[Flang] Move builtin .mod generation into runtimes by Meinersbur · Pull Request #679 · ROCm/llvm-project

Meinersbur · 2025-11-25T14:31:19Z

llvm#169638 pre-merged into amd-staging

github-actions · 2025-11-25T14:31:42Z

Thank you for submitting a Pull Request (PR) to the LLVM Project!

This PR will be automatically labeled and the relevant teams will be notified.

If you wish to, you can add reviewers by using the "Reviewers" section on this page.

If this is not working for you, it is probably because you do not have write permissions for the repository. In which case you can instead tag reviewers by name in a comment by using @ followed by their GitHub username.

If you have received no comments on your PR for a week, you can request a review by "ping"ing the PR by adding a comment “Ping”. The common courtesy "ping" rate is once a week. Please remember that you are asking for valuable time from other developers.

If you have further questions, they may be answered by the LLVM GitHub User Guide.

You can also ask questions in a comment on this PR, on the LLVM Discord or on the forums.

z1-cciauto · 2025-11-25T14:32:26Z

PSDB Link: https://compiler-ci.amd.com/job/compiler-psdb-amd-staging/2964

z1-cciauto · 2025-11-25T15:15:25Z

PSDB Link: https://compiler-ci.amd.com/job/compiler-psdb-amd-staging/2965

z1-cciauto · 2025-11-25T16:08:55Z

PSDB Link: https://compiler-ci.amd.com/job/compiler-psdb-amd-staging/2967

z1-cciauto · 2025-12-01T19:16:52Z

PSDB Link: https://compiler-ci.amd.com/job/compiler-psdb-amd-staging/3036

z1-cciauto · 2025-12-02T13:42:31Z

PSDB Link: https://compiler-ci.amd.com/job/compiler-psdb-amd-staging/3049

z1-cciauto · 2025-12-09T11:37:46Z

PSDB Link: https://compiler-ci.amd.com/job/compiler-psdb-amd-staging/3195

z1-cciauto · 2025-12-09T12:20:46Z

PSDB Link: https://compiler-ci.amd.com/job/compiler-psdb-amd-staging/3196

z1-cciauto · 2025-12-09T12:38:42Z

PSDB Link: https://compiler-ci.amd.com/job/compiler-psdb-amd-staging/3198

Meinersbur · 2025-12-09T12:47:09Z

@ronlieb Merged-in llvm@7675fc7. This includes all the patches since the last merge. Had to resolve two merge conflicts unrelated to the PR:

llvm/lib/Transforms/Vectorize/LoopVectorize.cpp: upstream inlined method
llvm/runtimes/CMakeLists.txt: ROCm 6.3 handling seems to be gone

The test 'flang/test/Integration/amdgpu/debug-target-var.f90' is failing. I think it is unrelated to llvm#169638

These were automatically resolved by kdiff3:

flang/test/Lower/OpenMP/target-enter-data-default-openmp52.f90
llvm/test/Transforms/LoopVectorize/AArch64/predicated-costs.ll

Had to add
74f496e

For aomp to build the amgdcn-amd-amdhsa modules, ROCm/aomp#1737 needs to be applied.

z1-cciauto · 2025-12-09T12:55:57Z

PSDB Link: https://compiler-ci.amd.com/job/compiler-psdb-amd-staging/3199

z1-cciauto · 2025-12-09T13:32:04Z

PSDB Link: https://compiler-ci.amd.com/job/compiler-psdb-amd-staging/3201

z1-cciauto · 2025-12-10T14:14:16Z

PSDB Link: https://compiler-ci.amd.com/job/compiler-psdb-amd-staging/3216

Meinersbur · 2026-05-29T15:05:42Z

This PR is intended @ronlieb to resolve a merge conflict of amd-staging with upstream, it is not intended for an additional review, but GitHub seems to have automatically added reviewers. Sorry for the noise.

…eapply llvm#137828) (llvm#171515)"" This reverts commit ce78ca1.

Before this PR, omp_lib.h is emitted to `${PREFIX}/include` or `${PREFIX}/lib/clang/<version>/include` (install prefix) and `${PREFIX}/runtime/src/omp_lib.h` (builddir prefix). It is never found there because the driver only adds `${PREFIX}/include/flang/OpenMP` to the include path. Fix the `omp_lib.h` include by using the same mechanism as the omp_lib.mod; that is, move it to `${PREFIX}/lib/clang/<version>/finclude/flang/<target-triple>`. The search path is already added by the driver via `-fintrinsics-modules-path` by the driver. Although omp_lib.h currently does not contain anything target-specific, it could do so in the future and I don't think it is worth the effort to add a mechanism without the target triple. It should also me consistent with omp_lib.mod. The changes in detail consist of: 1. Move the omp_lib.h output in the builddir to `${RUNTIMES_OUTPUT_RESOURCE_MOD_DIR}`. This already takes care of whether it is a bootstrapping build or not. 3. Move the omp_lib.h install dir to `${RUNTIMES_INSTALL_RESOURCE_MOD_PATH}`. 4. Remove the implicit search path `include/flang/OpenMP` from the frontend. There is nothing in there anyway. 5. Hardcoding the include search path for testing to the `LLVM_RUNTIME_TARGETS=default` build becomes unnecessary. This was way the `include-omp-header.f90` test was still passing, although it would not work outside the regression tests. Essentially, it tested lit.site.cfg, not the driver. 6. Replace the old `include-omp-header.f90` test. It created a temporary `omp_lib.h` that interferes with any other test that use `include omp_lib.h` (there currently isn't but I originally intended to add the replacement as an additional omp_lib.h test which resulted in a flaky test while the original test was there). Due to how the %flang(_fc1) substitution works, lookup order may also vary.

…) (llvm#171515) Reapplication of llvm#137828 and llvm#169638 Move building the .mod files from openmp/flang to openmp/flang-rt using a shared mechanism. Motivations to do so are: 1. Most modules are target-dependent and need to be re-compiled for each target separately, which is something the LLVM_ENABLE_RUNTIMES system already does. Prime example is `iso_c_binding.mod` which encodes the target's ABI. Constants such as `c_long_double` also have different values). Most other modules have `#ifdef`-enclosed code as well. For instance this caused offload targets nvptx64-nvidia-cuda/amdgpu-amd-amdhsa to use the modules files compiled for the host which may contrain uses of the types REAL(10) or REAL(16) not available for nvptx/amdgpu. 2. CMake has support for Fortran that we should use. Among other things, it automatically determines module dependencies so there is no need to hardcode them in the CMakeLists.txt. Currently the dependencies between builtin modules and `omp_lib.mod` seem to be incorrect: llvm#174474 (comment) 3. It allows using Fortran itself to implement Flang-RT. Currently, only `iso_fortran_env_impl.f90` emits object files that are needed by Fortran applications (llvm#89403). The workaround of llvm#95388 could be reverted (PR If using Flang for cross-compilation or target-offloading, flang-rt must now be compiled for each target not only for the library, but also to get the target-specific module files. For instance in a bootstrapping runtime build, this can be done by adding: `-DLLVM_RUNTIME_TARGETS=default;nvptx64-nvidia-cuda;amdgpu-amd-amdhsa`. Some new dependencies come into play: * openmp depends on flang-rt for building `lib_omp.mod` and `lib_omp_kinds.mod`. Currently, if flang-rt is not found then the modules are not built. * check-flang depends on flang-rt: If not found, the majority of tests are disabled. If not building in a bootstrpping build, the location of the module files can be pointed to using `-DFLANG_INTRINSIC_MODULES_DIR=<path>`, e.g. in a flang-standalone build. Alternatively, the test needing any of the intrinsic modules could be marked with `REQUIRES: flangrt-modules`. * check-flang depends on openmp: Not a change; tests requiring `lib_omp.mod` and `lib_omp_kinds.mod` those are already marked with `openmp_runtime`. As intrinsic are now specific to the target, their location is moved from `include/flang` to `<resource-dir>/finclude/flang/<triple>`. The mechnism to compute the location have been moved from flang-rt (previously used to compute the location of `libflang_rt.*.a`) to common locations in `cmake/GetToolchainDirs.cmake` and `runtimes/CMakeLists.txt` so they can be used by both, openmp and flang-rt. Potentially the mechnism could also be shared by other libraries such as compiler-rt. `finclude` was chosen because `gfortran` uses it as well and avoids misuse such as `#include <flang/iso_c_binding.mod>`. The search location is now determined by `ToolChain` in the driver, instead of by the frontend. Another subdirectory `flang` avoids accidental inclusion of gfortran-modules which due to compression would result in user-unfriendly errors. Now the driver adds `-fintrinsic-module-path` for that location to the frontend call (Just like gfortran does). Since the driver determines the location, tests invoking `flang -fc1` and `bbc` must also be passed the location by llvm-lit. This works like llvm-lit does for finding the include dirs for Clang using `-print-file-name=...`.

Before this PR, omp_lib.h is emitted to `${PREFIX}/include` or `${PREFIX}/lib/clang/<version>/include` (install prefix) and `${PREFIX}/runtime/src/omp_lib.h` (builddir prefix). It is never found there because the driver only adds `${PREFIX}/include/flang/OpenMP` to the include path. Fix the `omp_lib.h` include by using the same mechanism as the omp_lib.mod; that is, move it to `${PREFIX}/lib/clang/<version>/finclude/flang/<target-triple>`. The search path is already added by the driver via `-fintrinsics-modules-path` by the driver. Although omp_lib.h currently does not contain anything target-specific, it could do so in the future and I don't think it is worth the effort to add a mechanism without the target triple. It should also me consistent with omp_lib.mod. The changes in detail consist of: 1. Move the omp_lib.h output in the builddir to `${RUNTIMES_OUTPUT_RESOURCE_MOD_DIR}`. This already takes care of whether it is a bootstrapping build or not. 3. Move the omp_lib.h install dir to `${RUNTIMES_INSTALL_RESOURCE_MOD_PATH}`. 4. Remove the implicit search path `include/flang/OpenMP` from the frontend. There is nothing in there anyway. 5. Hardcoding the include search path for testing to the `LLVM_RUNTIME_TARGETS=default` build becomes unnecessary. This was way the `include-omp-header.f90` test was still passing, although it would not work outside the regression tests. Essentially, it tested lit.site.cfg, not the driver. 6. Replace the old `include-omp-header.f90` test. It created a temporary `omp_lib.h` that interferes with any other test that use `include omp_lib.h` (there currently isn't but I originally intended to add the replacement as an additional omp_lib.h test which resulted in a flaky test while the original test was there). Due to how the %flang(_fc1) substitution works, lookup order may also vary.

With llvm#201297 flang-rt-mod is required for running tests. Disable tests by default if module files are not built.

…/rocm_flang_builtin-mods

llvm#200863 added a new `-foffload-device` argument for informing the frontend that it compiling for the device-side (and as a consequence must not overwrite any module files compiled for the host), but the driver was mistakenly adding `-offload-device`. Also fix the condition and add a regression test for the driver.

…e' into HEAD

ronlieb · 2026-06-09T09:57:07Z

!PSDB

Meinersbur force-pushed the users/meinersbur/rocm_flang_builtin-mods branch from 0f93940 to a41ffe8 Compare November 25, 2025 16:08

Meinersbur force-pushed the users/meinersbur/rocm_flang_builtin-mods branch from a41ffe8 to 4a8fef3 Compare December 1, 2025 19:14

Meinersbur mentioned this pull request Dec 2, 2025

Support for per-target Fortran modules ROCm/aomp#1737

Open

Meinersbur marked this pull request as ready for review December 2, 2025 14:42

Meinersbur force-pushed the users/meinersbur/rocm_flang_builtin-mods branch from 80634ae to 021eaa2 Compare December 10, 2025 14:13

Meinersbur requested review from b-sumner, chinmaydd, david-salinas and lamb-j as code owners May 28, 2026 11:22

Meinersbur force-pushed the users/meinersbur/rocm_flang_builtin-mods branch from 82abb14 to 4b2ee45 Compare May 29, 2026 15:01

Meinersbur requested review from fabianmcg, ftynse, stellaraccident and vangthao95 as code owners May 29, 2026 15:01

Meinersbur removed request for ftynse and stellaraccident May 29, 2026 15:06

Meinersbur requested review from ronlieb and removed request for b-sumner, chinmaydd, david-salinas, fabianmcg, lamb-j and vangthao95 May 29, 2026 15:06

Meinersbur and others added 4 commits June 4, 2026 08:33

Revert "Revert "[Flang] Move builtin .mod generation into runtimes (R…

4945df7

…eapply llvm#137828) (llvm#171515)"" This reverts commit ce78ca1.

[OpenMP] FIx omp_lib.mod compilation for the GPU (llvm#201377)

7932581

Fix tests that have been removed upstream

77b8814

Meinersbur force-pushed the users/meinersbur/rocm_flang_builtin-mods branch from 964b66d to 77b8814 Compare June 4, 2026 13:39

Fox unresolved merge conflicts

64b690d

Meinersbur force-pushed the users/meinersbur/rocm_flang_builtin-mods branch from 3873b24 to 64b690d Compare June 4, 2026 13:55

Meinersbur and others added 6 commits June 4, 2026 09:57

[Flang-RT] Disable tests by default without modules (llvm#201311)

fd88df6

With llvm#201297 flang-rt-mod is required for running tests. Disable tests by default if module files are not built.

[OpenMP] FIx omp_lib.mod compilation for the GPU (llvm#201377)

8f07871

[openmp] runtime omp_lib.h fix

b686b38

Merge remote-tracking branch 'ROCm/amd-staging' into users/meinersbur…

f82ce63

…/rocm_flang_builtin-mods

Meinersbur force-pushed the users/meinersbur/rocm_flang_builtin-mods branch from 5c80b71 to 826eb63 Compare June 5, 2026 12:50

Put the 'f' prefix into -foffload-device

ebbfdf3

Meinersbur force-pushed the users/meinersbur/rocm_flang_builtin-mods branch from 826eb63 to ebbfdf3 Compare June 5, 2026 14:46

Meinersbur and others added 4 commits June 5, 2026 18:12

Remove test that requires preinstall vendor SDK

c1ea859

Merge remote-tracking branch 'ROCm/amd/dev/rlieberm/landModRestructur…

f65e823

…e' into HEAD

merge in amd-staging

6dd7184

ronlieb merged commit 382d708 into amd-staging Jun 9, 2026
105 of 111 checks passed

ronlieb deleted the users/meinersbur/rocm_flang_builtin-mods branch June 9, 2026 10:53

Conversation

Meinersbur commented Nov 25, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented Nov 25, 2025

Uh oh!

z1-cciauto commented Nov 25, 2025

Uh oh!

z1-cciauto commented Nov 25, 2025

Uh oh!

z1-cciauto commented Nov 25, 2025

Uh oh!

z1-cciauto commented Dec 1, 2025

Uh oh!

z1-cciauto commented Dec 2, 2025

Uh oh!

z1-cciauto commented Dec 9, 2025

Uh oh!

z1-cciauto commented Dec 9, 2025

Uh oh!

z1-cciauto commented Dec 9, 2025

Uh oh!

Meinersbur commented Dec 9, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

z1-cciauto commented Dec 9, 2025

Uh oh!

z1-cciauto commented Dec 9, 2025

Uh oh!

z1-cciauto commented Dec 10, 2025

Uh oh!

Meinersbur commented May 29, 2026

Uh oh!

ronlieb commented Jun 9, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Meinersbur commented Nov 25, 2025 •

edited

Loading

Meinersbur commented Dec 9, 2025 •

edited

Loading