Skip to content
Closed
Show file tree
Hide file tree
Changes from 18 commits
Commits
Show all changes
33 commits
Select commit Hold shift + click to select a range
d6e053a
Part of gemm + softmax, Add gemm + reduceMax
rocking5566 Apr 3, 2022
cbbc7e5
Refine the comment
rocking5566 Apr 6, 2022
3e811cc
Add device op for elementwise 2d
rocking5566 Apr 10, 2022
0277c89
Merge branch 'develop' into gemm_softmax
rocking5566 Apr 10, 2022
6818b58
Fix compile error
rocking5566 Apr 10, 2022
e3a09b5
Add gridwise_elementwise_2d api
rocking5566 Apr 11, 2022
cb1c473
Merge remote-tracking branch 'origin/develop' into gemm_softmax
rocking5566 Apr 11, 2022
a760a73
A kernel of elementwise_2d (except global store)
rocking5566 Apr 12, 2022
c8b4ac2
Add global write
rocking5566 Apr 13, 2022
f2540aa
Add exponential
rocking5566 Apr 13, 2022
30348da
[What] Refine naming
rocking5566 Apr 13, 2022
b05a594
Add reduce sum for denominator of softmax
rocking5566 Apr 13, 2022
6a781e5
Add broadcast div, the final step of softmax
rocking5566 Apr 13, 2022
dba65b1
Rewrite the gridwise_elementwise_
rocking5566 Apr 14, 2022
fe65950
Add verication of softmax
rocking5566 Apr 15, 2022
e83b22e
[What] Use half_float::half instead of ck::half_t for host reduction
rocking5566 Apr 18, 2022
21802fd
[What] Sync input of each host kernel and device kernel
rocking5566 Apr 18, 2022
c16f789
Merge remote-tracking branch 'origin/develop' into gemm_softmax
rocking5566 Apr 18, 2022
cf32669
[What] Use F32 as the acc of reduce sum
rocking5566 Apr 20, 2022
0f421d6
[What] Add ComputeDataType to the eltwise kernel
rocking5566 Apr 20, 2022
5fa209a
Add padding
rocking5566 Apr 20, 2022
0e6bf34
Rename elementwise p[ to binary elementwise
rocking5566 Apr 20, 2022
d7112d3
Fix the padding
rocking5566 Apr 20, 2022
88d621a
Merge remote-tracking branch 'origin/develop' into gemm_softmax
rocking5566 Apr 21, 2022
5d36f7a
Rewrite the elementwise operation.
rocking5566 Apr 21, 2022
680cfaa
Fix the meaning of broadcast dim parameter
rocking5566 Apr 22, 2022
a41f548
1. Fix coding style
rocking5566 Apr 25, 2022
f919809
Move threadPerBlock to argument
rocking5566 Apr 26, 2022
976815e
Prevent compile error when user pass rvalue, eg {3, 4}
rocking5566 Apr 28, 2022
ea09fd3
Fix typo
rocking5566 Apr 28, 2022
bfc8076
[What] Fix data type for host reduction
rocking5566 Apr 29, 2022
d92fb7e
Merge commit 'a3c910ac6cdd0c5b724449af312255abe5b531e1' into gemm_sof…
rocking5566 May 9, 2022
b6fe118
Fix typo
rocking5566 May 9, 2022
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions example/19_gemm_softmax/CMakeLists.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
add_example_executable(example_gemm_softmax_xdl_fp16 gemm_softmax_xdl_fp16.cpp)
Loading