[RFC][groovy][PATCH 0/1] UBUNTU: SAUCE: RDMA/core: Introduce peer memory interface

[dann frazier]

On Mon, Jun 28, 2021 at 3:18 AM Stefan Bader <stefan.bader at canonical.com> wrote:
>
> On 25.06.21 15:38, dann frazier wrote:
> > On Fri, Jun 25, 2021 at 2:01 AM Stefan Bader <stefan.bader at canonical.com> wrote:
> >>
> >> On 24.06.21 21:36, dann frazier wrote:
> >>> This is a backport of a SAUCE patch we're already carrying in hirsute[*].
> >>> In a conversation w/ Terry, he mentioned that he'd like to see some amount
> >>> of real world functional testing be completed before we'd consider SRU'ing
> >>> to other kernels. That testing will take some scheduling/engineering effort
> >>> so, in order to minimize the risk of respins/retests, I'm submitting this as
> >>> an RFC. My goal here is to try and fish out any aspects of this patch that
> >>> are likely to get NAK'd - even if testing were to pass.
> >>
> >> My primary argument against any larger scale change to Groovy is that it goes
> >> EOL in July. So why bothering?
> >
> > That's a good question Stefan. The ultimate goal is to have it
> > backported to focal/5.4, this is just to prevent focal-ga -> focal-hwe
> > regressions. Is it correct that the last cycle that will have both 5.4
> > and 5.8 would be 2021.07.12? In which case, we're unlikely to have a
> > 5.4 version ready in that time frame anyway, so I'd agree we should
> > just skip 5.8.
>
> The plan at the moment is to have 5.11 become the default hwe kernel for Focal
> by end of this cycle (if test results look good enough). Next cycle would be
> only done if things do not work out.

Understood, then I see no reason to proceed. Thanks for the comments!

  -dann