Conversation
meta-cla
Bot
added
the
CLA Signed
Contributor
|
@xgz2 has exported this pull request. If you are a Meta employee, you can view the originating Diff in D122385730. |
xgz2
added a commit
to xgz2/torchcomms
that referenced
this pull request
Oct 1, 2026
Summary: Pull Request resolved: meta-pytorch#4290 Compile the UniFlow implementation as position-independent object targets and collect them in one reusable target graph. Both the installable library and the Python extension can consume that graph without maintaining separate source lists. Differential Revision: D122385730
xgz2
force-pushed
the
export-D122385730
branch
from
October 1, 2026 05:26
ccefca2 to
f8a78a4
Compare
meta-codesync
Bot
changed the title
xgz2
added a commit
to xgz2/torchcomms
that referenced
this pull request
Oct 1, 2026
Summary: Pull Request resolved: meta-pytorch#4290 Compile the UniFlow implementation as position-independent object targets and collect them in one reusable target graph. Both the installable library and the Python extension can consume that graph without maintaining separate source lists. Differential Revision: D122385730
xgz2
force-pushed
the
export-D122385730
branch
from
October 1, 2026 05:59
f8a78a4 to
4ad8bf5
Compare
xgz2
added a commit
to xgz2/torchcomms
that referenced
this pull request
Oct 1, 2026
Summary: Pull Request resolved: meta-pytorch#4290 Compile the UniFlow implementation as position-independent object targets and collect them in one reusable target graph. Both the installable library and the Python extension can consume that graph without maintaining separate source lists. Differential Revision: D122385730
xgz2
force-pushed
the
export-D122385730
branch
from
October 1, 2026 06:06
4ad8bf5 to
1d93c2e
Compare
xgz2
added a commit
to xgz2/torchcomms
that referenced
this pull request
Oct 1, 2026
Summary: Pull Request resolved: meta-pytorch#4290 Compile the UniFlow implementation as position-independent object targets and collect them in one reusable target graph. Both the installable library and the Python extension can consume that graph without maintaining separate source lists. Differential Revision: D122385730
xgz2
force-pushed
the
export-D122385730
branch
from
October 1, 2026 07:03
1d93c2e to
6a48ef6
Compare
Summary: Define one explicit list of the headers that make up UniFlow's public C++ API. This gives installation and compile checks a single source of truth and prevents internal headers from becoming public by accident. Differential Revision: D122385726
Summary: Make UniFlow configure from any source or build directory without a parent-tree ROOT setting. CMake creates the expected include layout in the build directory, scopes include paths to targets, uses RelWithDebInfo as the standalone default, and discovers either system fmt and spdlog packages or pinned bundled copies. Differential Revision: D122385727
Summary: Add options that control whether CUDA, RDMA, NVLink, TCP, tests, benchmarks, and Python support are built. CUDA builds now require CUDA Toolkit 12.8 or newer. A build with CUDA disabled skips all CUDA-dependent code, while existing HIP and non-CMake behavior remains unchanged. Differential Revision: D122385728
Summary: Read the UniFlow version from one VERSION file, then install the C++ library, reviewed public headers, fmt and spdlog dependencies when bundled, and CMake package files into standard directories. A separate project can find and link UniFlow without access to its source tree. Differential Revision: D122385729
xgz2
added a commit
to xgz2/torchcomms
that referenced
this pull request
Oct 1, 2026
Summary: Pull Request resolved: meta-pytorch#4290 Compile the UniFlow implementation as position-independent object targets and collect them in one reusable target graph. Both the installable library and the Python extension can consume that graph without maintaining separate source lists. Differential Revision: D122385730
xgz2
force-pushed
the
export-D122385730
branch
from
October 1, 2026 21:04
6a48ef6 to
16f5377
Compare
Summary: Pull Request resolved: meta-pytorch#4290 Compile the UniFlow implementation as position-independent object targets and collect them in one reusable target graph. Both the installable library and the Python extension can consume that graph without maintaining separate source lists. Differential Revision: D122385730
xgz2
force-pushed
the
export-D122385730
branch
from
October 2, 2026 01:44
16f5377 to
81f417b
Compare
This branch has not been deployed
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
1 participant
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:
Compile the UniFlow implementation as position-independent object targets and collect them in one reusable target graph. Both the installable library and the Python extension can consume that graph without maintaining separate source lists.
Differential Revision: D122385730