On 02.01.24 14:50, Michael Schaller wrote:
> On 01.01.24 23:15, Bjorn Helgaas wrote:
>> On Mon, Jan 01, 2024 at 07:57:40PM +0100, Michael Schaller wrote:
>>> On 01.01.24 19:13, Bjorn Helgaas wrote:
>>>> On Mon, Dec 25, 2023 at 07:29:02PM +0100, Michael Schaller wrote:
>>>> ...
>>
>>>> So unless somebody has a counter-argument, I plan to queue a revert of
>>>> 08d0cc5f3426 ("PCI/ASPM: Remove pcie_aspm_pm_state_change()") for
>>>> v6.7.
>>>
>>> If it helps I could also try if a partial revert of 08d0cc5f3426
>>> would be
>>> sufficient. This might also narrow down the issue and give more insight
>>> where the issue originates from.
>>
>> We're so close to the v6.7 final release that I doubt we can figure
>> out the problem and test a fix before v6.7. I'm sure Kai-Heng would
>> appreciate any additional data, but I don't think it's urgent at this
>> point.
>
> We're indeed close to the final v6.7 release, which in turn means that a
> last minute revert of a 16 month old commit might cause even more
> regressions as there have been quite a few ASPM changes afterwards and
> there won't be much testing anymore before the final release.
>
> Furthermore, given the age of the commit and that it has been backported
> to kernel 5.15, the question is also if the revert would be backported
> to the affected LTS kernels?
>
> If this regression risk is acceptable then I'm all for reverting the
> commit now and then working on a fix.
FWIW (just in case some of you might not be aware of this): Linus not
that long ago said this about regressions that are somewhat older:
"""
There's obviously a time limit: if that "regression in an earlier
release" was a year or more ago, and just took forever for people to
notice, and it had semantic changes that now mean that fixing the
regression could cause a _new_ regression, then that can cause me to
go "Oh, now the new semantics are what we have to live with".
"""
For full context see:
https://lore.kernel.org/all/CAHk-=wis_qQy4oDNynNKi5b7Qhosmxtoj1jxo5wmB6SRUwQUBQ@xxxxxxxxxxxxxx/
Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
--
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
If I did something stupid, please tell me, as explained on that page.
