Bugzilla – Bug 1215285
[RPi3] dwc2 3f980000.usb: dwc2_hc_halt() Channel can't be halted
Last modified: 2023-12-05 10:00:51 UTC
## Observation openQA test in scenario opensuse-Tumbleweed-JeOS-for-RPi-aarch64-jeos@RPi3 fails in [journal_check](https://openqa.opensuse.org/tests/3568327/modules/journal_check/steps/9) New kernel 6.5.2 shows the following trace in journal on RPi3: dwc2 3f980000.usb: dwc2_hc_halt() Channel can't be halted ## Test suite description Maintainer: fvogt, mnowak Start JeOS from the HDD image, configure it using the firstboot wizard and then run basic tests. console=tty0 added as needed for aarch64. ## Reproducible Fails since (at least) Build [20221024](https://openqa.opensuse.org/tests/2833524) ## Expected result Last good: (unknown) (or more recent) ## Further details Always latest result in this scenario: [latest](https://openqa.opensuse.org/tests/latest?arch=aarch64&distri=opensuse&flavor=JeOS-for-RPi&machine=RPi3&test=jeos&version=Tumbleweed)
Despite being reported as an error, the result of that happening should be harmless since it's intended that periodic usb transaction, under some circumstances (served by DMA, etc.) cannot be halted. Dwc2 core should halt the relevant channel on end of frame, so the recover should be guarantee. Even if the net result may not be that much of a problem though, the point worth investigating is why the transaction has been requested to be halted in the first place. This may stem from some other issue...
Hi Guillame, I saw that the latest tests linked in your original ticket (tests executed 3 days ago) has passed. Is this ticket still relevant or maybe the issue is just non deterministic? Many thanks
(In reply to Andrea della Porta from comment #2) > Hi Guillame, > I saw that the latest tests linked in your original ticket (tests executed 3 > days ago) has passed. Is this ticket still relevant or maybe the issue is > just non deterministic? > Many thanks The problem is sporadic. Sometimes it fails, sometimes it passes... You can see that on the test history here (increase number of lines displayed): https://openqa.opensuse.org/tests/latest?arch=aarch64&distri=opensuse&flavor=JeOS-for-RPi&machine=RPi3&test=jeos&version=Tumbleweed#next_previous
This is most likely a duplicate of an issue a customer reported on RPi in that bug report *** This bug has been marked as a duplicate of bug 1214286 ***
I am not allowed to access bug#1214286 Could you fix that please?
(In reply to Guillaume GARDET from comment #5) > I am not allowed to access bug#1214286 > Could you fix that please? I am afraid it would not be possible, but the outcome is this: https://patchwork.kernel.org/project/linux-usb/patch/20231115144514.15248-1-oneukum@suse.com/
Just as a head up, please note that there is a good chance that the proposed fix in comment #6 will also fix the issue you are seeing, but the symptoms here are a little different so it may also be due to other cause. Its definitely worth trying the patch, though. Thanks
... and since the patch has just being pushed upstream, we may end up resolving (hopefully) this issue for free
The patch from comment 6 is a part of 6.6.4 stable and now merged to our stable branch so it should appear in TW soon. Guillaume, please let us know the outcome. It might be the fix even though the symptoms are different. As mentioned previously.