github/ROCm
1
1
Fork 0
mirror of https://github.com/ROCm/ROCm.git synced 2026-01-09 14:48:06 -05:00
Code Issues Packages Projects Releases Wiki Activity

Update RELEASE.md (#5349)

Browse Source 
* Update RELEASE.md

update llvm-project URL

* Update .wordlist.txt

add spelling errors
This commit is contained in:
randyh62
2025-09-16 09:38:23 -07:00
committed by GitHub
parent 8c40d14d7e
commit 37faf170b1
2 changed files with 8 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
.wordlist.txt
View File

@@ -72,6 +72,7 @@ CU
CUDA CUDA
CUs CUs
CXX CXX
CX
Cavium Cavium
CentOS CentOS
ChatGPT ChatGPT
@@ -118,6 +119,8 @@ Dependabot
Deprecations Deprecations
DevCap DevCap
DirectX DirectX
Disaggregated
disaggregated
Dockerfile Dockerfile
Dockerized Dockerized
Doxygen Doxygen
@@ -127,6 +130,7 @@ ENDPGM
EPYC EPYC
ESXi ESXi
EoS EoS
etcd
fas fas
FBGEMM FBGEMM
FIFOs FIFOs
@@ -178,6 +182,7 @@ GPUs
Graphbolt Graphbolt
GraphSage GraphSage
GRBM GRBM
GRE
GenAI GenAI
GenZ GenZ
GitHub GitHub
@@ -301,6 +306,7 @@ MirroredStrategy
Mixtral Mixtral
MosaicML MosaicML
MoEs MoEs
Mooncake
Mpops Mpops
Multicore Multicore
Multithreaded Multithreaded
@@ -445,6 +451,7 @@ SKU
SKUs SKUs
SLES SLES
SLURM SLURM
Slurm
SMEM SMEM
SMFMA SMFMA
SMI SMI

2
RELEASE.md
View File

@@ -172,7 +172,7 @@ Key compiler enhancements include:
* Comgr: * Comgr:
* Added support for an in-memory virtual file system (VFS) for storing temporary files generated during intermediate compilation steps. This is designed to improve performance by reducing on-disk file I/O. Currently, VFS is supported only for the device library link step, with plans for expanded support in future releases. * Added support for an in-memory virtual file system (VFS) for storing temporary files generated during intermediate compilation steps. This is designed to improve performance by reducing on-disk file I/O. Currently, VFS is supported only for the device library link step, with plans for expanded support in future releases.
* SPIR-V: * SPIR-V:
* Improved [target-specific extensions](https://github.com/ROCm/llvm-project/blob/c2535466c6e40acd5ecf6ba1676a4e069c6245cc/clang/docs/LanguageExtensions.rst): * Improved [target-specific extensions](https://github.com/llvm/llvm-project/blob/main/clang/docs/LanguageExtensions.rst#target-specific-extensions):
* Added a new target-specific builtin ``__builtin_amdgcn_processor_is`` for late or deferred queries of the current target processor. * Added a new target-specific builtin ``__builtin_amdgcn_processor_is`` for late or deferred queries of the current target processor.
* Added a new target-specific builtin ``__builtin_amdgcn_is_invocable``, enabling fine-grained, per-builtin feature availability. * Added a new target-specific builtin ``__builtin_amdgcn_is_invocable``, enabling fine-grained, per-builtin feature availability.
* The compiler driver now uses parallel code generation by default when compiling using full LTO (including when using the `-fgpu-rdc` option) for HIP. This divides the optimized LLVM IR module into roughly equal partitions before instruction selection and lowering, which can help improve build times. * The compiler driver now uses parallel code generation by default when compiling using full LTO (including when using the `-fgpu-rdc` option) for HIP. This divides the optimized LLVM IR module into roughly equal partitions before instruction selection and lowering, which can help improve build times.