Skip to content

Fix Intel/LLVM openmp build #407

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 10 commits into
base: main
Choose a base branch
from
Open

Fix Intel/LLVM openmp build #407

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
20aa25c
Fix Intel/LLVM openmp build
isuruf Jul 31, 2025
c3d300d
use llvm openmp everywhere
isuruf Aug 11, 2025
e8228ab
skip not win
isuruf Aug 11, 2025
245972f
debug
isuruf Aug 22, 2025
29c2d5c
debug
isuruf Aug 22, 2025
0a96b31
fix test
isuruf Aug 22, 2025
1992483
try forwarding instead of copying
isuruf Aug 23, 2025
4d1d683
no temp dir
isuruf Aug 25, 2025
39fb3bc
remove debug
isuruf Aug 25, 2025
ef25adc
use clang-cl
isuruf Sep 3, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion recipe/meta.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
# if you wish to build release candidate number X, append the version string with ".rcX"
{% set version = "2.7.1" %}
{% set build = 4 %}
{% set build = 5 %}

# Use a higher build number for the CUDA variant, to ensure that it's
# preferred by conda's solver, and it's preferentially
Expand Down Expand Up @@ -69,6 +69,7 @@ source:
- patches/0016-skip-test_norm_matrix_degenerate_shapes-on-numpy-2.3.patch
# backport https://github.com/pytorch/pytorch/pull/127702
- patches/0017-Define-PY_SSIZE_T_CLEAN-before-include-Python.h.patch
- patches/0018-Use-Intel-LLVM-openmp.patch
- patches_submodules/fbgemm/0001-remove-DESTINATION-lib-from-CMake-install-directives.patch # [win]
- patches_submodules/tensorpipe/0001-switch-away-from-find_package-CUDA.patch

Expand Down
31 changes: 31 additions & 0 deletions recipe/patches/0018-Use-Intel-LLVM-openmp.patch
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,31 @@
From ffbdd8bb8fa38397ad6d7958a8c4ddd80ff738eb Mon Sep 17 00:00:00 2001
From: Isuru Fernando <[email protected]>
Date: Thu, 31 Jul 2025 12:34:30 -0500
Subject: [PATCH] Use Intel/LLVM openmp

This make MSVC use __kmp_* symbols available in both libiomp.dll
and libiomp5md.dll instead of vcomp* symbols which are only in
vcomp140.dll and libiomp5md.dll, but are missing in libiomp.dll
---
cmake/Modules/FindOpenMP.cmake | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/cmake/Modules/FindOpenMP.cmake b/cmake/Modules/FindOpenMP.cmake
index 8a9abff398..336a9a927a 100644
--- a/cmake/Modules/FindOpenMP.cmake
+++ b/cmake/Modules/FindOpenMP.cmake
@@ -122,9 +122,9 @@ function(_OPENMP_FLAG_CANDIDATES LANG)
set(OMP_FLAG_MIPSpro "-mp")
if(__header_dir MATCHES ".*Microsoft Visual Studio.*")
# MSVC header. No need to pass it as additional include.
- set(OMP_FLAG_MSVC "-openmp:experimental" "-openmp")
+ set(OMP_FLAG_MSVC "-openmp:llvm" "-openmp:experimental" "-openmp")
else()
- set(OMP_FLAG_MSVC "-openmp:experimental -I${__header_dir}" "-openmp -I${__header_dir}")
+ set(OMP_FLAG_MSVC "-openmp:llvm" "-openmp:experimental -I${__header_dir}" "-openmp -I${__header_dir}")
endif()
set(OMP_FLAG_PathScale "-openmp")
set(OMP_FLAG_NAG "-openmp")
--
2.45.2

Loading