Re: [PATCH v2] Subject: x86/PAT: Report PAT on CPUs that support PAT without MTRR

From: Chuck Zmudzinski
Date: Wed Aug 17 2022 - 12:39:36 EST


On 7/14/2022 10:07 PM, Chuck Zmudzinski wrote:
> On 7/14/2022 1:30 AM, Thorsten Leemhuis wrote:
>
> >
> > Sorry, have to ask: is this supposed to finally fix this regression?
> > https://lore.kernel.org/regressions/YnHK1Z3o99eMXsVK@mail-itl/
>
> Yes that's the first report I saw to lkml about this isssue.

Hi Thorsten,

Actually, now I realize that was not the first report to lkml about this
issue, although it *was* the first report to the regressions mailing list.

Actually, the first report to lkml about this issue was Jan's patch that
will hopefully soon make it to linux-next and mainline. So the proper
Link: tag for this issue in the actual patch to be committed to the mainline
kernel is to Jan's patch that was originally posted to lkml before any
user reported it to the regressions mailing list. To know there was a
regression from Jan's original patch, one would need to have read his
commit message since he did not actually report it as a regression to the
regressions mailing list at that time, nor did he identify a culprit commit
at that time.

Bottom line: everything seems OK right now because the patch moving
towards mainline does have the correct Link: tag. Thanks for all the
work you have done on this regression.

Best regards,

Chuck