vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Add some minimal optimizations for CDNA (#10498)

Browse source 
* Add some minimal optimizations for CDNA

* ggml_cuda: set launch bounds also for GCN as it helps there too
This commit is contained in:
uvos 2024-11-27 17:10:08 +01:00 • committed by GitHub
parent 46c69e0e75
commit 3ad5451f3b
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
6 changed files with 36 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

8
ggml/src/ggml-cuda/vendors/hip.h vendored
View file

@ -95,6 +95,14 @@
#define __CUDA_ARCH__ 1300
#if defined(__gfx803__) || defined(__gfx900__) || defined(__gfx906__)
#define GCN
#endif
#if defined(__gfx908__) || defined(__gfx90a__) || defined(__gfx942__)
#define CDNA
#endif
#if defined(__gfx1100__) || defined(__gfx1101__) || defined(__gfx1102__) || defined(__gfx1103__) || \
defined(__gfx1150__) || defined(__gfx1151__)
#define RDNA3