victorkingi added inline comments.

================
Comment at: flang/test/Driver/frontend-forwarding.f90:26
 
+! RUN: %flang -### %s 2>&1 \
+! RUN:     -foptimization-record-file=%t.opt.yaml \
----------------
awarzynski wrote:
> victorkingi wrote:
> > awarzynski wrote:
> > > Is a dedicated driver invocation needed for this test?
> > since -fsave-optimization-record and foptimization-record-file both produce 
> > opt-record-file and opt-record-format flags, I couldn't find an easier way 
> > to test for both calls without having to use 2 separate flang-new 
> > invocations.
> > 
> > We could ignore the foptimization-record-file test and assume if 
> > fsave-optimization-record passes, then the former passes as well
> Ah, I see what's happening here. Neither `-fsave-optimization-record ` nor 
> `-foptimization-record-file` are "forwarded" from `flang-new` to `flang-new 
> -fc1` (*). So, shouldn't really be tested in this file. Please remove this 
> change and I'll update the comment in this file to clarify this for our 
> future selves. Apologies for the confusion.
> 
> (*) Forwarding in this context means "passing verbatim from `flang-new` to 
> `flang-new -fc1`"
moved the test to flang/test/Driver/fsave-optimization-record.f90


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D155452/new/

https://reviews.llvm.org/D155452

_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to