Skip to content
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

Jump to bottom

Move CUDA builds from C++14 to C++17 #354

Merged
merged 24 commits into from
Jan 26, 2022
Merged

Move CUDA builds from C++14 to C++17 #354

merged 24 commits into from
Jan 26, 2022

Conversation

valassi
Copy link
Member

@valassi valassi commented Jan 26, 2022

This is the PR to fix #333.

Still WIP for now but will complete and merge soon

@valassi valassi self-assigned this Jan 26, 2022
valassi added 16 commits January 26, 2022 11:04
@valassi
[cpp17] bug fix in OMPFLAGS - effectively reenable OpenMP (code was n…
db70f9c 
…ot built with OpenMP for a long time)

(Originally from the nvccsimd branch in Dec 2021)
@valassi
[cpp17] remove OMPFLAGS support for gcc<9.3 in Makefile
f4c80a1
@valassi
[cpp17] require gcc>=9.3 (madgraph5#96) - it was observed in the past…
9f86299 
… that OpenMP required this (madgraph5#269)
@valassi
[cpp17] fix OpenMP MT builds after effectively reenabling OpenMP
d93bfb9
@valassi
[cpp17] remove CUFLAGS CXXFLAGS CPPFLAGS from link instructions (keep…
be34933 
… for compilation)

(Originally from the nvccsimd branch in Dec 2021)
@valassi
[cpp17] add back OMPFAGS to linker after removing CXXFLAGS
d4feb80 
(Originally from the nvccsimd branch in Dec 2021)
@valassi
[cpp17] Use __CUDACC_VER_MAJOR__ instead of __CUDACC__ in getCompiler…
5c7ebbe 
… to detect if nvcc was used

(Originally from the nvccsimd branch in Dec 2021)
@valassi
[cpp17] improve the previous patch: use __NVCC__ instead of __CUDA_MA…
7415f1f 
…JOR_VERSION__ to detect if nvcc was used as compiler
@valassi
[cpp17] TEMPORARELY DISABLE OMP (need to reassess MT)
5245d4b
@valassi
[cpp17] further use __NVCC__ instead of __CUDACC__ throughout getComp…
d2725d6 
…iler
@valassi
[cpp17] complete backport to codegen and regenerate eemumu auto
87bde06
@valassi
[cpp17] BUG FIX IN LOGS! recover gcc102 baselines from c2e67b4
e8979bc 
These had been overridden by mistakes in 1e7fd92
@valassi
[cpp17] rerun eemumu basic log, check all is ok
157d430
@valassi
[cpp17] add ggtg and ggttggg to sync script
0654f26
@valassi
[cpp17] regenerate ggtt* auto and resync manu
fd6d8de
@valassi
[cpp17] ** COMPLETE CPP17 ** rerun ggtt/ggttg/ggttgg, all ok
b26408b 
(ggttggg takes too long to build, cannot be used routinely...)
@valassi
Copy link
Member Author

valassi commented Jan 26, 2022

This is now ready to be merged, it includes

@valassi
Copy link
Member Author

valassi commented Jan 26, 2022

I have added a (TEMPORARY?) patch to allow gcc<9.3 when it is used as toolchain for clang, else the CI was failing (#355)

This was referenced Jan 26, 2022
Closed
Closed
@valassi
Copy link
Member Author

valassi commented Jan 26, 2022

All checks pass, I am self merging

@valassi valassi merged commit 06522ae into madgraph5:master Jan 26, 2022
This was referenced Jan 26, 2022
Closed
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees

@valassi valassi

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

Move nvcc builds back from c++14 to c++17
1 participant
@valassi