Arch Linux

The HDMI / Displayport ist not shown on System Preferences - Audio - Output either with kernel 6.1.1.arch1-1. With 6.0.12 it's shown.

There is also an strange behaviour with the external display conntected to the laptop. Unplug an plug gin again, nothing is displayed. Also after upgrade to 6.1 and gone after downgrade

> https://bugzilla.kernel.org/show_bug.cgi?id=216836

For some reason it wants a login...cannot view.

The standard advice for kernel regressions [1]. Forum thread [2].

[1] https://wiki.archlinux.org/title/Kernel#Debugging_regressions
[2] https://bbs.archlinux.org/viewtopic.php?id=282239

Hi, I got as well the issue about the external display connected to the laptop after upgrading to 6.1.1-arch1-1 (only one of them was visible). Got this kernel stack trace:

Dec 25 11:33:47 matebook kernel: ------------[ cut here ]------------
Dec 25 11:33:47 matebook kernel: WARNING: CPU: 13 PID: 156 at drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc_link.c:3533 update_mst_stream_alloc_table+0x150/0x160 [amdgpu]
Dec 25 11:33:47 matebook kernel: Modules linked in: snd_rn_pci_acp3x vfat snd_hwdep irqbypass snd_acp_config snd_pcm pcc_cpufreq(-) fat hid_multitouch(+) cfg80211(+) crct10dif_pclmul fjes(-) crc32_pclmul snd_so>
Dec 25 11:33:47 matebook kernel: CPU: 13 PID: 156 Comm: kworker/13:1 Not tainted 6.1.1-arch1-1 #1 9bd09188b430be630e611f984454e4f3c489be77
Dec 25 11:33:47 matebook kernel: Hardware name: HUAWEI CREM-WXX9/CREM-WXX9-PCB-B1, BIOS 2.23 01/19/2022
Dec 25 11:33:47 matebook kernel: Workqueue: events_long drm_dp_mst_link_probe_work [drm_display_helper]
Dec 25 11:33:47 matebook kernel: RIP: 0010:update_mst_stream_alloc_table+0x150/0x160 [amdgpu]
Dec 25 11:33:47 matebook kernel: Code: 00 00 75 2d 48 81 c4 98 00 00 00 5b 5d 41 5c c3 cc cc cc cc 41 0f b7 40 04 4d 89 19 49 89 59 08 66 41 89 41 10 e9 71 ff ff ff <0f> 0b e9 fe fe ff ff e8 24 b6 00 f6 0f 1f 4>
Dec 25 11:33:47 matebook kernel: RSP: 0018:ffffa5564071f680 EFLAGS: 00010202
Dec 25 11:33:47 matebook kernel: RAX: 0000000000000002 RBX: 0000000000000000 RCX: 0000000000000000
Dec 25 11:33:47 matebook kernel: RDX: 0000000000000000 RSI: ffffa5564071f680 RDI: ffffa5564071f710
Dec 25 11:33:47 matebook kernel: RBP: ffff92bfc4400aa0 R08: ffffa5564071f740 R09: ffffa5564071f488
Dec 25 11:33:47 matebook kernel: R10: ffff92bfcbf66000 R11: ffff92bfc9d14300 R12: 0000000000000002
Dec 25 11:33:47 matebook kernel: R13: ffff92bfc61fe800 R14: ffffffffc0b7d4c0 R15: 0000000000000000
Dec 25 11:33:47 matebook kernel: FS: 0000000000000000(0000) GS:ffff92c2ce140000(0000) knlGS:0000000000000000
Dec 25 11:33:47 matebook kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Dec 25 11:33:47 matebook kernel: CR2: 00007f4cd8bd4000 CR3: 0000000104270000 CR4: 0000000000750ee0
Dec 25 11:33:47 matebook kernel: PKRU: 55555554
Dec 25 11:33:47 matebook kernel: Call Trace:
Dec 25 11:33:47 matebook kernel: <TASK>
Dec 25 11:33:47 matebook kernel: dc_link_allocate_mst_payload+0x99/0x2a0 [amdgpu 895e2b3772442c7d04dbf61a65c8a3690bb074b6]
Dec 25 11:33:47 matebook kernel: core_link_enable_stream+0x7d0/0x980 [amdgpu 895e2b3772442c7d04dbf61a65c8a3690bb074b6]
Dec 25 11:33:47 matebook kernel: ? optc1_set_drr+0x13a/0x1e0 [amdgpu 895e2b3772442c7d04dbf61a65c8a3690bb074b6]
Dec 25 11:33:47 matebook kernel: dce110_apply_ctx_to_hw+0x67b/0x720 [amdgpu 895e2b3772442c7d04dbf61a65c8a3690bb074b6]
Dec 25 11:33:47 matebook kernel: ? __free_pages_ok+0x2a7/0x530
Dec 25 11:33:47 matebook kernel: ? dcn10_wait_for_mpcc_disconnect+0x3d/0x150 [amdgpu 895e2b3772442c7d04dbf61a65c8a3690bb074b6]
Dec 25 11:33:47 matebook kernel: dc_commit_state_no_check+0x38c/0xc70 [amdgpu 895e2b3772442c7d04dbf61a65c8a3690bb074b6]
Dec 25 11:33:47 matebook kernel: dc_commit_state+0x96/0x110 [amdgpu 895e2b3772442c7d04dbf61a65c8a3690bb074b6]
Dec 25 11:33:47 matebook kernel: amdgpu_dm_atomic_commit_tail+0x4a4/0x2ae0 [amdgpu 895e2b3772442c7d04dbf61a65c8a3690bb074b6]
Dec 25 11:33:47 matebook kernel: ? dcn21_fast_validate_bw+0x3c2/0x490 [amdgpu 895e2b3772442c7d04dbf61a65c8a3690bb074b6]
Dec 25 11:33:47 matebook kernel: ? dcn21_validate_bandwidth_fp+0x115/0x6e0 [amdgpu 895e2b3772442c7d04dbf61a65c8a3690bb074b6]
Dec 25 11:33:47 matebook kernel: ? resource_build_scaling_params+0x908/0xeb0 [amdgpu 895e2b3772442c7d04dbf61a65c8a3690bb074b6]

After googling "update_mst_stream_alloc_table" I ended up in https://gitlab.freedesktop.org/drm/amd/-/issues/2171.
I rolled it back to 6.0.12-arch1-1 and both displays are now visible. No "update_mst_stream_alloc_table" present in the log.

Quick update as most discussions happened in the forum:

The responsible commit has been identified as ef6f5494faf6a37c74990689a3bb3cee76d2544c and a user has shared patched kernel packages:
https://bbs.archlinux.org/viewtopic.php?pid=2075553#p2075553

mixin, the issue you mentioned is also a quite severe regression in Linux 6.1 but as far as I can tell not related to the audio issue described here.

Thanks Michael, I've added my comment to https://bugs.archlinux.org/task/76934

It seems as if this issue has been thoroughly debugged and even bisected; which is good, but has anyone been able to verify that this triage is reported upstream? I see https://bugzilla.kernel.org/show_bug.cgi?id=216836 but I have no way to view this to either alert the maintainers myself, nor to ensure that the maintainers have been alerted. Can anyone confirm if this information has reached the places it needs to?

I tested this bug on an Intel (12th gen) machine too. Worked. It seems to be only an AMD Issue.

no reaction on: https://bugzilla.kernel.org/show_bug.cgi?id=216836

Not to get too off-topic in this bug report, but that's not what I mean. I cannot see the bug, even logged in to bugzilla. My question is, specifically: does this bug report convey the fact that a) this issue was bisected and b) reverting the commit from the bisection causes the regression to go away. If these two specific things are not in that bug report it is vital information that is missing that could delay a fix for this getting into 6.1.2 or 6.1.3.

@LaserEyess, i added you to the cc list of this bug. i hope this is ok for you. Don't know why this tickt is only visible to reporter or cc list. Perhaps you can find out more.

Arthur, you misreported that bug, saying it is a kernel.org issue, it needs to be under Drivers/Sound(ALSA) otherwise the maintainers won't ever see it.

@LaserEyess,thanks for your help. I changed the issue type

[Ignore this, browser refreshed another tab and submitted a post again]

There is now a fix that has been submitted upstream and will hopefully soon be applied to the stable series as well: https://lore.kernel.org/all/20221228125714.16329-1-tiwai%40suse.de/

(Flyspray seems to be confused by that URL. It should read: "https://lore.kernel.org/all/20221228125714.16329-1-tiwai @ suse.de/" without the spaces around the @.)

> Flyspray seems to be confused by that URL

It's documented in the wiki, see the Tip here [1]. Have fixed this one for you.

[1] https://wiki.archlinux.org/title/Bug_reporting_guidelines#Summary

The fix has now been merged into mainline: https://github.com/torvalds/linux/commit/090ddad4c7a9fefd647c762093a555870a19c8b2