mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-04-20 14:59:17 +00:00
Add json dump support to output details from CK/CKTile Examples. (#2551)
* Adding RapidJson Library * Adding Json Dumps in all CK_Tile Examples Not verified yet * Adding json to cktile Batched Transpose * adding json dumps to layernorm2d_fwd * Adding json dump to flatmm_basic * Adding RapidJson Library * Adding Json Dumps in all CK_Tile Examples Not verified yet * Adding json to cktile Batched Transpose * adding json dumps to layernorm2d_fwd * Adding json dump to flatmm_basic * Adding json in 03_gemm * Add json dump to 16_batched_gemm * Add json dump to gemm_multi_d_fp16 * Add json dump to grouped_gemm * fix fmha_bwd/fwd * Fix clang-format errors exclude include/rapidjson in jenkins as its a third-party library * Saparating function and defination. * Update Documentation of 03_gemm * Refactoring as per code review * Disable fp8 instances on unsupported targets (#2592) * Restrict building of gemm_universal_preshuffle_f8 instances to specific targets in CMakeLists.txt * Add condition to skip gemm_xdl_universal_preshuffle_f8 instances for unsupported targets in CMakeLists.txt * Add conditions to skip unsupported targets for gemm_universal_preshuffle_f8 and gemm_xdl_universal_preshuffle_f8 instances in CMakeLists.txt * Refine conditions to exclude gemm_universal_preshuffle_f8 instances for unsupported targets in CMakeLists.txt --------- Co-authored-by: AviralGoelAMD <aviralgoel@amd.com> * fix clang format * remove duplicate lines of code from library/src/tensor_operation_instance/gpu/CMakeLists.txt * Fixing Readme and unifying jsondumps * adding moe_smoothquant * adding fused_moe * Fixing Readme for batched_gemm * Fixing Readme for grouped_gemm * adding flatmm * adding gemm_multi_d_fp16 * adding elementwise * adding File name when json is dumped * Fixing Reduce after merge * adding batched_transpose * Adding Warptile in Gemm * Fixing Clang Format --------- Co-authored-by: Aviral Goel <aviral.goel@amd.com> Co-authored-by: AviralGoelAMD <aviralgoel@amd.com> Co-authored-by: illsilin_amdeng <Illia.Silin@amd.com>
This commit is contained in:
rahjain-amd
@@ -6,17 +6,34 @@ This folder contains example for Rmsnorm2D forward using ck_tile tile-programmin
|
||||
```
|
||||
# in the root of ck_tile
|
||||
mkdir build && cd build
|
||||
../script/cmake-ck-dev.sh ../ <arch> # you can replace this <arch> to gfx90a, gfx942...
|
||||
make tile_rmsnorm2d_fwd -j
|
||||
sh ../script/cmake-ck-dev.sh ../ <arch> # you can replace this <arch> to gfx90a, gfx942...
|
||||
make tile_rmsnorm2d_fwd -j`nproc`
|
||||
```
|
||||
This will result in an executable `build/bin/tile_rmsnorm2d_fwd`
|
||||
|
||||
## cmdline
|
||||
```
|
||||
args:
|
||||
-m m dimension (default:3328)
|
||||
-n m dimension (default:4096)
|
||||
-e epsilon (default:1e-5)
|
||||
-v cpu validation or not (default:1)
|
||||
-prec precision (default:fp16)
|
||||
-m m dimension (default:3328)
|
||||
-n n dimension (default:4096)
|
||||
-x_stride x row_stride, if -1 then equal to n (default:-1)
|
||||
-xr_stride x residule row_stride, if -1 then equal to n (default:-1)
|
||||
-y_stride y row_stride, if -1 then equal to n (default:-1)
|
||||
-yr_stride y residule row_stride, if -1 then equal to n (default:-1)
|
||||
-e epsilon (default:1e-5)
|
||||
-save_rms save rms(invrms) or not. set to 1 in training case (default:0)
|
||||
-save_unquant save result before quant (default:0)
|
||||
-v cpu validation or not (default:1)
|
||||
-kname print kernel name or not (default:1)
|
||||
-prec_i input precision (default:fp16)
|
||||
-prec_o output precision, set auto will be the same as input (default:auto)
|
||||
-prec_sm output quant scale type, set auto will use fp32. used when fquant=1 (default:auto)
|
||||
-prec_sy output quant scale type, set auto will use fp32. used when fquant=1 or 2 (default:auto)
|
||||
-fadd fused-add, 0:no fused add, 1:preadd+store, 2:preadd only (default:0)
|
||||
-fquant fused-quant, 0:no, 1:smooth-dynamic-quant, 2:dynamic-quant (default:0)
|
||||
-warmup cold iter (default:5)
|
||||
-repeat hot iter (default:20)
|
||||
-s sensitive model mode, 0: for no specific model, 1: for T5-like model (default:0)
|
||||
-json 0: No Json, 1: Dump Results in Json format (default:0)
|
||||
-jsonfile json file name to dump results (default:rmsnorm2d_fwd.json)
|
||||
```
|
||||
|
||||
@@ -1,6 +1,7 @@
|
||||
#include "ck_tile/host.hpp"
|
||||
#include "rmsnorm2d_fwd.hpp"
|
||||
#include <cstring>
|
||||
#include "json_dump.hpp"
|
||||
|
||||
// different threshold for different dtype
|
||||
template <typename DataType>
|
||||
@@ -53,7 +54,9 @@ auto create_args(int argc, char* argv[])
|
||||
.insert("fquant", "0", "fused-quant, 0:no, 1:smooth-dynamic-quant, 2:dynamic-quant")
|
||||
.insert("warmup", "5", "cold iter")
|
||||
.insert("repeat", "20", "hot iter")
|
||||
.insert("s", "0", "sensitive model mode, 0: for no specific model, 1: for T5-like model");
|
||||
.insert("s", "0", "sensitive model mode, 0: for no specific model, 1: for T5-like model")
|
||||
.insert("json", "0", "0: No Json, 1: Dump Results in Json format")
|
||||
.insert("jsonfile", "rmsnorm2d_fwd.json", "json file name to dump results");
|
||||
|
||||
bool result = arg_parser.parse(argc, argv);
|
||||
return std::make_tuple(result, arg_parser);
|
||||
@@ -437,6 +440,23 @@ bool run(const ck_tile::ArgParser& arg_parser)
|
||||
std::cout << ", valid:" << (pass ? "y" : "n") << std::flush << std::endl;
|
||||
}
|
||||
|
||||
if(arg_parser.get_int("json") == 1)
|
||||
{
|
||||
dump_rmsnorm2d_fwd_json(arg_parser.get_str("jsonfile"),
|
||||
prec_str,
|
||||
m,
|
||||
n,
|
||||
x_stride,
|
||||
xr_stride,
|
||||
y_stride,
|
||||
yr_stride,
|
||||
use_model_sensitive_rmsnorm,
|
||||
ave_time,
|
||||
0,
|
||||
gb_per_sec,
|
||||
pass);
|
||||
}
|
||||
|
||||
return pass;
|
||||
}
|
||||
|
||||
|
||||
Reference in New Issue
Block a user