Bug 218849 - ASRock TRX90 WS EVO fails to suspend to S3
Summary: ASRock TRX90 WS EVO fails to suspend to S3
Status: REOPENED
Alias: None
Product: ACPI
Classification: Unclassified
Component: Power-Sleep-Wake (show other bugs)
Hardware: All Linux
: P3 normal
Assignee: acpi_power-sleep-wake
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2024-05-16 13:43 UTC by Jacob Kaulike
Modified: 2024-06-14 16:11 UTC (History)
5 users (show)

See Also:
Kernel Version:
Subsystem:
Regression: No
Bisected commit-id:


Attachments
acpidump for ASRock WRX90 WS EVO (254.97 KB, application/x-xz)
2024-05-16 15:54 UTC, Jacob Kaulike
Details

Description Jacob Kaulike 2024-05-16 13:43:01 UTC
Testing with Ubuntu 22.04.3 with 6.5.0 kernel, and Ubuntu 24.04 with 6.8.0 kernel, and also Pop!_OS 22.04 with the pop 6.8.0 kernel, the ASRock TRX90 WS EVO motherboard does not suspend. Instead when I attempt to suspend the display blanks, fans spin up, the board remains powered on, and the system becomes unresponsive.
Comment 1 Mario Limonciello (AMD) 2024-05-16 14:08:24 UTC
As I understand it, reverting

d410ee5109a1 ("ACPICA: avoid "Info: mapping multiple BARs. Your kernel is fine."")

Which is the "fix" for https://bugzilla.kernel.org/show_bug.cgi?id=14445 fixes this issue, right?
Comment 2 Jacob Kaulike 2024-05-16 14:21:52 UTC
Yes, that is correct. Reverting that commit does fix this issue for us.
Comment 3 Mario Limonciello (AMD) 2024-05-16 14:36:54 UTC
I CC'ed the original reporter, Martin to the bug.

Martin - can you revert that on a current kernel and see if your original problem still exists?
Comment 4 Mario Limonciello (AMD) 2024-05-16 14:38:39 UTC
Jacob,

Can you please add an acpidump and kernel log (from a faulty kernel without that revert)
Comment 5 Martin Mokrejs 2024-05-16 14:51:05 UTC
Mario, https://lkml.org/lkml/2008/12/30/128 that was year 2008 and my ASUS L3C/S laptop. I don't have it anymore.
Comment 6 Mario Limonciello (AMD) 2024-05-16 14:54:10 UTC
Ah, got it.  Thanks for checking.  We *might* be best off reverting, but I'd like to see what Rafael and others have to say.
Comment 7 Jacob Kaulike 2024-05-16 15:54:46 UTC
Created attachment 306301 [details]
acpidump for ASRock WRX90 WS EVO

Here's an acpidump for the WRX90 WS EVO motherboard from Pop!_OS 22.04. I had to compress it since it's ~6MB
Comment 8 Mario Limonciello (AMD) 2024-05-22 09:52:16 UTC
This fix was posted, can you please try it? 

https://github.com/acpica/acpica/pull/954
Comment 9 Jacob Kaulike 2024-05-23 17:06:58 UTC
I've just given that a try and it does fix this issue
Comment 10 Mario Limonciello (AMD) 2024-05-26 20:48:30 UTC
I wouldn't say this should be marked resolved until that patch is accepted. It just confirms root cause.
Comment 11 Mario Limonciello (AMD) 2024-06-12 16:19:27 UTC
As an update here; the fix was accepted into ACPICA.  It will be pulled in the next ACPICA import or if an early patch is pulled into Linux kernel.
Comment 12 Mario Limonciello (AMD) 2024-06-14 16:11:58 UTC
Here is the submission backporting from ACPICA directly to Linux:

https://lore.kernel.org/linux-acpi/20240614140149.1654251-1-Raju.Rangoju@amd.com/T/#u

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