ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-04 21:51:28 +00:00
Code Issues Packages Projects Releases Wiki Activity

[CK_TILE] wa prec, remove sgpr offset for inline asm (#1356)

Browse Source 
* wa prec, remove sgpr offset for inline asm

* macro for set tile

* ignore unused param if no kernel instances in host API

* fix more prec issue

* cache buffer resource

* fix

* support pre-nop

* clear tile by vector type members

* add workaround to reduce scratch memory

* conditionally enable workaround code

* enable workaround start from certain build version

* fallback set_tile() implementation from certain build version

* undo template argument changes

* put dummy asm in load_raw()

* fix comments, refactor s_nop inside buffer_load

---------

Co-authored-by: PoYen, Chen <PoYen.Chen@amd.com>
This commit is contained in:
carlushuang
2024-07-09 02:09:55 +08:00
committed by GitHub
parent eb44e0472a
commit 8182976c37
13 changed files with 581 additions and 243 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
include/ck_tile/core/config.hpp
View File

@@ -21,6 +21,7 @@
#define __gfx12__
#endif
#include "hip/hip_version.h"
#ifndef CK_TILE_DONT_USE_HIP_RUNTIME_HEADERS
#include "hip/hip_runtime.h"
#include "hip/hip_fp16.h"
@@ -147,6 +148,14 @@
#define CK_TILE_WORKAROUND_SWDEV_XXXXXX_INT8_DS_WRITE_ISSUE 1
#endif
#ifndef CK_TILE_WORKAROUND_ROCM_6_1_SCRATCH_MEMORY_ISSUE
#if HIP_VERSION_MAJOR == 6 && HIP_VERSION_MINOR == 1 && HIP_VERSION_PATCH >= 40091
#define CK_TILE_WORKAROUND_ROCM_6_1_SCRATCH_MEMORY_ISSUE 1
#else
#define CK_TILE_WORKAROUND_ROCM_6_1_SCRATCH_MEMORY_ISSUE 0
#endif
#endif
#ifndef CK_TILE_DEBUG_LOG
#define CK_TILE_DEBUG_LOG 0
#endif