Bug 217282

Summary: Regression: ath11k hang on boot w/ Qualcomm FastConnect 6900 on HP Pro x360 435 G9
Product: Drivers Reporter: Carsten Hatger (xmb8dsv4)
Component: network-wirelessAssignee: drivers_network-wireless (drivers_network-wireless)
Status: RESOLVED UNREPRODUCIBLE    
Severity: blocking CC: regressions
Priority: P1    
Hardware: AMD   
OS: Linux   
Kernel Version: 6.1.22 Subsystem:
Regression: Yes Bisected commit-id:
Attachments: dmesg of 6.1.22 vanilla boot

Description Carsten Hatger 2023-03-31 14:49:06 UTC
Created attachment 304068 [details]
dmesg of 6.1.22 vanilla boot

Dear all,

ath11k hangs when booting v6.1.22 on a HP Pro x360 G9 w/ Qualcomm FastConnect 6900 using firmware WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.23.

However, v.6.1.21 works fine.

Please let me know if I can further assist in resolving this issue - testing would be fine.

Yours,
Carsten
Comment 1 The Linux kernel's regression tracker (Thorsten Leemhuis) 2023-04-02 08:07:39 UTC
I forwarded your report to the developers:
https://lore.kernel.org/regressions/83966474-658d-7e2f-3e7f-eb66100660e9@leemhuis.info/T/#u

I'm not one of them, nevertheless a few things:

* a "git bisect" would be really helpful here, as there were no ath11k changes between 6.1.21 and 6.1.22, hence the real problem is likely somewhere else; but before starting going down that path maybe wait a day or two, maybe one of the developers might have an idea what's wrong
* in some cases it can be helpful to have the dmesg of a working kernel at hand, consider attaching it
* would be good to know if mainline (e.g. 6.3-rc) is affected as well
Comment 2 Carsten Hatger 2023-04-03 08:03:41 UTC
Dear all.

I've rebuild kernel v6.1.22 to check if something went wrong during the build. Usually I build kernels in memory on a sufficiently large sized ramdisk to speed things up a bit.

The outcome is that I've not been able to reproduce the issue while booting into the rebuild. Silly me removed the old one prior to installing the rebuild - so no chance to check for (binary) diffs in kernel image and object files.

The only explanation I have for this result is that there could be something wrong with the RAM of my device since there have been no changes at all to the system:
- no updates at all
- no boot into W11 
- no full battery drain leaving hw in some strange state 
 
However, the system is stable: no reboots required so far. Strange thing :-|

Thanks for feedback so far. I'll close the issue marking it as unreproducible.

Yours,
Carsten