Add comment about bf16 to CMake docs

This commit is contained in:
Robert Schulze 2024-11-15 11:50:08 +00:00
parent 3b0b42cbbb
commit 9c5017323d
No known key found for this signature in database
GPG Key ID: 26703B55FB13728A

View File

@ -74,6 +74,7 @@ elseif (ARCH_AARCH64)
# introduced as optional, either in v8.2 [7] or in v8.4 [8]. # introduced as optional, either in v8.2 [7] or in v8.4 [8].
# rcpc: Load-Acquire RCpc Register. Better support of release/acquire of atomics. Good for allocators and high contention code. # rcpc: Load-Acquire RCpc Register. Better support of release/acquire of atomics. Good for allocators and high contention code.
# Optional in v8.2, mandatory in v8.3 [9]. Supported in Graviton >=2, Azure and GCP instances. # Optional in v8.2, mandatory in v8.3 [9]. Supported in Graviton >=2, Azure and GCP instances.
# bf16: Bfloat16, a half-precision floating point format developed by Google Brain. Optional in v8.2, mandatory in v8.6.
# #
# [1] https://github.com/aws/aws-graviton-getting-started/blob/main/c-c%2B%2B.md # [1] https://github.com/aws/aws-graviton-getting-started/blob/main/c-c%2B%2B.md
# [2] https://community.arm.com/arm-community-blogs/b/tools-software-ides-blog/posts/making-the-most-of-the-arm-architecture-in-gcc-10 # [2] https://community.arm.com/arm-community-blogs/b/tools-software-ides-blog/posts/making-the-most-of-the-arm-architecture-in-gcc-10