[Flang][OpenMP] Explicit member mapping fortran offload tests now supported through extending derived type map support in Flang and the OpenMP dialect #81508
Conversation
Created using spr 1.3.4
agozillon
requested review from
razvanlupusoru,
ergawy,
clementval,
skatrak,
jsjodin,
kiranchandramohan,
raghavendhra,
DominikAdamski,
dpalermo,
kiranktp,
vzakhari and
kparzysz
Created using spr 1.3.4
Created using spr 1.3.4
Created using spr 1.3.4
|
a little ping for some reviewer attention if at all possible please to get some forward momentum on the PR stack, thank you very much ahead of time! |
|
Hello, this branch name is too long and prevents me from checking out llvm on Windows. Can you please delete it and recreate the pr with a shorter branch name? This ends up using 203 bytes of the allowed 260 on Windows. |
Will do, sorry for the bother. |
agozillon
deleted the
users/agozillon/spr/flangopenmp-explicit-member-mapping-fortran-offload-tests-now-supported-through-extending-derived-type-map-support-in-flang-and-the-openmp-dialect
branch
|
Will re-open with a shorter name in a bit |
|
Thanks! |
You're welcome, sorry again, please do let me know if the newly opened ones are an issue (if I have calculated correctly I don't believe they are). If they are I am happy to delete them again. and next time I open them I'll avoid utilising SPR. |
|
The real issue isn't SPR, it's that a commit title should really be short: try to stay <80char for the title and provide a complete description. See https://cbea.ms/git-commit/ for some details and examples. |
Sounds good |
| !CHECK: 10 | ||
| !CHECK: 2. | ||
| !CHECK: (2.,10.) | ||
|
|
There was a problem hiding this comment.
Nit: Missing newline, or extra?
| print *, scalar | ||
| end program | ||
|
|
||
| !CHECK: 10 |
There was a problem hiding this comment.
Nit: same here, looks like missing newline.
This is a large series of runtime tests that help to add coverage for the specific cases intended to be supported by the PR stack
that extends derived type map support in Flang+OpenMP. Primarily this will add functionality coverage, there's cases where
things may work, but not optimally (or at least similarly to the status quo in Clang), addiitonal IR tests are added in the
relevant segments of the related PRs to test for breakages like that.