Re: The Annnoying Affair has No eXcuses - Bulletin #2

On Mon, 4 Nov 2019, speck for Ben Hutchings wrote:
> On Mon, 2019-11-04 at 13:09 +0100, speck for Thomas Gleixner wrote:
> > Folks,
> > 
> > I've merged Linus tree into master to pick up the KVM fix and applied
> > Paolo's V9 series. The commits are considered stable from now on. Any
> > further tweaks need to go on top.
> > 
> > The stable branches linux-5.3.y, linux-4.19.y and linix-4.14.y have been
> > updated to the latest stable and the fallout in the backports of TAA has
> > been fixed.
> > 
> > The non KVM parts of the NX series has been backported to them. Paolo will
> > provide the KVM backports including the commits which need to be
> > additionally backported as prerequisites for 5.3, 4.19 and 4.14.
> 
> Based on a quick comparison between your 4.19 branch and mine:
> 
> * You've included "x86/cpu: Add Atom Tremont (Jacobsville)" and
> "x86/cpu: Add Tremont to the cpu vulnerability whitelist", but Paolo
> seemed to say these weren't worth adding so I've dropped them.

Hmm. Not sure about that, but it doesn't really matter.

> * Your backport of "KVM: x86: use Intel speculation bugs and features
> as derived in generic x86 code" is messed up; at least the indentation
> is wrong.

Yes indeed.
 
> * I backported commit 3ff519f29d98 "KVM: x86: adjust kvm_mmu_page
> member to save 8 bytes" as this made it easier to apply the other
> changes in arch/x86/include/asm/kvm_host.h.

For which part TAA or NX?

Thanks,

	tglx