User Details
- User Since
- Nov 15 2019, 2:59 AM (137 w, 6 d)
Tue, Jul 5
Leftovers, if any, can be covered in a follow-up patch.
Wed, Jun 22
Ping
Fri, Jun 17
clang-formatted.
Tue, Jun 14
Added comments for parseTargeID() .
Forgot to invert the condition in an if block.
Wed, Jun 8
Jun 2 2022
Could you please add support for HSA_ISA_INFO_NAME using hsa_isa_get_info_alt?
This one gives triple and target ID information, like amdgcn-amd-amdhsa--gfx908:sramecc-:xnack-
Jun 1 2022
Added the multi-entry logic in libomptarget. Yet to move the image compatibility testing to plugin.
May 26 2022
Thanks for the detailed review. I will update rest of the patch soon.
Addressed some simple review changes. Will update remaining in the next iteration.
May 25 2022
Changed the embedding scheme to add ImageInfo field in __tgt_device_image.
May 6 2022
Thanks!
LGTM.
Ideally we could just put this on the linker itself, but nvlink doesn't seem to support .a files.
Looks good to me.
May 2 2022
May 1 2022
Apr 28 2022
Apr 27 2022
Apr 12 2022
Apr 11 2022
Apr 8 2022
Mar 30 2022
Hey @jhuber6 , as discussed in multi-company meeting, I think that we will need at least an arch field somewhere in this. We would like to create multi-arch binaries so that runtime can load the compatible one on its own.
You may even consider using TargetID Format to store the list of archs.
Mar 29 2022
LGTM
Mar 2 2022
Mar 1 2022
Feb 28 2022
Feb 22 2022
Will it be possible to use Bundle Entry ID format [1] for naming sections and being sort of the the primary key to refer an offload object?
The arch string following the triple will be just offload-arch (or march, or mcpu) for others, but for amdgpu it may contain some more information.
Feb 8 2022
Feb 4 2022
Feb 3 2022
Restored device ancestor codegen unit test with requires reverese_offload.
Jan 31 2022
Can you please add this tool's doc in clang/docs?
Jan 27 2022
Jan 19 2022
It seems that this patch along with D117156 and D117246 is giving patch application failed error [https://buildkite.com/llvm-project/diff-checks/builds/82688].
arc patch is also giving the same error.
Jan 18 2022
Ping :-)
Jan 12 2022
Jan 11 2022
- Used a common diagnostic warning warn_unknown_declare_variant_isa_trait for ParseOpenMP and SemaOpenMP for decalre variant and metadirectives.
- Split lit codegen tests into two files, one requiring amdgpu-registered target and another for host only.
- Added warning message lit test at an appropriate place.
Jan 10 2022
Fixed the lit test failing in pre-check build bot.
Jan 7 2022
Added diagnostic remarks for when ISA trait is not selected.
Jan 6 2022
Added target specific tests for ISA traits, for CPU as well as GPU.
Jan 4 2022
Jan 3 2022
Specs 5.0 and 5.1 linked above doesn't contain PGI as a vendor. But, we can leave it as is.
- Added nec as compiler vendor.
- Fixed clang test cases.
- I don't know why "pgi" is present here as a vendor. Should it be renamed as nvidia instead? @Meinersbur @tra
- Should "cray" be renamed as "hpe" here? @sandoval
- Should this vendor list be reordered as per Spec 5.1 [https://www.openmp.org/wp-content/uploads/OpenMP-API-Additional-Definitions-2-0.pdf] ?
Updated comments as per suggestion. Apologies for the delay.
Oct 12 2021
clang-format(ed).
Thanks, Michael and Artem!
Used comment style for named parameters.
Oct 11 2021
Fixed typo
- Changed the option from path to nvlink-command.
- Command line arguments are now parsed using proper API.
Oct 9 2021
Oct 8 2021
Thanks for reporting. Taking a look.