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 flag to dump mlir modules to mxrs #3182

Merged
merged 14 commits into from
Oct 21, 2024
Merged

Add flag to dump mlir modules to mxrs #3182

merged 14 commits into from
Oct 21, 2024

Conversation

pfultz2
Copy link
Collaborator

@pfultz2 pfultz2 commented Jun 14, 2024

No description provided.

@pfultz2 pfultz2 requested a review from causten as a code owner June 14, 2024 01:58
@codecov Codecov
Copy link

codecov bot commented Jun 14, 2024 •
edited
Loading

Codecov Report

Attention: Patch coverage is 0% with 13 lines in your changes missing coverage. Please review.

Project coverage is 92.16%. Comparing base (275f854) to head (fff0dc9).
Report is 12 commits behind head on develop.

Files with missing lines Patch % Lines
src/shape.cpp 0.00% 9 Missing ⚠️
src/program.cpp 0.00% 4 Missing ⚠️
Additional details and impacted files 
@@             Coverage Diff             @@
##           develop    #3182      +/-   ##
===========================================
- Coverage    92.17%   92.16%   -0.02%     
===========================================
  Files          512      512              
  Lines        21391    21394       +3     
===========================================
  Hits         19718    19718              
- Misses        1673     1676       +3

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

umangyadav
umangyadav reviewed Jun 17, 2024
View reviewed changes
src/targets/gpu/jit/mlir.cpp
@@ -198,8 +200,14 @@ struct mlir_compiler : compiler<mlir_compiler>
const operation&,
bool exhaustive) const
{
static const auto mxr_loc = string_value_of(MIGRAPHX_MLIR_DUMP_TO_MXR{});
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

add to ENV var document.

Also what's the usefulness of having MXRs for each MLIR module ?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added docs.

@pfultz2 pfultz2 requested a review from a team as a code owner August 8, 2024 22:42
@pfultz2 pfultz2 mentioned this pull request Oct 16, 2024
Closed
@migraphx-bot
Copy link
Collaborator

Test Batch Rate new
fff0dc		 Rate old
ddc4c0		 Diff Compare
torchvision-resnet50 64 3,261.58 3,261.96 -0.01% ✅
torchvision-resnet50_fp16 64 6,999.88 6,986.50 0.19% ✅
torchvision-densenet121 32 2,435.79 2,435.05 0.03% ✅
torchvision-densenet121_fp16 32 4,088.41 4,098.64 -0.25% ✅
torchvision-inceptionv3 32 1,637.98 1,637.72 0.02% ✅
torchvision-inceptionv3_fp16 32 2,763.46 2,760.21 0.12% ✅
cadene-inceptionv4 16 776.37 776.83 -0.06% ✅
cadene-resnext64x4 16 807.12 812.01 -0.60% ✅
slim-mobilenet 64 7,535.95 7,539.31 -0.04% ✅
slim-nasnetalarge 64 211.52 211.58 -0.03% ✅
slim-resnet50v2 64 3,505.14 3,507.45 -0.07% ✅
bert-mrpc-onnx 8 1,146.38 1,148.38 -0.17% ✅
bert-mrpc-tf 1 469.29 462.39 1.49% ✅
pytorch-examples-wlang-gru 1 426.36 416.01 2.49% ✅
pytorch-examples-wlang-lstm 1 383.45 387.25 -0.98% ✅
torchvision-resnet50_1 1 788.74 799.30 -1.32% ✅
cadene-dpn92_1 1 424.20 413.07 2.69% ✅
cadene-resnext101_1 1 383.67 384.54 -0.23% ✅
onnx-taau-downsample 1 342.74 343.19 -0.13% ✅
dlrm-criteoterabyte 1 33.37 33.33 0.12% ✅
dlrm-criteoterabyte_fp16 1 52.70 52.75 -0.10% ✅
agentmodel 1 8,153.22 8,254.64 -1.23% ✅
unet_fp16 2 58.97 58.74 0.39% ✅
resnet50v1_fp16 1 964.71 942.04 2.41% ✅
resnet50v1_int8 1 989.95 993.52 -0.36% ✅
bert_base_cased_fp16 64 1,170.46 1,171.05 -0.05% ✅
bert_large_uncased_fp16 32 363.49 363.48 0.00% ✅
bert_large_fp16 1 198.82 200.70 -0.94% ✅
distilgpt2_fp16 16 2,203.34 2,201.26 0.09% ✅
yolov5s 1 538.02 537.45 0.11% ✅
tinyllama 1 43.56 43.68 -0.28% ✅
vicuna-fastchat 1 176.45 175.99 0.27% ✅
whisper-tiny-encoder 1 418.43 418.69 -0.06% ✅
whisper-tiny-decoder 1 427.33 428.55 -0.28% ✅

This build is OK for merge ✅

@migraphx-bot
Copy link
Collaborator


     ✅ bert-mrpc-onnx: PASSED: MIGraphX meets tolerance

     ✅ bert-mrpc-tf: PASSED: MIGraphX meets tolerance

     ✅ pytorch-examples-wlang-gru: PASSED: MIGraphX meets tolerance

     ✅ pytorch-examples-wlang-lstm: PASSED: MIGraphX meets tolerance

     ✅ torchvision-resnet50_1: PASSED: MIGraphX meets tolerance

     ✅ cadene-dpn92_1: PASSED: MIGraphX meets tolerance

     ✅ cadene-resnext101_1: PASSED: MIGraphX meets tolerance

     ✅ dlrm-criteoterabyte: PASSED: MIGraphX meets tolerance

     ✅ agentmodel: PASSED: MIGraphX meets tolerance

     ✅ unet: PASSED: MIGraphX meets tolerance

     ✅ resnet50v1: PASSED: MIGraphX meets tolerance

     ✅ bert_base_cased_fp16: PASSED: MIGraphX meets tolerance

🔴bert_large_uncased_fp16: FAILED: MIGraphX is not within tolerance - check verbose output


     ✅ bert_large: PASSED: MIGraphX meets tolerance

     ✅ yolov5s: PASSED: MIGraphX meets tolerance

     ✅ tinyllama: PASSED: MIGraphX meets tolerance

     ✅ vicuna-fastchat: PASSED: MIGraphX meets tolerance

     ✅ whisper-tiny-encoder: PASSED: MIGraphX meets tolerance

     ✅ whisper-tiny-decoder: PASSED: MIGraphX meets tolerance

     ✅ distilgpt2_fp16: PASSED: MIGraphX meets tolerance

@causten causten merged commit c3a5367 into develop Oct 21, 2024
37 of 44 checks passed
@causten causten deleted the dump-mlir branch October 21, 2024 19:31
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@shivadbhavsar shivadbhavsar shivadbhavsar approved these changes

@umangyadav umangyadav umangyadav approved these changes

@causten causten Awaiting requested review from causten causten is a code owner

Assignees

@pfultz2 pfultz2

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
5 participants
@pfultz2 @migraphx-bot @umangyadav @shivadbhavsar @causten