Re: [GIT PULL] x86 updates for v4.15

[Ingo Molnar]

* Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:

> On Tue, Nov 14, 2017 at 1:48 AM, Borislav Petkov <bp@xxxxxxxxx> wrote:
> >
> > Just did 2 suspend cycles (once to RAM and once to disk) on my x230
> > with your tree from right now and it looks ok so far. So it could be
> > machine- and config-specific...
> 
> .. and it's not repeatable for me. I rebooted pretty quickly, and
> didn't gather a lot of information (well, 'dmesg' would SIBGUS, so..)
> and it hasn't happened again.
> 
> Will ignore until I have more information.

Haven't seen such behavior or got such reports - although admittedly laptop 
suspend/resume testing is done only sporadically, as it isn't easily automated.

As per the symptoms one thing that _could_ produce SIGSEGVs are the 
CONFIG_X86_INTEL_UMIP changes: the upcoming changes that make any UMIP action more 
verbose should make it more apparent if that's the case.

Plus, of course, anything entry code related. We did a few harmless (looking...) 
x86/mm changes as well, but none stands out at the moment.

Thanks,

	Ingo