[pull] master from tensorflow:master - #8799

Merged

pull[bot] merged 19 commits into

Cache-Cloud:masterfrom

tensorflow:master

Sep 1, 2026

Commits on Mar 10, 2026

Bump markdown in /ci/official/requirements_updater/numpy1_requirements
dependabot[bot]
authored
5d630bd

Commits on Aug 29, 2026

fix segfault by adding initialize for deletion
hphng
committed
e923cf4

Commits on Aug 30, 2026

fix the join() function to not waiting when start = false|none
hphng
committed
5287881
add testcase when worker or dispatcher start failed combine with join() function
hphng
committed
b08bc3d
Merge branch 'master' of https://github.com/hphng/tensorflow into fix-data-service-server-lib-crash
hphng
committed
38f75f0

Commits on Aug 31, 2026

ci: trigger rebuild
hphng
committed
f1e8404

Commits on Sep 1, 2026

revert join function back to void
hphng
committed
0fae1d1
[XLA:GPU] Rename collective_kernel_api to multi_gpu_barrier.

PatriosTheGreat
authored and
tensorflower-gardener
committed
9ddd501
Disable experimental GEMM fusion v2 in HoistFusedBitcastsReshapeTest.

vwbaker
authored and
tensorflower-gardener
committed
e26cbf5
[XLA:GPU] Support duplicating cheap operands in cuDNN conv fusion.

derdrdirk
authored and
tensorflower-gardener
committed
e3b7ac8
Skip fusion computations in HloComputationDeduplicator.

akuegel
authored and
tensorflower-gardener
committed
91fb2c3
Automated Code Change
tensorflower-gardener
committed
b07caa3
Update TritonGemmFusionTest to be parametrized and test GemmFusionV1 & V2 in preparation of V2 launch.

vwbaker
authored and
tensorflower-gardener
committed
0de11b4
Merge pull request #126442 from hphng:fix-data-service-server-lib-crash
tensorflower-gardener
committed
731ee0d
Propagate constraints across kCall boundaries in ConstraintPropagator.

bhatuzdaname
authored and
tensorflower-gardener
committed
93066fa
Merge pull request #111739 from tensorflow:dependabot/pip/ci/official/requirements_updater/numpy1_requirements/markdown-3.8.1
tensorflower-gardener
committed
674373b
Get TAP ready for Gemm Fusion V2 flag flip - explicitly set experimental_gemm_fusion flag in each case that it's meant to represent. These will get cleaned up after the flag flip, but explicitly se…

vwbaker
authored and
tensorflower-gardener
committed
26a3cc9
Leverage input_devices_for_output_map in RemapArrays for PJRT IFRT.

junwhanahn
authored and
tensorflower-gardener
committed
2d7a460
Replace TF_ASSERT_OK_AND_ASSIGN with ASSERT_OK_AND_ASSIGN in triton_fusion_analysis_test.cc

vwbaker
authored and
tensorflower-gardener
committed
7bd3888