Skip to content

GitLab

Unverified Commit a29d2b76 authored by Haocong WANG's avatar Haocong WANG Committed by GitHub
Browse files

Merge pull request #1 from aska-0096/e2e_v2 

Merging some forgot changes
parents 2c265ebd 87ae9b71
Hide whitespace changes
Inline Side-by-side
Showing with 4 additions and 4 deletions
example/32_batched_gemm_scale_softmax_gemm/batched_gemm_scale_softmax_gemm_permute_wmma_fp16.cpp
View file @ a29d2b76
...@@ -96,12 +96,12 @@ using DeviceGemmInstance = ...@@ -96,12 +96,12 @@ using DeviceGemmInstance =
256, 256,
// Gemm 0 // Gemm 0
128, // MPerBlock 128, // MPerBlock
64, // LPerBlock 64, // LPerBlock
32, // KPerBlock 64, // KPerBlock
8, // K1 8, // K1
// Gemm 1 // Gemm 1
64, // NPerBlock 64, // NPerBlock
32, // LTilePerBlock 64, // LTilePerBlock
8, // L1 8, // L1
16, // MPerWMMA 16, // MPerWMMA
16, // LPerWMMA 16, // LPerWMMA
......
include/ck/ck.hpp
View file @ a29d2b76
...@@ -36,7 +36,7 @@ ...@@ -36,7 +36,7 @@
#elif defined(__gfx1030__) // for GPU code #elif defined(__gfx1030__) // for GPU code
#define CK_BUFFER_RESOURCE_3RD_DWORD 0x31014000 #define CK_BUFFER_RESOURCE_3RD_DWORD 0x31014000
#elif defined(__gfx1100__) || defined(__gfx1101__) || defined(__gfx1102__) // for GPU code #elif defined(__gfx1100__) || defined(__gfx1101__) || defined(__gfx1102__) // for GPU code
#define CK_BUFFER_RESOURCE_3RD_DWORD 0x10020000 #define CK_BUFFER_RESOURCE_3RD_DWORD 0x31004000
#endif #endif
// FMA instruction // FMA instruction
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment