[PATCH] D69825: [Clang][Driver] Re-use the calling process instead of creating a new process for the cc1 invocation

Fri Feb 7 18:02:19 PST 2020

plotfi added a comment.

@thakis I got sidetracked by some other things and had not made more progress, if progress is not made yes I think -emit-interface-stubs should imply -fno-integrated-cc1 if that seems like the right thing to do to you. @erichkeane Thanks for filing the bugzilla bug. I was having some problems with my bugzilla login myself and wasn't able to do so. I'll track progress if this bug and decide shortly if my feature should imply no-integraed-cc1. Thanks Guys.

In D69825#1859504 <https://reviews.llvm.org/D69825#1859504>, @thakis wrote:

> In D69825#1837243 <https://reviews.llvm.org/D69825#1837243>, @hctim wrote:
>
> > For tracking purposes.
> >
> > The ASan leaks have been worked around in two separate patches in order to make the bots green again:
> >
> > - c38e42527b21acee8d01a016d5bfa2fb83202e29 <https://reviews.llvm.org/rGc38e42527b21acee8d01a016d5bfa2fb83202e29>
> > - e174da447c180b586719cb28f7bd556e30625762 <https://reviews.llvm.org/rGe174da447c180b586719cb28f7bd556e30625762>
> >
> >   @plotfi is graciously debugging :)
>
>
> How is the debugging going? If it takes a while to fix, maybe it's better to make  -emit-interface-stub imply -fno-integrated-cc1 until it's fixed for real, instead of adding that in the tests? As-is, the tests are happy but the feature is likely broken (?)

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D69825/new/

https://reviews.llvm.org/D69825