[Bug 216516] s2ram freezes screen (Ryzen-5650U incl. Radeon GPU)

[bugzilla-daemon@kernel.org]

https://bugzilla.kernel.org/show_bug.cgi?id=216516

--- Comment #26 from kolAflash (kolAflash@kolahilft.de) ---
Created attachment 301896
  --> https://bugzilla.kernel.org/attachment.cgi?id=301896&action=edit
kernel log for s2idle: v6.0-rc7 with bug 216516 comment #25 patches

@Mario
Thank you for all the great help!!!




(In reply to Mario Limonciello (AMD) from comment #25)
> [...]
> Can you please do the following:
> 
> 1) add right after the usleep_range() you added before this:
> 
> amd_pmc_idlemask_read(pdev, pdev->dev, NULL);
> 
> 2) Add
> https://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git/commit/
> ?h=bleeding-edge&id=cb3e7d624c3ff34a300587929c82af7364cf5c09
> 
> That will get us some more debugging information at least when this happens.

Compiled with all patches from comment #14, 1) and 2).

Here are the first results.

1. s2idle was without sysfs settings. System was woken by closing the lid.

2. s2idle was with sysfs settings (see log). Directly after sleep I (un)plugged
the power, opened/closed the lid and pressed keys on the keyboard. System was
finally fully woken by pressing the power button.

3. s2idle was with sysfs settings (see log). System was going to sleep while
the lid was closed. Lid was opened after about 18 seconds. System was fully
woken by pressing the power button.

I'll now do another test for at least 30 minutes. So we can see the power
consumption.




> [...]
> As a W/A can you have a try with acpi.ec_no_wakeup=1 on your kernel command
> line?  That should stop EC from issuing any wakeups and triggering this
> other firmware bug.

That workaround works very well! Thanks :-)

Didn't make any sysfs of procfs settings. Opening/closing the lid, pressing
keyboard keys and disconnecting AC power didn't wake the system. And the
"deepest state" log reported the full time being in that state, so it didn't
return from "deepest state" early.
I didn't have time to do a long time test with the ec_no_wakeup workaround.
Tested it for max. 90 seconds.
(workaround is not in the attached log)

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.