mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-05-14 02:02:46 +00:00
67dcc552b6
* create files for xdlops
* working on blockwise_gemm_xdlops
* add KReduction
* add m/n repeats
* add 2x2 pipeline
* added 128x128 wavegemm
* use StaticBuffer of vector_type
* break vector type to blk_size
* add kpack into xldops_gemm and blockwise_gemm
* abroadcast only
* add fp32 mfma instructions
* adding fp16 mfma
* pack half4_t
* rename kperwave to kpack
* add 32x32x8fp16
* add fp16 mfma
* clean code
* clean code
* V4r4 xdlops kpack (#35)
* add kpack with incorrect results
* bug fix for make_dynamic_naive_tensor_descriptor_aligned_v2
* add 1x1 kernel
* add gridwise_gemm_v2 - single_buffer
* enabled dwordx4 for fp16
Co-authored-by: Chao Liu <chao.liu2@amd.com>
* refactor fwd-v4r4-xdlops
* add v4r4-nhwc-xdlop
* improve some perf of nhwc and nchw by tuning parameters, and change scheuduling in gridwise-gemm loop
* tweak scheduling in gridwise gemm
* add v4r3 with a single output copy
* init commit: output with slice win
* adding sliceWin
* add multiple repeats pattern
* starting adding bwd-v4r1-xdlops
* use tuple as SrcBuffer
* adding bwd-data v4r1 nhwc xdlops
* fix bug in make_dynamic_naive_tensor_descriptor_aligned_v2()
* fix bug in host bwd-data conv
* initial implementation of bwd-data v4r1 nhwc xdlops
* add launch bound flags
* enable launch bound
* add m/nrepeat=4
* tweak bwd-data v4r1 nhwc xdlops
* added bwd-data v4r1 nhwc xlops with output A and weight B
* add fwd-v4r4 nhwc xdlops, A input, B weight, C output
Co-authored-by: Chao Liu <chao.liu2@amd.com>
[ROCm/composable_kernel commit: 3835318cc3]
47 lines
2.3 KiB
Bash
Executable File
47 lines
2.3 KiB
Bash
Executable File
#!/bin/bash
|
|
|
|
## GPU visibility
|
|
export ROCR_VISIBLE_DEVICE=0
|
|
export GPU_DEVICE_ORDINAL=0
|
|
|
|
## Boost
|
|
#export LD_LIBRARY_PATH=/usr/local/lib:$LD_LIBRARY_PATH
|
|
|
|
## Compiling
|
|
export OLC_DEBUG_HIP_VERBOSE=1
|
|
export OLC_DEBUG_HIP_DUMP=1
|
|
export OLC_DEBUG_SAVE_TEMP_DIR=1
|
|
|
|
#make -j conv_driver_v2
|
|
#make -j conv_bwd_data_driver_v2
|
|
make -j conv_driver_v2_olc
|
|
|
|
rm -rf /root/_hip_binary_kernels_/
|
|
rm -rf /tmp/olCompile*
|
|
|
|
LAYOUT=$1
|
|
ALGO=$2
|
|
VERIFY=$3
|
|
INIT=$4
|
|
LOG=$5
|
|
REPEAT=$6
|
|
|
|
################################ layout algo verify init log repeat N__ K___ C___ Y X Hi_ Wi__ Strides Dilations LeftPads RightPads
|
|
./conv_driver_v2 $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 128 256 192 3 3 71 71 2 2 1 1 1 1 1 1
|
|
#./conv_driver_v2 $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 128 256 1024 1 7 17 17 1 1 1 1 0 3 0 3
|
|
#./conv_driver_v2 $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 256 256 2048 3 3 14 14 1 1 1 1 1 1 1 1
|
|
#./conv_driver_v2 $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 256 256 256 3 3 14 14 1 1 1 1 1 1 1 1
|
|
#./conv_driver_v2 $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 256 512 512 3 3 7 7 1 1 1 1 1 1 1 1
|
|
|
|
#./conv_driver_v2 $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 128 512 192 3 3 35 35 2 2 1 1 0 0 0 0
|
|
#./conv_driver_v2 $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 256 256 256 3 3 30 30 2 2 1 1 0 0 0 0
|
|
#./conv_driver_v2 $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 256 512 512 3 3 16 16 2 2 1 1 0 0 0 0
|
|
|
|
#./conv_driver_v2 $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 256 2048 1024 1 1 14 14 2 2 1 1 0 0 0 0
|
|
#./conv_driver_v2 $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 256 256 1024 1 1 14 14 1 1 1 1 0 0 0 0
|
|
#./conv_driver_v2 $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 256 512 2048 1 1 7 7 1 1 1 1 0 0 0 0
|
|
|
|
#./conv_bwd_data_driver_v2 $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 256 256 256 3 3 14 14 1 1 1 1 1 1 1 1
|
|
|
|
./conv_driver_v2_olc $LAYOUT $ALGO $VERIFY $INIT $LOG $REPEAT 128 256 192 3 3 71 71 2 2 1 1 1 1 1 1
|