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

merge CUDA and ROCm codes in op #2847

Merged
merged 25 commits into from
Sep 26, 2023
Merged

merge CUDA and ROCm codes in op #2847

merged 25 commits into from
Sep 26, 2023

Conversation

njzjz
Copy link
Member

@njzjz njzjz commented Sep 20, 2023

No description provided.

njzjz added 16 commits September 19, 2023 15:43
@njzjz
use the same macro between CUDA and ROCm
2fa777f 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
checkpoint
3f8d05d 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
gpuMemset
6ee416e 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
checkpoint
79b31ca 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
merge locate_xx_se_a and locate_xx_se_r: the same function
972919f 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
merge cmake files
797dad2 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
renaming
6ef697f 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
Revert "merge locate_xx_se_a and locate_xx_se_r: the same function"
c32407c 
This reverts commit 972919f.
@njzjz
fix typo
a1cdc20 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
fix trap
f7bfadb 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
typo
245964c 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
fix tabulate_fusion_se_a_fifth_order_polynomial for rocm
15f5729 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
use GPU_DYNAMIC_SHARED_MEM_DECL
f36ecc6 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
clean never used shared memory decalare
074bbdc 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
fix submodule
4edce5f 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz
merge CUDA and ROCm codes in op
30b61e7 
Signed-off-by: Jinzhe Zeng <[email protected]>
@njzjz njzjz mentioned this pull request Sep 20, 2023
Closed
7 tasks
@github-actions github-actions bot added the OP label Sep 20, 2023
@codecov
Copy link

codecov bot commented Sep 20, 2023 •
edited
Loading

Codecov Report

Patch coverage: 100.00% and project coverage change: +0.43% 🎉

Comparison is base (c2c6476) 75.38% compared to head (b627c86) 75.81%.

Additional details and impacted files 
@@            Coverage Diff             @@
##            devel    #2847      +/-   ##
==========================================
+ Coverage   75.38%   75.81%   +0.43%     
==========================================
  Files         242      243       +1     
  Lines       24424    24713     +289     
  Branches     1580     1617      +37     
==========================================
+ Hits        18411    18736     +325     
+ Misses       5082     5015      -67     
- Partials      931      962      +31
Files Changed Coverage Δ
source/lib/include/gpu_cuda.h 70.00% <ø> (ø)
source/op/gelu_multi_device.cc 28.33% <ø> (+3.33%) ⬆️
source/op/prod_env_mat_multi_device_nvnmd.cc 69.44% <ø> (+0.42%) ⬆️
source/op/prod_force_grad_multi_device.cc 69.60% <ø> (+2.60%) ⬆️
source/op/prod_force_multi_device.cc 65.93% <ø> (+1.56%) ⬆️
source/op/prod_virial_grad_multi_device.cc 69.04% <ø> (+2.11%) ⬆️
source/op/prod_virial_multi_device.cc 72.41% <ø> (+0.98%) ⬆️
source/op/tabulate_multi_device.cc 61.64% <ø> (+1.58%) ⬆️
source/op/unaggregated_grad.cc 67.51% <ø> (ø)
source/op/prod_env_mat_multi_device.cc 72.81% <100.00%> (+12.38%) ⬆️

... and 41 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

@njzjz njzjz marked this pull request as draft September 20, 2023 01:59
@njzjz
Copy link
Member Author

njzjz commented Sep 20, 2023

The OP codes contain cudaMemcpy, so we must wait for #2844.

@github-actions github-actions bot added the Core label Sep 21, 2023
@njzjz njzjz added the Test CUDA Trigger test CUDA workflow label Sep 21, 2023
@github-actions github-actions bot removed the Test CUDA Trigger test CUDA workflow label Sep 21, 2023
@njzjz njzjz marked this pull request as ready for review September 22, 2023 14:01
@njzjz njzjz added CUDA ROCM Docs Test CUDA Trigger test CUDA workflow labels Sep 22, 2023
@github-actions github-actions bot removed the Test CUDA Trigger test CUDA workflow label Sep 22, 2023
@njzjz njzjz requested a review from wanghan-iapcm September 22, 2023 16:57
@njzjz njzjz linked an issue Sep 26, 2023 that may be closed by this pull request
[Feature Request] Merge CUDA and ROCm codes #2838
Closed
7 tasks
@wanghan-iapcm wanghan-iapcm merged commit 938cea2 into deepmodeling:devel Sep 26, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@wanghan-iapcm wanghan-iapcm wanghan-iapcm approved these changes

Assignees
No one assigned
Labels
Core OP
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

[Feature Request] Merge CUDA and ROCm codes
2 participants
@njzjz @wanghan-iapcm