Re: [PATCH -v2] MAINTAINERS: Fix Intel TDX kernel file patterns

[Sean Christopherson]

On Mon, Nov 02, 2026, Rick P Edgecombe wrote:
> On Fri, 2026-10-30 at 20:01 -0700, Borislav Petkov wrote:
> > From: "Borislav Petkov (AMD)" <bp@xxxxxxxxx>
> > 
> > A K: is a content regex file pattern which matches any text in a patch
> > so if a commit message contains "tdx" as in
> > 
> >   has_tdxe_auto_clear
> > 
> > for example, see
> > 
> >   https://lore.kernel.org/r/20261026-g2ul_sci-v3-3-b3097163edd5@xxxxxxxxxxxxxx
> > 
> > get_maintainer.pl erroneously adds the x86 gang to the Cc list for what
> > obviously is a completely unrelated patch.
> > 
> > The same happens when someone uses it to check
> > 
> >   drivers/power/reset/tdx-ec-poweroff.c
> > 
> > patches.
> > 
> > So switch the x86 TDX patterns to files and directories (F:) wildcard
> > type patterns and group them by path, for less false positives.
> > 
> > Signed-off-by: Borislav Petkov (AMD) <bp@xxxxxxxxx>
> > Link:
> > https://lore.kernel.org/r/20261027055750.GBauA9XlPykc1c1Xhv@fat_crate.local
> > ---
> >  MAINTAINERS | 7 +++++--
> >  1 file changed, 5 insertions(+), 2 deletions(-)
> > 
> > diff --git a/MAINTAINERS b/MAINTAINERS
> > index 7a162e9bcab1..015cf66afb96 100644
> > --- a/MAINTAINERS
> > +++ b/MAINTAINERS
> > @@ -30347,8 +30347,11 @@ L:	linux-coco@xxxxxxxxxxxxxxx
> >  L:	kvm@xxxxxxxxxxxxxxx
> >  S:	Supported
> >  T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86/tdx
> > -N:	tdx
> > -K:	\b(tdx)
> > +F:	arch/x86/**/*tdx*
> > +F:	Documentation/**/*tdx*
> > +F:	drivers/virt/coco/tdx*/
> > +F:	include/uapi/linux/tdx
> > +F:	tools/testing/selftests/tdx/
> 
> This drops arch/x86/kvm/vmx/tdx.c. Not that we are sending any PRs based on this
> maintainer entry, but I'd like to continue to keep tabs on that code.
> 
> Sean do you have any thoughts about how this entry applies to the KVM side of
> TDX? I'm not sure we actually ever had a specific discussion.

Honestly, it annoys me.  I often temporarily delete the TDX MAINTAINERS entry when
sending patches that affect arch/x86/kvm/vmx/tdx.c, because many, if not most, of
the KVM changes are purely mechanical and/or KVM-internal that don't have much to
do with TDX as a whole.  Although maybe that's just a reflection of the types of
patches I tend to send.

But I do that purely to try and not pollute x86@ and linux-coco@ with stuff I'm
99.9% certain they don't care about.  If y'all want to sign up for spam, I won't
stop you. :-D

> In practice it has been a cross subsystem wrangler role. With TDX connect it
> will likely grow more tentacles.