Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add riscv float32 gemm #4903

Merged
merged 46 commits into from Oct 20, 2023
Merged

Add riscv float32 gemm #4903

Show file tree
Hide file tree
Changes from 7 commits
Commits
Show all changes
46 commits
Select commit Hold shift + click to select a range
c4be138
add defination of gemm_riscv
Xinyu302 Aug 2, 2023
8d0b0c7
add pack_A_tile pack_B_tile in gemm_riscv
Xinyu302 Aug 19, 2023
d8255f9
transpose and gemm
Xinyu302 Aug 19, 2023
523a548
compile right, before add pipeine
Xinyu302 Aug 19, 2023
b674cf1
finish gemm_riscv but has bug
Xinyu302 Aug 19, 2023
8cb184f
add create_pipeline function in gemm_riscv
Xinyu302 Aug 21, 2023
fcb974f
fix bug in transpose_unpack_output_tile
Xinyu302 Aug 21, 2023
28dcb7e
add #if __riscv_vector to support device which cannot run RISCV-V
Xinyu302 Sep 13, 2023
2a32961
add C906 macro, in other case, now use naive implementation
Xinyu302 Sep 13, 2023
45aaf1f
modify transpose kernel
Xinyu302 Sep 13, 2023
f98d0e7
change C906 macro location
Xinyu302 Sep 13, 2023
c373531
modify store_float32_v2, store_float_v4
Xinyu302 Sep 13, 2023
a5c2a90
delete useless functions
Xinyu302 Sep 13, 2023
9424f85
delete annotations
Xinyu302 Sep 13, 2023
71f382b
Add #include cpu.h
Xinyu302 Sep 14, 2023
b0022d6
add static for pack_A_tile
Xinyu302 Sep 14, 2023
366d6cb
delete annotation
Xinyu302 Sep 14, 2023
5d6a55f
replace vlseg2e32_v_f32m1x2 with vlseg2e32_v_f32m1
Xinyu302 Sep 15, 2023
fca6d4b
fix small bugs
Xinyu302 Sep 15, 2023
31db10d
remove C906 macro
Xinyu302 Sep 15, 2023
c4a4580
add nT = 0
Xinyu302 Sep 15, 2023
d40c309
fix stupid error
Xinyu302 Sep 15, 2023
ad1196f
Merge branch 'fix-vlseg-bug' into add-riscv-gemm, modify PC<=>outptr
Xinyu302 Sep 15, 2023
882b502
change float32_t to float
Xinyu302 Sep 15, 2023
364d5d9
add matmul_riscv.cpp
Xinyu302 Sep 16, 2023
3f4b415
add gemm model
Xinyu302 Sep 16, 2023
013eb15
add model
Xinyu302 Sep 16, 2023
6835ad5
add timing code
Xinyu302 Sep 17, 2023
951ec58
modify some intrinsic usage
Xinyu302 Sep 17, 2023
e532b61
modify vfmaq_laneq_f32_riscv
Xinyu302 Sep 17, 2023
90de9dd
replace vdupq_n_f32_riscv func call with vfmv_v_f_f32m1 and improve a…
Xinyu302 Sep 17, 2023
0ff2893
fix no compute bug
Xinyu302 Sep 17, 2023
a78c5b8
add time record to all functions
Xinyu302 Sep 17, 2023
eafc7b7
use vfmadd_vf_f32m1 and offset
Xinyu302 Sep 17, 2023
086f4ef
delete time code
Xinyu302 Sep 17, 2023
762ce79
delete useless include
Xinyu302 Sep 17, 2023
fd9d83c
delete useless printf
Xinyu302 Sep 17, 2023
00f97c9
apply code-format changes
Xinyu302 Sep 17, 2023
9276321
Merge pull request #1 from Xinyu302/gemm-time-test
Xinyu302 Sep 17, 2023
02feba4
delete useless examples
Xinyu302 Sep 17, 2023
eebe280
delte useless function in riscv_usability.h
Xinyu302 Sep 18, 2023
578a8e8
LAYER_GEMM_RISCV_H
Xinyu302 Sep 20, 2023
c48b8c4
delete define VL
Xinyu302 Sep 20, 2023
c9dc401
add annotation
Xinyu302 Sep 20, 2023
6ccb081
apply code-format changes
Xinyu302 Sep 20, 2023
0f67295
delete riscv_zfh comment
Xinyu302 Sep 22, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view