Re: [PATCH v2] usb: core: reduce power-on-good delay time of root hub

[Thorsten Leemhuis <linux@leemhuis.info>]

On 01.11.21 17:46, Thorsten Leemhuis wrote:
> On 01.11.21 04:33, Walt Jr. Brake wrote:
>> Three month ago, My USB 3.1 device cannot be detected after upgrade
>> kernel from version 5.10.41 to 5.10.42.
>>
>> And I found some other people ran into the same issue:
>>
>> https://forum.manjaro.org/t/kernel-5-10-59-doesnt-detect-a-connected-usb-external-disk/79320
>>
>> https://forum.manjaro.org/t/how-to-make-active-usb-hub-work-in-manjaro/84548
>>
>>
>> Two days ago, I try to fix the issue myself, and I found out that it was
>> cause by this patch. After revert this patch, my USB 3.1 device can be
>> detect again.
>>
>> I have already file a bug on here:
>> https://bugzilla.kernel.org/show_bug.cgi?id=214875.
> 
> #regzbot ^introduced 90d28fb53d4a51299ff324dede015d5cb11b88a2
> #regzbot link: https://bugzilla.kernel.org/show_bug.cgi?id=214875

#regzbot fixed-by e1959faf085b004e6c3afaaaa743381f00e7c015
#regzbot ignore-activity

BTW, to anyone is reading this and wondering why it's needed: regzbot
needed the above, as the fix for this regression(¹) lacked a Link: tag
with the URL to the report. It it had been there, regzbot would have
automatically noticed the fix when it showed up in next.

Thinking aloud: The fix was posted in the thread with the regression
report. Regzbot could be enhanced to detect such cases automatically.

Thinking aloud, too: regzbot could be enhanced to notice if a mail or a
commit mentioned a commit-id that is known to cause a tracked regression.

I guess the former might not be worth it, the latter OTOH might be a
good idea I will keep in mind.

Ciao, Thorsten


(¹)
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?id=e1959faf085b004e6c3afaaaa743381f00e7c015