Re: [PATCH 0/3] x86: make pat and mtrr independent from each other

[Chuck Zmudzinski <brchuckz@netscape.net>]

On 7/16/2022 7:32 AM, Chuck Zmudzinski wrote:
> On 7/15/2022 10:25 AM, Juergen Gross wrote:
> > Today PAT can't be used without MTRR being available, unless MTRR is at
> > least configured via CONFIG_MTRR and the system is running as Xen PV
> > guest. In this case PAT is automatically available via the hypervisor,
> > but the PAT MSR can't be modified by the kernel and MTRR is disabled.
> >
> > As an additional complexity the availability of PAT can't be queried
> > via pat_enabled() in the Xen PV case, as the lack of MTRR will set PAT
> > to be disabled. This leads to some drivers believing that not all cache
> > modes are available, resulting in failures or degraded functionality.
> >
> > The same applies to a kernel built with no MTRR support: it won't
> > allow to use the PAT MSR, even if there is no technical reason for
> > that, other than setting up PAT on all cpus the same way (which is a
> > requirement of the processor's cache management) is relying on some
> > MTRR specific code.
> >
> > Fix all of that by:
> >
> > - moving the function needed by PAT from MTRR specific code one level
> >   up
> > - adding a PAT indirection layer supporting the 3 cases "no or disabled
> >   PAT", "PAT under kernel control", and "PAT under Xen control"
> > - removing the dependency of PAT on MTRR
> >
> > Juergen Gross (3):
> >   x86: move some code out of arch/x86/kernel/cpu/mtrr
> >   x86: add wrapper functions for mtrr functions handling also pat
> >   x86: decouple pat and mtrr handling
> >
> >  arch/x86/include/asm/memtype.h     |  13 ++-
> >  arch/x86/include/asm/mtrr.h        |  27 ++++--
> >  arch/x86/include/asm/processor.h   |  10 +++
> >  arch/x86/kernel/cpu/common.c       | 123 +++++++++++++++++++++++++++-
> >  arch/x86/kernel/cpu/mtrr/generic.c |  90 ++------------------
> >  arch/x86/kernel/cpu/mtrr/mtrr.c    |  58 ++++---------
> >  arch/x86/kernel/cpu/mtrr/mtrr.h    |   1 -
> >  arch/x86/kernel/setup.c            |  12 +--
> >  arch/x86/kernel/smpboot.c          |   8 +-
> >  arch/x86/mm/pat/memtype.c          | 127 +++++++++++++++++++++--------
> >  arch/x86/power/cpu.c               |   2 +-
> >  arch/x86/xen/enlighten_pv.c        |   4 +
> >  12 files changed, 289 insertions(+), 186 deletions(-)
> >
>
> This patch series seems related to the regression reported
> here on May 5, 2022:

I'm sorry, the date of that report was May 4, 2022, not
May 5, 2022 - just to avoid any doubt about which regression
I am referring to.

Chuck

>
> https://lore.kernel.org/regressions/YnHK1Z3o99eMXsVK@mail-itl/
>
> I am experiencing that regression 

or a very similar regression that is caused by the same commit:

bdd8b6c98239cad
("drm/i915: replace X86_FEATURE_PAT with pat_enabled()")

> and could test this patch
> on my system.
>
> Can you confirm that with this patch series you are trying
> to fix that regression?
>
> Chuck

Chuck