both were backported to the 15.4 kernel in mid/late September.

Right, those should have been already covered. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c9 --- Comment #9 from Andreas --- Created attachment 854476 --> http://bugzilla.opensuse.org/attachment.cgi?id=854476&action=edit hwinfo -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c11 --- Comment #11 from Andreas --- There were such errors in Leap 15.3, but there were fewer of them, in my opinion 2 messages. There are also these errors in Tumbleweed, I tested up to kernel 5.15.5 -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c13 --- Comment #13 from Andreas --- Created attachment 854480 --> http://bugzilla.opensuse.org/attachment.cgi?id=854480&action=edit hwinfo -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c15 --- Comment #15 from Takashi Iwai --- There has been a fix for a similar issue https://gitlab.freedesktop.org/drm/intel/-/issues/203 and the upstream commit f542d671ffcec772a561cd41c7e2394392d9dafb was merged in 5.11. I wonder whether this is the same problem reappearing. Does this happen no matter whether you plug the HDMI monitor (if any) before or after the boot? The bug above mentioned that it happens when the machine was booted without HDMI plugged. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c17 --- Comment #17 from Andreas --- I also connected a monitor via Display Port, but it didn't help. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c19 --- Comment #19 from Takashi Iwai --- In anyway, please boot with drm.debug=0x1e boot option, and upload the dmesg output again. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c21 --- Comment #21 from Andreas --- Created attachment 854484 --> http://bugzilla.opensuse.org/attachment.cgi?id=854484&action=edit dmesg drm.debug=0x1e -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c25 --- Comment #25 from Andreas --- There is a post on reddit with almost the same laptop and LSPCON errors, HDMI also does not work. The second monitor does not work for me either with Intel, but I thought it was normal, but the person writes that the second monitor works under Windows. https://www.reddit.com/r/openSUSE/comments/s0uzw3/leap_153_cant_get_hdmi_or_... The same errors are on the ThinkPad P15 Gen 1 with Comet-Lake https://gitlab.freedesktop.org/drm/intel/-/issues/4458 -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c30 --- Comment #30 from Patrik Jakobsson --- Are you still experiencing this problem or is it solved in upstream? -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c32 Daike Yu changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |yu.daike@suse.com --- Comment #32 from Daike Yu --- I have also been bitten by this issue on a Thinkpad P15 Gen1, with a Quadro T1000M and an i9-10885H. This issue exists on both Leap 15.4 (kernel 5.14.21) and Tumbleweed (kernel 6.0.8). Tested nvidia driver includes G05 (on Leap 15.4 and Tumbleweed) and G06 (on Tumbleweed). The version of nvidia driver doesn't seem to affect the behavior. dmesg output is almost identical to the attachment provided by Andreas earlier (I'll attach them nonetheless). Some observations I would like to add: Since this issue seems to be related to the HDMI output (as indicated by the fact that this is related to LSPCON, and other bug reports also mentioned that HDMI is not working unless plugged in before boot up), I did some experience with whether HDMI is working in different situations: 1. nvidia driver disabled (with prime-select), and nvidia module unloaded: HDMI output does not work no matter HDMI cable plugged in before boot or not. 2. nvidia driver enabled (with prime-select): 2.1. HDMI cable plugged in before boot up: HDMI output works, LSPCON error message still appears. 2.2. HDMI cable unplugged before boot up: HDMI output not working during X11 session. However plymouth splash does show up normally on the external monitor after `systemctl reboot`. I'm guessing in this scenario the discrete GPU is providing the HDMI output. 3. nvidia driver enabled, iGPU disabled in BIOS: same as 2, except LSPCON error messages disappeared (makes sense since i915 is not loaded at all) Unfortunately AFAIK there's no way to disable discrete GPU only in BIOS on this machine, so I cannot test the iGPU-only situation. If anybody know how to disable discrete GPU on this machine please kindly let me know. Also, on the same machine HDMI is working without flaw if I dual-boot into Windows 10, although I'm not sure in such situation which GPU is in use (since I don't know how to enable only the iGPU). So I suspect there's two separate issue exists here: 1. nvidia driver is not handling HDMI hot plug correctly (therefore we need to plug in HDMI cable before boot to get HDMI working with discrete GPU) 2. i915 have some problem with LSPCON (so HDMI output with iGPU is not working at all). Hope this info helps, and please let me know if there's anything else I can help with (debug log/testing etc). -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c34 --- Comment #34 from Daike Yu --- Created attachment 863135 --> http://bugzilla.opensuse.org/attachment.cgi?id=863135&action=edit dmesg output, with HDMI plugged-in before boot -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 Daike Yu changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #863135|dmesg output, with HDMI |dmesg output, with iGPU description|plugged-in before boot |disabled -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1193599 http://bugzilla.opensuse.org/show_bug.cgi?id=1193599#c36 Takashi Iwai changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |sndirsch@suse.com --- Comment #36 from Takashi Iwai --- (In reply to Daike Yu from comment #32)

So I suspect there's two separate issue exists here:

1. nvidia driver is not handling HDMI hot plug correctly (therefore we need to plug in HDMI cable before boot to get HDMI working with discrete GPU) 2. i915 have some problem with LSPCON (so HDMI output with iGPU is not working at all).

Hope this info helps, and please let me know if there's anything else I can help with (debug log/testing etc).

Thanks for the detailed analysis! Adding Stefan to Cc, as this sounds more like a Intel/Nvidia combo case (being always a headache). -- You are receiving this mail because: You are on the CC list for the bug.