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

R2.1 #5651

Closed
wants to merge 24 commits into from
Closed

R2.1 #5651

wants to merge 24 commits into from

Conversation

ManfeiBai
Copy link
Collaborator

No description provided.

vanbasten23 and others added 24 commits August 29, 2023 01:09
@vanbasten23 @wonjoolee95
Disable cxx abi in ansible when building pt/xla for branch r2.0 (#5332)
3ac818e
@wonjoolee95
Update pytorch git tag for r2.1 (#5529)
3b34ae1 
Update more places

Add torch_pin
@wonjoolee95 @alanwaketan
Enable megacore_dense by default (#5520) (#5531)
bd1958f 
Summary:
This change enables megacore_dense by default to allow asynchorous cc
ops especailly for GSPMD.

Test Plan:
CI

Co-authored-by: Jiewen Tan <jwtan@google.com>
@will-cromar
Add option to unbundle libtpu (#5534) (#5536)
7a01093 
* Add optiona to unbundle libtpu

* Add clarifying note
@wonjoolee95
Revert 2.1 terraform changes (#5537)
a1d3651
@wonjoolee95 @Liyang90
Fix FSDP for Models with Frozen Weights (#5484) (#5539)
89574d2 
* Fix fsdp not freeing forzen full params

* add test

* formatting

* remove unnecessary env var in test

Co-authored-by: Liyang90 <liyanglu@google.com>
@will-cromar
Update r2.1 wheel to be compatible with PyPI (#5550)
691838f 
* Update project metadata and remove useless files

* Update README

* Add manylinux platform tag

* formatting
@wonjoolee95 @lsy323
Add resnet50-weight-quant colab notebook (#5407) (#5556)
9bcf3f7 
* Add resnet50-weight-only-quant colab notebook

* update notebook with llama blog link

Co-authored-by: Siyuan Liu <lsiyuan@google.com>
@ManfeiBai @alanwaketan
[Pin Update] Version 20230826(#5527) (#5555)
2c07df9 
Co-authored-by: Jiewen Tan <jwtan@google.com>
@will-cromar
Cherry pick pjrt:// init method rename and doc updates (#5562)
ee72332 
* Change `pjrt://` init method to `xla://` (#5560)

* Update PJRT documentation for the 2.1 release (#5557)

* Update PJRT documentation for the 2.1 release

* clarify plugins

* clarify PJRT doc

* Update `pjrt://` to `xla://`
@wonjoolee95
Update torch pin (#5575)
02858b9
@JackCaoG
Handle dynamo function without input (#5565) (#5577)
f0f9126
@JackCaoG
Make cpu tensor on XLA dynamo backend a warning instead of error (#5549…
c55ef0b 
…) (#5576)
@wonjoolee95 @aws-kingrj @seanlatias
[author: jluntamazon] Adding more explicit HLO lowering control by ex…
4ec0997 
…posing LoweringContext… (#5431) (#5580)

* Adding more explicit HLO lowering control by exposing LoweringContext (and utilities) to python for Neuron

* fixing linter issues

* fixing spacing

* apply comments and fix compilation errors

* add test for new apis

* fix linter

* update test

* update test

* modify test

* reverse back to GetIrValue()

* update test inputs with random numbers

* skip unittest because it only fails in CI

---------

Co-authored-by: aws-kingrj <78175353+aws-kingrj@users.noreply.github.com>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-3-186.us-west-2.compute.internal>
Co-authored-by: seanlatias <seanlatias@gmail.com>
@will-cromar @aws-kingrj
fixing num_local_processes typo (#5573) (#5579)
cbb1602 
Co-authored-by: aws-kingrj <78175353+aws-kingrj@users.noreply.github.com>
@JackCaoG
Move where clear pending IR is called to avoid crash (#5552) (#5582)
3e2bf8a 
* Move where clear pending IR is called to avoid crash

* fix CI

* fix CI and add some debugging messages
@will-cromar
Fix release branch and tag patterns for GitHub Actions (#5587) (#5590)
79ce1c5
@ManfeiBai @yeounoh
Improve bernoulli rng-bit-generation memory footprint (#5581) (#5589)
fa5d132 
* Allow downcasting RngUniform genenration for Bernoulli

Co-authored-by: Yeounoh Chung <yeounoh@google.com>
@will-cromar @yeounoh
Enable xla:gpu autocast for bfloat16 if not restricted (#5570) (#5591)
1e6c11d 
* Enable autocast for XLA:GPU

* linter fix

* XLA autocast test for GPU and TPU

* linter fix

* Ensure that xla autocast is properly enabled for GPU and does not crash when torch cuda is not available.

* linter fix

* Add tests

* Support bf16

* linter fix

* exclude unsupported test cases

* increase GPU test timeout to 300

Co-authored-by: Yeounoh Chung <yeounoh@google.com>
@will-cromar
Cherry-pick: Don't trigger CI build on release tag push (#5595)
5fcf79c 
Copy of #5594 on release branch
@ManfeiBai @alanwaketan
XLA pin downgrade to 20230825 (#5592) (#5604)
53fa74c 
Summary:
The 20230826 version contains openxla/xla@3b8a539
whch regresses our LLaMA2 GSPMD training benchmark. Let's rollback to version
before it.

Test Plan:
CI.

Co-authored-by: Jiewen Tan <jwtan@google.com>
@will-cromar
Clean up README for 2.1 release branch. (#5618)
2d2c80c
@ManfeiBai @will-cromar
Fix log spam when libtpu is loaded (#5619) (#5621)
b62be6e 
* Fix log spam when libtpu is loaded (#5619)

* fix conflict

* only cherry-pick

---------

Co-authored-by: Will Cromar <wcromar@google.com>
@ManfeiBai @JackCaoG @zpcore
lower NativeDropoutBackward (#5642) and lower native_dropout (#5643) (#…
b840137 
…5647)

* lower NativeDropoutBackward (#5642)

* lower NativeDropoutBackward

* fix lowering and add python test

* lower native_dropout (#5643)

* prototype version (compiling error)

* Add native_dropout manual lowering.

* fix to tensor IR and add a simple native_dropout test

* fix data type issue and update test case

* fix IR hash issue

* fix corner case when probability==0

* remove typo line

* add test case when probability=0

---------

Co-authored-by: JackCaoG <59073027+JackCaoG@users.noreply.github.com>
Co-authored-by: zpcore <piz@google.com>
@ManfeiBai ManfeiBai closed this Sep 26, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@will-cromar will-cromar Awaiting requested review from will-cromar will-cromar will be requested when the pull request is marked ready for review will-cromar is a code owner

@JackCaoG JackCaoG Awaiting requested review from JackCaoG JackCaoG will be requested when the pull request is marked ready for review JackCaoG is a code owner

@lsy323 lsy323 Awaiting requested review from lsy323 lsy323 will be requested when the pull request is marked ready for review lsy323 is a code owner

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

Successfully merging this pull request may close these issues.
5 participants
@ManfeiBai @vanbasten23 @wonjoolee95 @will-cromar @JackCaoG