ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-14 02:02:46 +00:00
Code Issues Packages Projects Releases Wiki Activity

Overhaul to Reducton and its dependants (#237)

Browse Source 
* Tiny fix in dynamic_buffer.hpp to support vectorized AtomicAdd for double type

* Update to host layer and host reduction

* Merge and remove reduction kernels

* Merge and remove reduction device interfaces and update pooling device interface

* Merge and remove useless reduction device instances

* Update to reduction profiler and reduction ctests

* Update to reduction and pooling examples and add one reduction example

* Change to reduction examples to let them testable by ctest

* Add explicit pass checking for reduction and pooling examples

* Explicit assignment of tensor shapes in example reduce_blockwise_two_call

* Use atomic_add to repace atomicAdd and add atomic_add for double type

* Add reduce ctest support for double data type

* Replace to_int_vector() by using c++ std::vector::assign()

* Keep DeviceReduceThreadWise separated from DeviceReduceBlockWise

* Merge DeviceReduceBlockWise and DeviceReduceMultiBlockAtomicAdd into DeviceReduceMultiBlock

* Add GetAtomicOperationZeroValue() support for AtomicMax

* Tiny change to reduce example README.md

* Fix some tiny issues due to branch merging

* Revoke previous change in dynamic_buffer.hpp and add atomic_add for double2_t

* Add reduce multiblock_atomic_add instances for fp64 to verify vectorized atomic_add on fp64

* Renaming

* Clean the header includings in device_reduce instances header files

[ROCm/composable_kernel commit: 63eee2d999]
This commit is contained in:
Qianfeng
2022-05-25 01:19:12 +08:00
committed by GitHub
parent eb85e53cb6
commit 5bcaa25bb7
94 changed files with 2429 additions and 6785 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
script/test_reduce_no_index.sh
View File

@@ -15,6 +15,17 @@ bin/test_reduce_no_index -D 64,4,280,82 -R 1 0 2
bin/test_reduce_no_index -D 64,4,280,82 -R 2 0 2
bin/test_reduce_no_index -D 64,4,280,82 -R 3 0 2
## for float64
bin/test_reduce_no_index -D 64,4,280,82 -R 0,1,2,3 6 2
bin/test_reduce_no_index -D 64,4,280,82 -R 0,1,2 6 2
bin/test_reduce_no_index -D 64,4,280,82 -R 0,1,3 6 2
bin/test_reduce_no_index -D 64,4,280,82 -R 0,2,3 6 2
bin/test_reduce_no_index -D 64,4,280,82 -R 1,2,3 6 2
bin/test_reduce_no_index -D 64,4,280,82 -R 0 6 2
bin/test_reduce_no_index -D 64,4,280,82 -R 1 6 2
bin/test_reduce_no_index -D 64,4,280,82 -R 2 6 2
bin/test_reduce_no_index -D 64,4,280,82 -R 3 6 2
## for float16
bin/test_reduce_no_index -D 64,4,280,82 -R 0,1,2,3 1 2
bin/test_reduce_no_index -D 64,4,280,82 -R 0,1,2 1 2

11
script/test_reduce_with_index.sh
View File

@@ -15,6 +15,17 @@ bin/test_reduce_with_index -D 64,4,280,82 -R 1 0 2
bin/test_reduce_with_index -D 64,4,280,82 -R 2 0 2
bin/test_reduce_with_index -D 64,4,280,82 -R 3 0 2
## for float64
bin/test_reduce_with_index -D 64,4,280,82 -R 0,1,2,3 6 2
bin/test_reduce_with_index -D 64,4,280,82 -R 0,1,2 6 2
bin/test_reduce_with_index -D 64,4,280,82 -R 0,1,3 6 2
bin/test_reduce_with_index -D 64,4,280,82 -R 0,2,3 6 2
bin/test_reduce_with_index -D 64,4,280,82 -R 1,2,3 6 2
bin/test_reduce_with_index -D 64,4,280,82 -R 0 6 2
bin/test_reduce_with_index -D 64,4,280,82 -R 1 6 2
bin/test_reduce_with_index -D 64,4,280,82 -R 2 6 2
bin/test_reduce_with_index -D 64,4,280,82 -R 3 6 2
## for float16
bin/test_reduce_with_index -D 64,4,280,82 -R 0,1,2,3 1 2
bin/test_reduce_with_index -D 64,4,280,82 -R 0,1,2 1 2