mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-19 22:39:03 +00:00

Files

myamlak f03a1738d9 Resolution of issue #153 : Add compiler warning on comparing int and size_t (#212 )

* Turning compare warnings on

* Cleaning part I

* Cleaning part II

* Explicit static_cast to ck::type_convert

* Resolving large tensor size issue.

* format

* revert change to tensor descriptor; promote lementSpaceSize to 64bit

* use integer value for GEMM test

* Review remarks

* Review remarks + issues with (un)signed arithmetic

* Format fix

* Format

* Clang-format.

* fix 2gb limit issue

Co-authored-by: Chao Liu <chao.liu2@amd.com>
Co-authored-by: Adam Osewski <aosewski@amd.com>

2022-05-09 15:06:49 -05:00

CMakeLists.txt

Reorganize files, Part 1 (#119 )

2022-03-08 21:46:36 -06:00

pool2d_fwd.cpp

Resolution of issue #153 : Add compiler warning on comparing int and size_t (#212 )

2022-05-09 15:06:49 -05:00

README.md

Compile for gfx908 and gfx90a (#130 )

2022-03-31 12:33:34 -05:00

README.md

Instructions for `example_pool2d_fwd` Example

Run `example_pool2d_fwd`

#arg1: verification (0=no, 1=yes)
#arg2: initialization (0=no init, 1=single integer value, 2=scope integer value, 3=decimal value)
#arg3: run kernel # of times (>1)
#arg4 to 15: N, C, Y, X, Hi, Wi, Sy, Sx, LeftPy, LeftPx, RightPy, RightPx
./bin/example_pool2d_fwd 1 1 10

Result

in_n_c_hi_wi: dim 4, lengths {128, 192, 71, 71}, strides {967872, 1, 13632, 192}
out_n_c_ho_wo: dim 4, lengths {128, 192, 36, 36}, strides {248832, 1, 6912, 192}
launch_and_time_kernel: grid_dim {124416, 1, 1}, block_dim {64, 1, 1} 
Warm up
Start running 10 times...
Perf: 0.415453 ms, 1.37996 TFlops, 749.726 GB/s
error: 0
max_diff: 0, 1, 1

README.md

Instructions for example_pool2d_fwd Example

Run example_pool2d_fwd

Instructions for `example_pool2d_fwd` Example

Run `example_pool2d_fwd`