Re: runtime regression with "x86/mm/pat: Emulate PAT when it is disabled"

[Toshi Kani]

On Mon, 2016-03-07 at 17:56 -0700, Toshi Kani wrote:
> On Mon, 2016-03-07 at 18:53 -0500, Paul Gortmaker wrote:
> > [Re: runtime regression with "x86/mm/pat: Emulate PAT when it is
> > disabled"] On 07/03/2016 (Mon 16:38) Toshi Kani wrote:
> > 
> > > On Mon, 2016-03-07 at 16:08 -0500, Paul Gortmaker wrote:
> > > > [dropping oe list and lkml since attaching dmesg files.]
> > > > 
> > 
> > [...]
> > 
> > > > > Yes, please send me full dmesg files. ÂSince I do not know your
> > > > > original state, the diff does not give me the whole picture.
> > > > 
> > > > Attached.
> > > 
> > > Thanks for the dmesg files! ÂAs I suspected, there is no message from
> > > pat_init() in both cases. ÂThat is, you are missing the following
> > > message,
> > > which shows how PAT is configured to support cache attributes.
> > > 
> > > # dmesg | grep PAT
> > > [0.000000] x86/PAT: Configuration [0-7]: WBÂÂWCÂÂUC- UCÂÂWBÂÂWCÂÂUC-
> > > WTÂÂ
> > 
> > Interesting...
> > 
> > > 
> > > It may have seemed working before, but you did not have WC configured
> > > to PAT without calling pat_init(). ÂThere was not a proper check in
> > > place to detect this error before. ÂCan you please check your code to
> > > see what caused this skip of pat_init()? ÂIf you have a git tree, I
> > > can
> > > take a look as well.Â
> > 
> > You already have git copies of what I'm running, since it is vanilla
> > mainline commits.ÂÂNo code changes at this end whatsoever.ÂÂI did the
> > bisect on vanilla mainline.ÂÂAll I took from yocto was their ".config"
> > 
> > To recap, v4.1-rc5-21-g9dac62909451 works,ÂÂv4.1-rc5-22-g9cd25aac1f44
> > fails, and v4.5-rc6 also fails.ÂÂIf pat_init() isn't called then this
> > is a bug in current mainline.ÂÂI'll have a look later myself and see
> > if I can trace out how we expect to get to pat_init() and how that
> > might be skipped inadvertently unless someone beats me to it.
> 
> Oh, I see. ÂCan you send me the ".config" file?

And also an output of /proc/cpuinfo, please?

I think I know what's going on. ÂI noticed that you have the following
message in your dmesg files.

Â[ÂÂÂÂ0.000000] MTRR: Disabled

MTRR is set to disabled when your CPU is Intel but does not support MTRR.
ÂPerhaps, QEMU does not emulate MTRR?

pat_init() is not called when MTRR is disabled. ÂI think this dependency is
wrong, and it needs to be fixed.

This issue has been there for a long time, and you have been running
essentially as PAT disabled in the past. ÂThe commit in question simply
detected this issue.

Thanks,
-Toshi