[PATCH] D29203: Change debug-info-for-profiling from a TargetOption to a function attribute.
David Blaikie via llvm-commits
llvm-commits at lists.llvm.org
Wed Feb 1 17:08:03 PST 2017
On Wed, Feb 1, 2017 at 4:56 PM David L. Jones via Phabricator <
reviews at reviews.llvm.org> wrote:
> dlj added inline comments.
> Comment at: test/DebugInfo/Generic/gmlt.test:2
> ; REQUIRES: object-emission
> ; RUN: %llc_dwarf -O0 -filetype=obj < %S/../Inputs/gmlt.ll |
> llvm-dwarfdump - | FileCheck %S/../Inputs/gmlt.ll
> +; RUN: %llc_dwarf -O0 -filetype=obj < %S/../Inputs/gmlt_profiling.ll |
> llvm-dwarfdump - | FileCheck %S/../Inputs/gmlt_profiling.ll
> dblaikie wrote:
> > This line should probably be in gmlt.ll (both gmlt.ll and
> gmlt_profiling.ll would move from test/DebugInfo/Inputs to
> Moving this file to Generic/ means that other tests (notably, under X86/)
> now use files not within their Inputs, or within their parents' Inputs.
> Was that the intent of this comment?
Ah - hadn't looked at that. If it's worth keeping the distinct tests
sharing non-trivial IR, best to switch it back to the way it was. If the
IR's trivial enough, might be easier for each test to have its own (makes
the tests easier to read, etc).
Ah, I see - it's so that it's easier to filter the different tests, but
they're still pretty legible because all FileCheck logic is in the IR file.
Not sure why there's an X86 version of this test if there's also a generic
The intent wasn't to break any other tests, no - it looked like the test
was generic (still looks like it - so I'm a bit confused) so I wasn't sure
why it would be referring to a separate IR file (still not sure why,
I guess it's because the darwin case is XFAIL'd so it had to be in a
Looks like maybe this has all got a bit confused. I'll take a look.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the llvm-commits