[Submodule] Turning flash-attention integration into 3rd party submod#144120
Closed
drisspg wants to merge 22 commits intogh/drisspg/111/basefrom
Closed
[Submodule] Turning flash-attention integration into 3rd party submod#144120drisspg wants to merge 22 commits intogh/drisspg/111/basefrom
drisspg wants to merge 22 commits intogh/drisspg/111/basefrom
Conversation
Closed
drisspg
commented
Jan 3, 2025
Skylion007
reviewed
Jan 5, 2025
Contributor
Author
|
@drisspg has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
pytorchmergebot
pushed a commit
that referenced
this pull request
Jan 24, 2025
…145502) # Context Prototyped here: #144120, we are going to make flash-attention a 3rd party submodule. We will then use the c++ sources and include into our build of libtorch.so This requires various changes to work including external and internal changes. Since these require internal changes we need to co-dev and in the co-dev environment I haven't found a way to sync submodule changes + internal only changes. This is unused for now Pull Request resolved: #145502 Approved by: https://github.com/Skylion007
drisspg
commented
Jan 24, 2025
| # feature by default We dont currently document this feature because we don't | ||
| # Suspect users building from source will need this | ||
| add_definitions(-DFLASHATTENTION_DISABLE_ALIBI) | ||
| add_definitions(-DFLASHATTENTION_DISABLE_SOFTCAP) |
pytorch-bot bot
pushed a commit
that referenced
this pull request
Feb 4, 2025
…#144120) Summary: Pull Request resolved: #144120 # Summary ### Sticky points Cuda-graph rng handling has changed / deviated from original implementation. We will be left with a dangling 'offset' val and confusing naming due to BC ## Dependencies - Flash PR: Dao-AILab/flash-attention#1419 ### Other Points - The BC linter is complaining about losing generate.py and its functions which is not real BC surface cc albanD imported-using-ghimport Test Plan: Imported from OSS Building in dev `buck build @//mode/dev-nosan -c fbcode.nvcc_arch=h100a //caffe2:ATen-cu --show-full-output ` I and Nming the .so I do see that the flash symbols are correctly named: ``` 0000000001c3dfb0 t pytorch_flash::run_mha_bwd(pytorch_flash::Flash_bwd_params&, CUstream_st*)::$_0::operator()() const::{lambda()#1}::operator()() const::{lambda()#1}::operator()() const::{lambda()#7}::operator()() const 0000000001c36080 t pytorch_flash::run_mha_fwd(pytorch_flash::Flash_fwd_params&, CUstream_st*, bool)::$_0::operator()() const::{lambda()#2}::operator()() const::{lambda()#1}::operator()() const::{lambda()#6}::operator()() const 0000000001c360e0 t pytorch_flash::run_mha_fwd(pytorch_flash::Flash_fwd_params&, CUstream_st*, bool)::$_0::operator()() const::{lambda()#2}::operator()() const::{lambda()#1}::operator()() const::{lambda()#7}::operator()() const 0000000001c35fc0 t pytorch_flash::run_mha_fwd(pytorch_flash::Flash_fwd_params&, CUstream_st*, bool)::$_0::operator()() const::{lambda()#1}::operator()() const::{lambda()#1}::operator()() const::{lambda()#6}::operator()() const 0000000001c36020 t pytorch_flash::run_mha_fwd(pytorch_flash::Flash_fwd_params&, CUstream_st*, bool)::$_0::operator()() const::{lambda()#1}::operator()() const::{lambda()#1}::operator()() const::{lambda()#7}::operator()() const ``` Reviewed By: vkuzo Differential Revision: D68502879 Pulled By: drisspg
drisspg
added a commit
to drisspg/pytorch
that referenced
this pull request
Feb 5, 2025
…pytorch#146372) Summary: Pull Request resolved: pytorch#146372 Pull Request resolved: pytorch#144120 # Summary ### Sticky points Cuda-graph rng handling has changed / deviated from original implementation. We will be left with a dangling 'offset' val and confusing naming due to BC ## Dependencies - Flash PR: Dao-AILab/flash-attention#1419 ### Other Points - The BC linter is complaining about losing generate.py and its functions which is not real BC surface cc albanD imported-using-ghimport Test Plan: Imported from OSS Building in dev `buck build @//mode/dev-nosan -c fbcode.nvcc_arch=h100a //caffe2:ATen-cu --show-full-output ` I and Nming the .so I do see that the flash symbols are correctly named: ``` 0000000001c3dfb0 t pytorch_flash::run_mha_bwd(pytorch_flash::Flash_bwd_params&, CUstream_st*)::$_0::operator()() const::{lambda()#1}::operator()() const::{lambda()#1}::operator()() const::{lambda()pytorch#7}::operator()() const 0000000001c36080 t pytorch_flash::run_mha_fwd(pytorch_flash::Flash_fwd_params&, CUstream_st*, bool)::$_0::operator()() const::{lambda()#2}::operator()() const::{lambda()#1}::operator()() const::{lambda()pytorch#6}::operator()() const 0000000001c360e0 t pytorch_flash::run_mha_fwd(pytorch_flash::Flash_fwd_params&, CUstream_st*, bool)::$_0::operator()() const::{lambda()#2}::operator()() const::{lambda()#1}::operator()() const::{lambda()pytorch#7}::operator()() const 0000000001c35fc0 t pytorch_flash::run_mha_fwd(pytorch_flash::Flash_fwd_params&, CUstream_st*, bool)::$_0::operator()() const::{lambda()#1}::operator()() const::{lambda()#1}::operator()() const::{lambda()pytorch#6}::operator()() const 0000000001c36020 t pytorch_flash::run_mha_fwd(pytorch_flash::Flash_fwd_params&, CUstream_st*, bool)::$_0::operator()() const::{lambda()#1}::operator()() const::{lambda()#1}::operator()() const::{lambda()pytorch#7}::operator()() const ``` Reviewed By: vkuzo Differential Revision: D68502879 Pulled By: drisspg
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
Stack from ghstack (oldest at bottom):
Summary
Sticky points
Cuda-graph rng handling has changed / deviated from original implementation. We will be left with a dangling 'offset' val and confusing naming due to BC
Dependencies
Other Points
cc @albanD
Differential Revision: D68502879