Bazel#4528
Conversation
will-cromar
left a comment
There was a problem hiding this comment.
This is awesome! Thanks so much for your hard work on this
| bazel clean | ||
| popd | ||
| else | ||
| # Overlay llvm-raw secondary cache. The remote cache should be updated |
There was a problem hiding this comment.
@yeounoh regarding removing this secondary cahce, I think we added to make upstream CI more stable.
There was a problem hiding this comment.
Not sure what to do here. We can patch the TF repo with this change.
There was a problem hiding this comment.
I think that llvm tensorflow external repository already names both of these URLs so bazel will fetch from both and there's no need to do this anymore.
Not 100% sure but it looks like it.
wdyt?
There was a problem hiding this comment.
I've patched in TF with this 2 sed equivalents so we have it as a patch, not as a in-place replacement.
5a48f3c to
81912ed
Compare
|
CC @ManfeiBai |
|
@stgpetrovic The branch cut for the pytorch 2.0 is scheduled to be this Friday, I am a bit hesitant to merge this one right before the branch cut. Do you have any opinion? |
|
I think that we should first cut and then we have a large merge window for this large change; something's bound to explode here. |
f606826 to
6438e58
Compare
7e2d557 to
7aa34be
Compare
* Fix HLO dumping (#4619) * Update TF pin to 2/13 (#4615) * Update TF pin to 2/13 * Fix pinned commit * Add patch to revert TF 3e24055 * Add comment to new patch * Fix patch command in TPU CI (#4623) * Skip execution for extract_compiled_graph (#4612) * Only warm up cache for dynamo extract_graph step * Add missing config * Make sure warm up run does not cause place holder to be created * Fix tests * Disable failing `test_operations.py` tests on TPU (#4622) * Disable `test_operations.py` tests failing on TPU * Add to TPU CI * Bazel (#4528) * Replace tensorflow with a bazel external repository * Basic migration to bazel for xla_client. * Revert to blob * Add vscode config. * Update newlines * Merge with pjrt client test build changes. * Migrate tests to new build * Format test and plugin * Order imports * Conditionally apply tf patches; apply pt patches always. * Format python * configure formatters * Mirror TF pin update an fixes in bazel. * Support local and sandboxed build based on flags * Add cloud cache URLs for llvm. * Merge with upstream * Update TF pin * Fix patching regression * Revert "Bazel (#4528)" (#4631) This reverts commit 3a90f5a. --------- Co-authored-by: JackCaoG <59073027+JackCaoG@users.noreply.github.com> Co-authored-by: Will Cromar <wcromar@google.com> Co-authored-by: stgpetrovic <stgpetrovic@gmail.com>
* Replace tensorflow with a bazel external repository * Basic migration to bazel for xla_client. * Revert to blob * Add vscode config. * Update newlines * Merge with pjrt client test build changes. * Migrate tests to new build * Format test and plugin * Order imports * Conditionally apply tf patches; apply pt patches always. * Format python * configure formatters * Mirror TF pin update an fixes in bazel. * Support local and sandboxed build based on flags * Add cloud cache URLs for llvm. * Merge with upstream * Update TF pin * Fix patching regression
* Replace tensorflow with a bazel external repository * Basic migration to bazel for xla_client. * Revert to blob * Add vscode config. * Update newlines * Merge with pjrt client test build changes. * Migrate tests to new build * Format test and plugin * Order imports * Conditionally apply tf patches; apply pt patches always. * Format python * configure formatters * Mirror TF pin update an fixes in bazel. * Support local and sandboxed build based on flags * Add cloud cache URLs for llvm. * Merge with upstream * Update TF pin * Fix patching regression
* Replace tensorflow with a bazel external repository * Basic migration to bazel for xla_client. * Revert to blob * Add vscode config. * Update newlines * Merge with pjrt client test build changes. * Migrate tests to new build * Format test and plugin * Order imports * Conditionally apply tf patches; apply pt patches always. * Format python * configure formatters * Mirror TF pin update an fixes in bazel. * Support local and sandboxed build based on flags * Add cloud cache URLs for llvm. * Merge with upstream * Update TF pin * Fix patching regression
Bazel integration for PyTorch/XLA's
libxla_computation_client.so.This is a large change but should have no effect on the build.
Work done:
This enables for:
bazel build, no shell or the likeFuture work:
torch_xlato bazel too to get all perks named above for the whole codebase