Avoid graph breaks by disabling sourceless calls in instrument_w_nvtx #7081
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
This PR is a continuation of the effort to improve Deepspeed performance when using PyTorch compile. The instrument_w_nvtx decorator is used to instrument code with NVIDIA Tools Extension (NVTX) markers for profiling and visualizing code execution on GPUs. Along with executing the function itself, instrument_w_nvtx makes calls to nvtx.range_push and nvtx.range_pop which can't be traced by Dynamo. That's why this decorator causes a graph break. The impact on performnace can be significant due to numerous uses of the decorator throughout the code. We propose a simple solution: Don't invoke the sourceless functions when torch is compiling. Signed-off-by: Max Kovalenko <[email protected]>
tjruwase
previously approved these changes
Feb 26, 2025
tjruwase
reviewed
Feb 26, 2025
tjruwase
dismissed
their stale review
Requested usage of DeepSpeed utility to address CI failures.
tjruwase
reviewed
Feb 26, 2025
Signed-off-by: Max Kovalenko <[email protected]>
tjruwase
approved these changes
Feb 27, 2025
Signed-off-by: Max Kovalenko <[email protected]>
github-merge-queue
bot
removed this pull request from the merge queue due to failed status checks
ys950902
pushed a commit
to ys950902/DeepSpeed
that referenced
this pull request
Mar 6, 2025
…deepspeedai#7081) This PR is a continuation of the efforts to improve Deepspeed performance when using PyTorch compile. The `instrument_w_nvtx` decorator is used to instrument code with NVIDIA Tools Extension (NVTX) markers for profiling and visualizing code execution on GPUs. Along with executing the function itself, `instrument_w_nvtx` makes calls to `nvtx.range_push` and `nvtx.range_pop` which can't be traced by Dynamo. That's why this decorator causes a graph break. The impact on performance can be significant due to numerous uses of the decorator throughout the code. We propose a simple solution: Don't invoke the sourceless functions when torch is compiling. --------- Signed-off-by: Max Kovalenko <[email protected]> Co-authored-by: Logan Adams <[email protected]> Signed-off-by: yisheng <[email protected]>
raza-sikander
pushed a commit
to raza-sikander/DeepSpeed
that referenced
this pull request
Mar 6, 2025
…deepspeedai#7081) This PR is a continuation of the efforts to improve Deepspeed performance when using PyTorch compile. The `instrument_w_nvtx` decorator is used to instrument code with NVIDIA Tools Extension (NVTX) markers for profiling and visualizing code execution on GPUs. Along with executing the function itself, `instrument_w_nvtx` makes calls to `nvtx.range_push` and `nvtx.range_pop` which can't be traced by Dynamo. That's why this decorator causes a graph break. The impact on performance can be significant due to numerous uses of the decorator throughout the code. We propose a simple solution: Don't invoke the sourceless functions when torch is compiling. --------- Signed-off-by: Max Kovalenko <[email protected]> Co-authored-by: Logan Adams <[email protected]>
raza-sikander
pushed a commit
to raza-sikander/DeepSpeed
that referenced
this pull request
Mar 6, 2025
…deepspeedai#7081) This PR is a continuation of the efforts to improve Deepspeed performance when using PyTorch compile. The `instrument_w_nvtx` decorator is used to instrument code with NVIDIA Tools Extension (NVTX) markers for profiling and visualizing code execution on GPUs. Along with executing the function itself, `instrument_w_nvtx` makes calls to `nvtx.range_push` and `nvtx.range_pop` which can't be traced by Dynamo. That's why this decorator causes a graph break. The impact on performance can be significant due to numerous uses of the decorator throughout the code. We propose a simple solution: Don't invoke the sourceless functions when torch is compiling. --------- Signed-off-by: Max Kovalenko <[email protected]> Co-authored-by: Logan Adams <[email protected]> Signed-off-by: Shaik Raza Sikander <[email protected]>
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.
This PR is a continuation of the efforts to improve Deepspeed performance when using PyTorch compile.
The
instrument_w_nvtxdecorator is used to instrument code with NVIDIA Tools Extension (NVTX) markers for profiling and visualizing code execution on GPUs.Along with executing the function itself,
instrument_w_nvtxmakes calls tonvtx.range_pushandnvtx.range_popwhich can't be traced by Dynamo.That's why this decorator causes a graph break.
The impact on performance can be significant due to numerous uses of the decorator throughout the code.
We propose a simple solution: Don't invoke the sourceless functions when torch is compiling.