NACK/Cmnt: [PULL][Noble] backport arm64 THP improvements from 6.9

[Andrea Righi]

On Wed, Apr 03, 2024 at 07:07:25AM -0600, dann frazier wrote:
> On Wed, Apr 03, 2024 at 09:47:22AM +0200, Andrea Righi wrote:
> > On Thu, Mar 28, 2024 at 06:25:45AM -0600, dann frazier wrote:
> > > BugLink: https://bugs.launchpad.net/bugs/2059316
> > > 
> > > These are all clean cherry-picks from upstream, save one that required
> > > minor backporting due to an API change. This does touch generic code.
> > > I've regression tested as much as I've had time to given I just
> > > learned about this patchset last week. But with the performance
> > > improvements I've measured, and the GPU workload improvements reported
> > > upstream, I think this is something we need to seriously consider
> > > for our next LTS.
> > > 
> > > I plan to monitor upstream for Fixes commits during the 6.9
> > > development cycle, in case regressions are found/fixed.
> > 
> > Hi Dann,
> > 
> > after thinking and discussing more about this with the team, we could
> > try to mark all the patches as SAUCE and apply them to the noble/linux
> > kernel (this would allow us to better identify them).
> > 
> > Then if we find critical issues in the future that may delay security
> > updates we would be able to drop them quickly, if needed (it shouldn't
> > happen, but just to be extra safe...).
> > 
> > So, I'm NACK-ing this for now, but if you can send another PR simply by
> > updating the subject of the patches to add the keyword "SAUCE:" I'll be
> > happy to apply it.
> 
> Andrea - that sounds like a reasonable plan. While we wait for an
> official build, I'll try to get some stress-ng runs in on our ARM
> server kit to see if we can turn up any issues ourselves.

Perfect! Feel free to share any result/number/issue that you may find.

Thanks,
-Andrea