FYI: https://github.com/llvm/llvm-project/issues/57786

Thanks for the review @shraiysh & @peixin.
Changes:

• Updated taskgroup.f90 test to include allocate clause. task_reduction clause not added, for support missing/todo while lowering.
• Updated convert-to-llvm-openmp-and-fir.fir test based on feedback. (Some how that test still passed previously as reflected in build-status of the patch also).

Thanks @kiranchandramohan for reviewing this.
Changes:

• Addressed review comments.

NIT corrected.

In D104118#2813358, @stuart wrote:

Note: there is currently no DWARF language code defined for C++ for OpenCL, so we must use DW_LANG_C_plus_plus* if we wish to be able to determine whether output has been generated from C++ for OpenCL source or from OpenCL C source. I have raised DWARF issue 210514.1 to add a dedicated C++ for OpenCL code in the next version of DWARF, but for now I believe that it is best to use DW_LANG_OpenCL for OpenCL C only, and not for C++ for OpenCL.

I could perhaps add a note regarding this to the commit message but am concerned about overcomplicating the message.

Overall I support & I'd like to have one driver i.e flang-new considering all the effort that has gone in it and the long term plan associated with the driver(that will cater to the entire flang project as a primary interface). :)

It looks like the existing test is producing an empty range as a side-effect of the behaviour under test, and so it would be quite possible that the test coverage would be lost if this beahviour changed/the testing method changed for it. What do others think?

Please clang-format the patch :)
FYI https://llvm.org/docs/Contributing.html#how-to-submit-a-patch

Thanks @abidmalikwaterloo for the patch!

LG - Thanks! with one question inline :)

The new driver is still called flang-new

I presume eventually next step would be replace that as well ?

One step at a time :)

+1 :)

LGTM. Thanks for the patch! and congratulations on the milestone of bringing experimental driver to mainline!

Thanks! @fghanim for the patch :)

• Addressed @aprantl review comments. Thanks!.

Minor comment correction.

• Rebased & addressed @aprantl review comments.

Rebased

Changes:

• Rebased to the tip.
• Addressed @aprantl review comments.

Thanks for the review @jmorse!

• Addressed review comments by @jmorse.
• Reinforced checks for both isUnsignedConstant and isSignedConstant.
• Added an invalid expression test case covering isSignedConstant 6 arguments case with operand 0 and operand 2 matching.

• Minor comment updation in isSignedConstant Function in DebugInfoMetadata.cpp

Thanks @aprantl for reviewing this.
Changes:

• Rebased to tip.
• Addressed @dstenb review comments regarding covering fragments and added test case to excercise that.
• Addressed @aprantl review comments of combinining check to a single function i.e isConstant.
• Replaced most checks of isUnsignedConstant with isConstant adding appropriate test case modifications to exercise the changes.

Thanks for review @dstenb !

Patch is stemmed from the discussions from https://reviews.llvm.org/D97311

In D97311#2590688, @aprantl wrote:

Here's an example that seems to work:

...

code is pretty stale here, primary problem still persists. Abandoning for now.

In D97311#2588856, @aprantl wrote:

Have you considered instead introducing a bool isConstant field in DILocalVariable? This way we could reuse all the machinery for local/global variables and only need to switch out the TAG in AsmPrinter.

Could you please add a Unit test for this(as we don't have clang, flang, or MLIR interfacing for dynamic workshare loops).

Thanks for the patch! Could you please clang-format the patch(this is will remove lot of Lint warning/noise) and ease up the review.
If you're using arcanist then that will automatically take care of formatting(provided clang-format should be in system path.) Or if you're using web UI then you may have to do it manually (git-clang-format HEAD~).
More details: https://llvm.org/docs/Phabricator.html

Resolved the previous mentioned issue. Ready for review.

Rebase + clang-format

Thanks! for the patch.
This code touches some of the clang part, Anyway changes are pretty self-explanatory. I'll leave this one to you, if you want land it or wait from someone from clang community to take a look :)

Thanks! for the patch @awarzynski . I support this enhancement. LGTM(Not accepting formally :) ), Let's wait and hear from others also, what's their take on this.

Addressed @tskeith comments.

In D67563#1675361, @aprantl wrote:

In D67563#1675356, @aprantl wrote:

How do Fortran compilers represent regular versus inout arguments in DWARF?

According to godbolt.org, gfortran 8.2 emits the same DWARF for in and inout parameters. That doesn't help :-)

@mehdi_amini are you Okay with this patch ?

Can be tested locally as:

make(or ninja) check-flang-fir

Enabled tests for this in: https://reviews.llvm.org/D95958
Builds clean.
@mehdi_amini Apologies, I totally forgot and landed without your approval :(, if you have any concerns please share I'll try to address as a post commit thing or if you want I can revert it.