[PATCH] D72222: [Driver][CodeGen] Add -fpatchable-function-entry=N[,0]

Tue Feb 4 10:21:46 PST 2020

MaskRay added a comment.

In D72222#1856587 <https://reviews.llvm.org/D72222#1856587>, @hans wrote:

> In D72222#1849326 <https://reviews.llvm.org/D72222#1849326>, @hans wrote:
>
> > > I created D73680 <https://reviews.llvm.org/D73680> to place the patch label after BTI.
> > > 
> > > @hans Is there still time to cherry pick the patch to release/10.x? See above, Linux developers really want the Clang release to have compatible behavior with GCC.
> >
> > Yes, there is still time. Just let me know which commits to cherry-pick.
>
>
> Just to follow up: D73680 <https://reviews.llvm.org/D73680> was cherry-picked to 10.x. Does that mean all issues are resolved here, and the kernel folks are happy, or is there more work expected that might affect the release?

@hans Thanks for following up!

All known AArch64 -fpatchable-function-entry= issues are resolved after D73680 <https://reviews.llvm.org/D73680>.
D73760 <https://reviews.llvm.org/D73760> is an x86 counterpart of D73680 <https://reviews.llvm.org/D73680>. It changed the placement of the label `.Lpatch0`, which has been agreed to be superior.

hjl is working on the GCC side issue https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93492. I don't know whether that fix will be included in GCC 10.
>From the discussions, what we did in D73760 <https://reviews.llvm.org/D73760> is a consensus and GCC will eventually do it as well. D73760 <https://reviews.llvm.org/D73760> will be nice to be cherry picked if there is not much trouble, so we won't have a major release with unsatisfactory label placement if Linux x86 developers ever want to adopt -fpatchable-function-entry=.

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D72222/new/

https://reviews.llvm.org/D72222