[HIPIFY][Device][#1796][tests][fix] Fix device test failures on CUDA 11.8.0, 12.0.x, and 12.1.x #1846
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…UDA 11.8.0, 12.0.x, and 12.1.x + Added a separated file with device tests for CUDA < 11.8 || CUDA > 12.1 and the following device/host APIs: `__hmax2`, `__lows2bfloat162`, `__hmin2`, `__low2bfloat16`, and few others + [Reason] The above device/host APIs were erroneously decorated in those CUDA versions as `__device__` only, which led to clang compilation errors like: "no matching function for call to `bfloat162bfloat162`: cuda_bf16.hpp:1453:35: note: candidate function not viable: call to `__device__` function from `__host__` function" + The problem was fixed in `CUDA 12.2.0` when those APIs started to be decorated as device/host: `__CUDA_HOSTDEVICE_BF16_DECL__` + Added the `Changed` CUDA version to the above APIs + [ToDo] Revise the rest of the supported device functions
emankov
added a commit
to emankov/HIPIFY
that referenced
this pull request
Jan 29, 2025
…n to the rest device `fp16` APIs + Updated the regenerated `Device` `CUDA2HIP` docs accordingly
emankov
added a commit
that referenced
this pull request
Jan 29, 2025
[HIPIFY][Device][#1846][doc][fix] Added the `Changed` CUDA version to…
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
__hmax2,__lows2bfloat162,__hmin2,__low2bfloat16, and few others__device__only, which led to clang compilation errors like: "no matching function for call tobfloat162bfloat162: cuda_bf16.hpp:1453:35: note: candidate function not viable: call to__device__function from__host__function"CUDA 12.2.0when those APIs started to be decorated as device/host:__CUDA_HOSTDEVICE_BF16_DECL__ChangedCUDA version to the above APIs