[RELAND] Force synced KJT to trace unbacked SymInt (#108960) #109216
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/109216
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 93c783b with merge base afad0d0 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
This pull request was exported from Phabricator. Differential Revision: D49236757 |
ezyang
changed the title
ezyang
commented
Sep 13, 2023
| # they (1) generalize immediately and (2) unsoundly never compare equal to | ||
| # 0/1. This is not on by default as AOTAutograd/Inductor cannot currently | ||
| # compile this code; however, this can be useful for export. | ||
| force_unspec_int_unbacked_size_like_on_torchrec_kjt = False |
There was a problem hiding this comment.
The delta is we are now flag guarded.
ezyang
added a commit
to ezyang/pytorch
that referenced
this pull request
Sep 13, 2023
Summary: The basic concept behind this diff is to modify Dynamo's tracing behavior when it encounters a KeyedJaggedTensor that is synced (aka has `_length_per_key` and `_offset_per_key` populated). These fields are lists of integers; ordinarily, Dynamo will optimistically try to specialize on integers, however, for KJTs, we know that these integers will definitely vary from run-to-run. Furthermore, ordinarily, we would also specialize these integers if they are 0/1, but we will frequently expect features in KJTs to be 0/1. The fix is to detect KJTs and treat these integers as *unbacked integers*. This is NOT a universally sound optimization: when treating these integers as unbacked, we never report them as equal to zero or one. In return, we always generate graphs that generalize no matter the length of values on features. This is enough to trace through APS sparse arch, torchrec_dlrm and some small split-cat examples. The special integer behavior is triggered by a dynamically scoped `force_unspec_int_unbacked_size_like` variable on TracingContext, which we trigger when we wrap a KJT. There probably are other ways to do this, but this was simple and worked. Test Plan: ``` buck2 test mode/dev-nosan //pytorch/benchmark/fb/test_gpu:run_test_gpu ``` from aakhundov 1. first build feed_lower_benchmark: ``` buck2 build --show-output mode/opt -c python.package_style=inplace -c fbcode.enable_gpu_sections=true -c fbcode.platform=platform010 -c fbcode.split-dwarf=true hpc/new/models/feed/benchmark:feed_lower_benchmark ``` 2. then run the lowering of the model with it: ``` TORCHINDUCTOR_MAX_AUTOTUNE=1 TORCHINDUCTOR_UNIQUE_KERNEL_NAMES=1 TORCH_LOGS="output_code,graph_code" TORCH_COMPILE_DEBUG=1 ../buck-out/v2/gen/fbcode/79c6b019ee0f9469/hpc/new/models/feed/benchmark/__feed_lower_benchmark__/feed_lower_benchmark.par --load=manifold://ig_inference_model/tree/user/facebook/fblearner/predictor/960999465/60/gpu_lowering/input.predictor --skip-trt --skip-ait --sync-mode=0 --enable-aot-inductor --lower-presets="ig_stories" --gpu-trace ``` cf https://docs.google.com/document/d/1yD30xYrdmM8r2HTdmXnZTg0-MHVexfVrAa0294m1AUE/edit?pli=1#heading=h.qiv3fp7e6zg0 From torchrec: https://www.internalfb.com/intern/wiki/Torchrec/Development/Testing_production_models/ From ge0405 baseline (without your diff): f477293168 your diff: f477292363 ``` buck2 test //caffe2/test/dynamo:test_dynamo_torchrec buck2 run 'fbcode//mode/opt' fbcode//pytorch/benchmark/fb/test_gpu:run_test_gpu -- 'pytorch.benchmark.fb.test_gpu.test_gpu.TestBenchmarkFbGpu.test_train_blue_reels_vdd_v3_inductor_speedup' ``` Differential Revision: D49236757
ezyang
force-pushed
the
export-D49236757
branch
from
59fb225
to
003af11
Compare
|
This pull request was exported from Phabricator. Differential Revision: D49236757 |
voznesenskym
approved these changes
Sep 14, 2023
ezyang
force-pushed
the
export-D49236757
branch
from
003af11
to
67f4d18
Compare
|
This pull request was exported from Phabricator. Differential Revision: D49236757 |
ezyang
added a commit
to ezyang/pytorch
that referenced
this pull request
Sep 15, 2023
Summary: This is a re-land of pytorch#108960 The basic concept behind this diff is to modify Dynamo's tracing behavior when it encounters a KeyedJaggedTensor that is synced (aka has `_length_per_key` and `_offset_per_key` populated). These fields are lists of integers; ordinarily, Dynamo will optimistically try to specialize on integers, however, for KJTs, we know that these integers will definitely vary from run-to-run. Furthermore, ordinarily, we would also specialize these integers if they are 0/1, but we will frequently expect features in KJTs to be 0/1. The fix is to detect KJTs and treat these integers as *unbacked integers*. This is NOT a universally sound optimization: when treating these integers as unbacked, we never report them as equal to zero or one. In return, we always generate graphs that generalize no matter the length of values on features. This is enough to trace through APS sparse arch, torchrec_dlrm and some small split-cat examples. The special integer behavior is triggered by a dynamically scoped `force_unspec_int_unbacked_size_like` variable on TracingContext, which we trigger when we wrap a KJT. There probably are other ways to do this, but this was simple and worked. Test Plan: ``` buck2 test mode/dev-nosan //pytorch/benchmark/fb/test_gpu:run_test_gpu ``` from aakhundov 1. first build feed_lower_benchmark: ``` buck2 build --show-output mode/opt -c python.package_style=inplace -c fbcode.enable_gpu_sections=true -c fbcode.platform=platform010 -c fbcode.split-dwarf=true hpc/new/models/feed/benchmark:feed_lower_benchmark ``` 2. then run the lowering of the model with it: ``` TORCHINDUCTOR_MAX_AUTOTUNE=1 TORCHINDUCTOR_UNIQUE_KERNEL_NAMES=1 TORCH_LOGS="output_code,graph_code" TORCH_COMPILE_DEBUG=1 ../buck-out/v2/gen/fbcode/79c6b019ee0f9469/hpc/new/models/feed/benchmark/__feed_lower_benchmark__/feed_lower_benchmark.par --load=manifold://ig_inference_model/tree/user/facebook/fblearner/predictor/960999465/60/gpu_lowering/input.predictor --skip-trt --skip-ait --sync-mode=0 --enable-aot-inductor --lower-presets="ig_stories" --gpu-trace ``` cf https://docs.google.com/document/d/1yD30xYrdmM8r2HTdmXnZTg0-MHVexfVrAa0294m1AUE/edit?pli=1#heading=h.qiv3fp7e6zg0 From torchrec: https://www.internalfb.com/intern/wiki/Torchrec/Development/Testing_production_models/ From ge0405 baseline (without your diff): f477293168 your diff: f477292363 ``` buck2 test //caffe2/test/dynamo:test_dynamo_torchrec buck2 run 'fbcode//mode/opt' fbcode//pytorch/benchmark/fb/test_gpu:run_test_gpu -- 'pytorch.benchmark.fb.test_gpu.test_gpu.TestBenchmarkFbGpu.test_train_blue_reels_vdd_v3_inductor_speedup' ``` Reviewed By: voznesenskym Differential Revision: D49236757
Summary: This is a re-land of pytorch#108960 The basic concept behind this diff is to modify Dynamo's tracing behavior when it encounters a KeyedJaggedTensor that is synced (aka has `_length_per_key` and `_offset_per_key` populated). These fields are lists of integers; ordinarily, Dynamo will optimistically try to specialize on integers, however, for KJTs, we know that these integers will definitely vary from run-to-run. Furthermore, ordinarily, we would also specialize these integers if they are 0/1, but we will frequently expect features in KJTs to be 0/1. The fix is to detect KJTs and treat these integers as *unbacked integers*. This is NOT a universally sound optimization: when treating these integers as unbacked, we never report them as equal to zero or one. In return, we always generate graphs that generalize no matter the length of values on features. This is enough to trace through APS sparse arch, torchrec_dlrm and some small split-cat examples. The special integer behavior is triggered by a dynamically scoped `force_unspec_int_unbacked_size_like` variable on TracingContext, which we trigger when we wrap a KJT. There probably are other ways to do this, but this was simple and worked. Test Plan: ``` buck2 test mode/dev-nosan //pytorch/benchmark/fb/test_gpu:run_test_gpu ``` from aakhundov 1. first build feed_lower_benchmark: ``` buck2 build --show-output mode/opt -c python.package_style=inplace -c fbcode.enable_gpu_sections=true -c fbcode.platform=platform010 -c fbcode.split-dwarf=true hpc/new/models/feed/benchmark:feed_lower_benchmark ``` 2. then run the lowering of the model with it: ``` TORCHINDUCTOR_MAX_AUTOTUNE=1 TORCHINDUCTOR_UNIQUE_KERNEL_NAMES=1 TORCH_LOGS="output_code,graph_code" TORCH_COMPILE_DEBUG=1 ../buck-out/v2/gen/fbcode/79c6b019ee0f9469/hpc/new/models/feed/benchmark/__feed_lower_benchmark__/feed_lower_benchmark.par --load=manifold://ig_inference_model/tree/user/facebook/fblearner/predictor/960999465/60/gpu_lowering/input.predictor --skip-trt --skip-ait --sync-mode=0 --enable-aot-inductor --lower-presets="ig_stories" --gpu-trace ``` cf https://docs.google.com/document/d/1yD30xYrdmM8r2HTdmXnZTg0-MHVexfVrAa0294m1AUE/edit?pli=1#heading=h.qiv3fp7e6zg0 From torchrec: https://www.internalfb.com/intern/wiki/Torchrec/Development/Testing_production_models/ From ge0405 baseline (without your diff): f477293168 your diff: f477292363 ``` buck2 test //caffe2/test/dynamo:test_dynamo_torchrec buck2 run 'fbcode//mode/opt' fbcode//pytorch/benchmark/fb/test_gpu:run_test_gpu -- 'pytorch.benchmark.fb.test_gpu.test_gpu.TestBenchmarkFbGpu.test_train_blue_reels_vdd_v3_inductor_speedup' ``` Reviewed By: voznesenskym Differential Revision: D49236757
|
This pull request was exported from Phabricator. Differential Revision: D49236757 |
ezyang
force-pushed
the
export-D49236757
branch
from
67f4d18
to
93c783b
Compare
|
This pull request was exported from Phabricator. Differential Revision: D49236757 |
|
@pytorchbot merge (Initiating merge automatically since Phabricator Diff has merged) |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
The basic concept behind this diff is to modify Dynamo's tracing behavior when it encounters a KeyedJaggedTensor that is synced (aka has
_length_per_keyand_offset_per_keypopulated). These fields are lists of integers; ordinarily, Dynamo will optimistically try to specialize on integers, however, for KJTs, we know that these integers will definitely vary from run-to-run. Furthermore, ordinarily, we would also specialize these integers if they are 0/1, but we will frequently expect features in KJTs to be 0/1.The fix is to detect KJTs and treat these integers as unbacked integers. This is NOT a universally sound optimization: when treating these integers as unbacked, we never report them as equal to zero or one. In return, we always generate graphs that generalize no matter the length of values on features. This is enough to trace through APS sparse arch, torchrec_dlrm and some small split-cat examples.
The special integer behavior is triggered by a dynamically scoped
force_unspec_int_unbacked_size_likevariable on TracingContext, which we trigger when we wrap a KJT. There probably are other ways to do this, but this was simple and worked.Test Plan:
from aakhundov
cf https://docs.google.com/document/d/1yD30xYrdmM8r2HTdmXnZTg0-MHVexfVrAa0294m1AUE/edit?pli=1#heading=h.qiv3fp7e6zg0
From torchrec: https://www.internalfb.com/intern/wiki/Torchrec/Development/Testing_production_models/
From ge0405
baseline (without your diff): f477293168
your diff: f477292363
Differential Revision: D49236757
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @Xia-Weiwen @wenzhe-nrv @jiayisunx @chenyang78 @aakhundov @kadeng