| COMMIT |
0.20 |
Automated Code Change |
|
'Automated Code Change' is generic; poss |
2026-04-02 |
| PR |
0.20 |
Fix grammar and formatting in README installation section |
|
Slightly formal, but plausible for a min |
2026-03-13 |
| COMMIT |
0.10 |
Automated Code Change |
|
Generic 'Automated Code Change'; could i |
2026-04-01 |
| PR |
0.10 |
Improve AutoGraph if-expression error for mismatched branche |
|
Uses template sections, but concise and |
2026-04-02 |
| PR |
0.10 |
docs: Fix typo 'occured' to 'occurred' in metrics |
|
Mostly template-driven; free-text is pre |
2026-03-21 |
| PR |
0.10 |
Validate tf.name_scope names with spaces in eager and graph |
|
Some template sections, but free-text is |
2026-03-24 |
| PR |
0.10 |
Fix integer overflow in stablehlo_reduce_window output dimen |
|
Structured technical write-up, but no AI |
2026-04-01 |
| COMMIT |
0.00 |
Migrate multithreaded_compilation_test to PjRt runtime. |
|
Technical changelog, domain-specific, no |
2026-04-02 |
| COMMIT |
0.00 |
[tsl] Clean up #includes in errors.h and status.h |
|
Terse, uses abbreviations, concise goal |
2026-04-02 |
| COMMIT |
0.00 |
[XLA] Wrap loops containing only DCHECKs with ifndef NDEBUG. |
|
Domain-specific, terse, no AI markers. |
2026-04-02 |
| COMMIT |
0.00 |
Use salted hashing for key_hash so that we have fewer collis |
|
Informal, specific benchmarks, no AI phr |
2026-04-02 |
| COMMIT |
0.00 |
[XLA:CPU] Add Opt Preset FAST_COMPILE for interactive and te |
|
Changelog style, has domain specifics, n |
2026-04-02 |
| COMMIT |
0.00 |
Run `MakeArraysFromHostBuffer` and `CopyToHostBuffer` tests |
|
Direct, references concrete tests, clear |
2026-04-02 |
| COMMIT |
0.00 |
[IFRT IR] Add compile option to set strict memory reservatio |
|
Concise technical addition, no AI stylis |
2026-04-02 |
| COMMIT |
0.00 |
[XLA:CPU] Add Pi approximation to fusion microbenchmark. |
|
Direct technical language, context-rich, |
2026-04-02 |
| COMMIT |
0.00 |
Implement non-descending layout support in `CpuRawBuffer::Co |
|
Bug fix explained clearly, technical, no |
2026-04-02 |
| COMMIT |
0.00 |
Centralize PjRt Transpose Caching in CommonPjRtClient. |
|
Technical description, bulleted updates, |
2026-04-02 |
| COMMIT |
0.00 |
Drop deduplicating any calls on outliner. |
|
Technical jargon and concise, informal t |
2026-04-02 |
| COMMIT |
0.00 |
PR #40117: Update XLA oneAPI presubmit build targets |
|
Detailed, domain-specific PR with some f |
2026-04-02 |
| COMMIT |
0.00 |
Populate device_memory_bytes_limit attribute while reconstru |
|
Domain-specific content and phrasing; co |
2026-04-02 |
| COMMIT |
0.00 |
Reverts e6f36bd7a12f87aac6f21a5068719dfc91bed3eb |
|
Single-purpose revert message; standard |
2026-04-02 |
| COMMIT |
0.00 |
Remove unneeded condition variable from coordination service |
|
Terse, domain-specific commit message. |
2026-04-02 |
| COMMIT |
0.00 |
Make `HloProgram::name()` return a unique and stable name ev |
|
Technical and specific wording; informal |
2026-04-02 |
| COMMIT |
0.00 |
Rollback of PR #39854 |
|
Standard rollback message, typical commi |
2026-04-02 |
| COMMIT |
0.00 |
Original change is causing presubmit tests failure. Rolling |
|
Human rationale and informal explanation |
2026-04-02 |
| COMMIT |
0.00 |
HLO builder tuple_test is hardware-independent. |
|
Technical explanation with human-written |
2026-04-02 |
| COMMIT |
0.00 |
Reverts cd712c81dacec1d737af7e27dc01c8fc0a7f5020 |
|
Standard auto-generated revert message f |
2026-04-02 |
| COMMIT |
0.00 |
Consolidate sanitizer build checks for tfrt_session_python_t |
|
Terse, domain-specific commit. Likely hu |
2026-04-02 |
| COMMIT |
0.00 |
PR #40311: [xla:gpu] Delete vestigial WaitForStreamsThunk |
|
Domain jargon, imports, no AI markers. |
2026-04-02 |
| COMMIT |
0.00 |
Make XLA:GPU propagate all CPU HW features to FFI and XLA:CP |
|
Somewhat formal but uses technical detai |
2026-04-02 |
| COMMIT |
0.00 |
[mpmd] Use reserved_hbm_bytes attribute to reserve per-fragm |
|
Template style, technical context, not A |
2026-04-02 |
| COMMIT |
0.00 |
Bump rules_python 1.8.4 -> 1.8.5 to restore correct URL-base |
|
Specific technical context, informal and |
2026-04-02 |
| COMMIT |
0.00 |
Add TargetMachineOptions::Native() to create options for the |
|
Technical vocabulary, tests, not AI-like |
2026-04-02 |
| COMMIT |
0.00 |
[XLA:GPU] Do not run large HLO fusions concurrently in comma |
|
Short, technical, no AI text signals. |
2026-04-02 |
| COMMIT |
0.00 |
[XLA:GPU] Use XLA_VLOG_DEVICE where device id is useful in l |
|
Terse, log context, fits human commit. |
2026-04-02 |
| COMMIT |
0.00 |
Stabilize SymbolicExpr and SymbolicMap string representation |
|
CL note, details concise, domain-specifi |
2026-04-02 |
| COMMIT |
0.00 |
PR #39373: [ROCm] Fix issue with unsupported types combinati |
|
Uses PR template; free-text sections are |
2026-04-02 |
| COMMIT |
0.00 |
Automated Code Change |
|
Standard auto-generated commit; no AI ha |
2026-04-02 |
| COMMIT |
0.00 |
PR #40232: Add missing kAllGatherStart/kAllGatherDone verifi |
|
Template-driven; free-text is concise an |
2026-04-02 |
| COMMIT |
0.00 |
Migrate indexing analysis from MLIR Affine to SymbolicMap |
|
Technical, domain-specific, no AI-style |
2026-04-02 |
| COMMIT |
0.00 |
Remove unused `options` parameter from `CompileTargetBinary` |
|
Concise commit, clear changelog, no AI-w |
2026-04-02 |
| COMMIT |
0.00 |
Remove test timeout long from prng_test. |
|
Very brief, informal, and specific commi |
2026-04-02 |
| COMMIT |
0.00 |
[XLA:GPU] Accumulate TiledHloInstructions for the runtime va |
|
Domain-specific vocabulary, informal exp |
2026-04-02 |
| COMMIT |
0.00 |
[XLA:GPU] Introduce EmitterContext to group the params. |
|
Short, technical title; lacks AI writing |
2026-04-02 |
| COMMIT |
0.00 |
Avoid crashing in HloVerifier for unexpected shapes in AllGa |
|
Bugfix commit, informal language without |
2026-04-02 |
| COMMIT |
0.00 |
PR #40252: [xla:gpu] Use Global HangWatchdog in se_gpu_pjrt_ |
|
PR summary is short, direct, and technic |
2026-04-02 |
| COMMIT |
0.00 |
[NFC] Annotate forwarding headers with `INLINER_FORWARD_TO`. |
|
Technical and concise; normal engineerin |
2026-04-02 |
| COMMIT |
0.00 |
Make CpuTargetMachineOptions always available through GpuTop |
|
Human-written, clear summary, no AI sign |
2026-04-02 |
| COMMIT |
0.00 |
Migrate XLA codegen attributes and operations to the new Sym |
|
Contains domain-specific terminology and |
2026-04-02 |
| COMMIT |
0.00 |
[XLA:GPU] Normalize two batch dimensions (which may appear a |
|
Human style, includes jargon and straigh |
2026-04-02 |
| COMMIT |
0.00 |
Roll-forward Migrate XLA GPU emitters and HWIR to use Symbol |
|
Direct, technical, and detailed—no AI ha |
2026-04-02 |
| COMMIT |
0.00 |
PR #39951: [xla:gpu] Add missing scheduling for all-gather-s |
|
Mostly template import text plus concise |
2026-04-02 |
| COMMIT |
0.00 |
Improve parentheses in SymbolicExpr serialization |
|
Simple, informal explanation with minima |
2026-04-02 |
| COMMIT |
0.00 |
PR #39843: Bump jwalton/gh-find-current-pr from 1.3.3 to 1.3 |
|
— |
2026-04-02 |
| COMMIT |
0.00 |
PR #39622: [ROCm] Use BFCAllocator for scratch allocations n |
|
Slightly formal, but includes typos and |
2026-04-02 |
| COMMIT |
0.00 |
[XLA:GPU] Add symmetric memory to collective memory cache to |
|
Technical, informal, and uses in-domain |
2026-04-02 |
| COMMIT |
0.00 |
Always pass TargetMachineOptions to FFI calls from CustomCal |
|
Precise engineering note, no AI text ind |
2026-04-02 |
| COMMIT |
0.00 |
Automated Code Change |
|
Generic automated commit, no AI signals, |
2026-04-02 |
| COMMIT |
0.00 |
Automated Code Change |
|
Generic automated commit, no AI signals, |
2026-04-02 |
| COMMIT |
0.00 |
Update rules_ml_toolchain to the version with Hermetic Pytho |
|
Human-like brief update description, cle |
2026-04-02 |
| COMMIT |
0.00 |
[IFRT] Add more general `Layout::ByteSize` static methods |
|
Technical, detailed, domain‑jargon prese |
2026-04-02 |
| COMMIT |
0.00 |
[IFRT] Add GetName() method to HloProgram |
|
Issue-only commit, concise, no markers o |
2026-04-02 |
| COMMIT |
0.00 |
Fix heap-use-after-free in ConstraintPropagator::PropagateCo |
|
Domain-specific explanation, informal, n |
2026-04-02 |
| COMMIT |
0.00 |
Remove condition hints that are based on future's readiness |
|
Concise and technical rationale, informa |
2026-04-02 |
| COMMIT |
0.00 |
Mark coordination service movable but not copyable. |
|
Short, technical, lacks structure common |
2026-04-01 |
| COMMIT |
0.00 |
[GPU] Use CheckFabricHomogeneity in RaggedAllToAllThunk |
|
Technical, structured, uses domain terms |
2026-04-01 |
| COMMIT |
0.00 |
Reverts f6b594dc036bed5a4d59ee60e69698e8219501b2 |
|
Standard revert message, common human ph |
2026-04-01 |
| COMMIT |
0.00 |
[ReplicaGroupV3][Refactor][18/n] migrate more XLA functions |
|
Terse, technical, commit-style message w |
2026-04-01 |
| COMMIT |
0.00 |
[XLA] Relax memory ordering of `entry->next` access in `~Obj |
|
Technical writeup, natural phrasing, ref |
2026-04-01 |
| COMMIT |
0.00 |
Remove `allow_new_incarnation_to_reconnect` config from coor |
|
Brief removal message; no AI signals, do |
2026-04-01 |
| COMMIT |
0.00 |
Make an explicit ref type for PjRtDeviceEvent to allow easil |
|
Specific technical intent, colloquial wo |
2026-04-01 |
| COMMIT |
0.00 |
[ReplicaGroupV3][Refactor][17/n] migrate functions in spmd/ |
|
Refactor message, technical context, ter |
2026-04-01 |
| COMMIT |
0.00 |
Add PID metadata to XPlanes and TraceMeConsumers to enable p |
|
Describes technical change in one clear |
2026-04-01 |
| COMMIT |
0.00 |
[XLA] Make `ObjectPool::Entry::next` atomic |
|
Short, domain-specific commit; no AI-lik |
2026-04-01 |
| COMMIT |
0.00 |
Benchmark for traceme_encode. |
|
Minimal benchmark addition, no AI langua |
2026-04-01 |
| COMMIT |
0.00 |
[tfrt] Introduce DeviceInterconnectResource to bridge Global |
|
Contains some structure, but detailed te |
2026-04-01 |
| COMMIT |
0.00 |
Update Shardy to the latest commit. |
|
Routine update, minimal content, highly |
2026-04-01 |
| COMMIT |
0.00 |
[XLA] Remove obsolete shmap & sdy manual computation inliner |
|
Technical commit message with revert and |
2026-04-01 |
| COMMIT |
0.00 |
Grant users of XLA access to gpu_topology_util |
|
Brief, jargon-heavy message; no AI phras |
2026-04-01 |
| COMMIT |
0.00 |
Add a test for `jit(lambda x: x, out_shardings=unreduced)` a |
|
Concise, uses jargon and code refs, read |
2026-04-01 |
| COMMIT |
0.00 |
Internal visibility change. |
|
Extremely terse and generic; typical hum |
2026-04-01 |
| COMMIT |
0.00 |
Drop fully deduplicating on outliner and leave it to unflatt |
|
Uses domain-specific concepts; informal |
2026-04-01 |
| COMMIT |
0.00 |
Add SymbolicExpr::ReplaceDims(absl::Span |
|
Technical addition; notes requirement, n |
2026-04-01 |
| COMMIT |
0.00 |
Fix dependency issues. |
|
Minimal and typical engineering commit m |
2026-04-01 |
| COMMIT |
0.00 |
Drop func and call sharding conflict resolution on outliner. |
|
Concise, abrupt phrasing; uses specific |
2026-04-01 |
| COMMIT |
0.00 |
[XLA:CPU] Lower Cbrt to Libm call. |
|
Detailed technical description, no AI st |
2026-04-01 |
| COMMIT |
0.00 |
[XLA:GPU] Simplify and retune the splitk heuristic. |
|
Matter-of-fact, technical, typical commi |
2026-04-01 |
| COMMIT |
0.00 |
Handle V3 named sharding in windowed input resharding. |
|
Technical, concise, domain-specific and |
2026-04-01 |
| COMMIT |
0.00 |
Add original func names on shardy outliner. |
|
Concise, technical, contains no AI style |
2026-04-01 |
| COMMIT |
0.00 |
Fix use-after-free in `SymbolicMap::GetResults()` |
|
Domain jargon, detailed technical explan |
2026-04-01 |
| COMMIT |
0.00 |
[XLA:GPU] Fix Triton loads dropping the mask and causing OOB |
|
Technical jargon and direct explanation, |
2026-04-01 |
| COMMIT |
0.00 |
Add if_onednn_async to tf_copts. |
|
Short, domain-specific, no AI style dete |
2026-04-01 |
| COMMIT |
0.00 |
Run shardy/google/integrate_latest.sh to integrate. |
|
Minimal description, human style, typica |
2026-04-01 |
| COMMIT |
0.00 |
Add reshards for when func has no argument shardings but the |
|
Concise, domain-specific, human phrasing |
2026-04-01 |
| COMMIT |
0.00 |
Allow creating a GPU PjRtClient without devices |
|
Technical, explains code change, referen |
2026-04-01 |
| COMMIT |
0.00 |
Handle one size axis in `IsReplicated` checks |
|
Very brief, domain-specific term, human |
2026-04-01 |
| COMMIT |
0.00 |
Enforce consistent simplification at creation time in Symbol |
|
Technical, with spelling errors and info |
2026-04-01 |
| COMMIT |
0.00 |
[XLA:CPU] Add benchmarks for Cbrt |
|
Short, jargon-filled changelog typical o |
2026-04-01 |
| COMMIT |
0.00 |
Wrap XNNPack delegate check macros in `do { ... } while(0)`. |
|
Technical phrasing and concise explanati |
2026-04-01 |
| COMMIT |
0.00 |
[XLA:GPU] Mark NVSHMEM thunks as deprecated. |
|
Direct and technical; no AI hallmarks pr |
2026-04-01 |
| COMMIT |
0.00 |
PR #39725: [xla] Add global per-process hang watchdog |
|
Very technical, includes human-style sum |
2026-04-01 |
| COMMIT |
0.00 |
Use utility to walk on call graphs. |
|
Terse commit message, strong human style |
2026-04-01 |
| COMMIT |
0.00 |
[XLA:GPU] Remove NCCL from dependencies if no_nccl definitio |
|
Technical feature flag; clear human-auth |
2026-04-01 |
| COMMIT |
0.00 |
Run shardy/google/integrate_latest.sh to integrate. |
|
Build automation statement, typical of h |
2026-04-01 |
| COMMIT |
0.00 |
[XLA:GPU] Sort the tiled instructions taking RT variables in |
|
Long, technical, with domain context and |
2026-04-01 |
| PR |
0.00 |
Erase async collectives in XLA CPU. |
|
— |
2026-04-01 |
| PR |
0.00 |
Remove unused reset logic from coordination agent. |
|
— |
2026-04-02 |
| PR |
0.00 |
[StableHLO reference] Add support for broadcast/dot in inter |
|
— |
2026-04-02 |
| PR |
0.00 |
Eradicate Ruy from TFLite CPU Backend for GEMM. |
|
— |
2026-04-02 |
| PR |
0.00 |
[XLA:CPU] Rename is_fusion_emitters to use_fusion_emitters f |
|
— |
2026-04-02 |
| PR |
0.00 |
Migrate multithreaded_compilation_test to PjRt runtime. |
|
— |
2026-04-02 |
| PR |
0.00 |
Internal build change. |
|
— |
2026-04-02 |
| PR |
0.00 |
Refactor vhlo::TensorV1Attr to store DenseElementsAttr |
|
— |
2026-03-23 |
| PR |
0.00 |
[tsl] Clean up #includes in errors.h and status.h |
|
— |
2026-03-18 |
| PR |
0.00 |
[IFRT] Add `Value::ByteSize` method |
|
— |
2026-04-01 |
| PR |
0.00 |
Integrate StableHLO at openxla/stablehlo@05fdca09 |
|
— |
2026-03-30 |
| PR |
0.00 |
[XLA] Wrap loops containing only DCHECKs with ifndef NDEBUG. |
|
— |
2026-04-02 |
| PR |
0.00 |
Added a bounds check to FastParseExample |
|
— |
2026-04-02 |
| PR |
0.00 |
Recursively clean up some unused TSL dependencies |
|
— |
2026-04-02 |
| PR |
0.00 |
[XLA:MSA] Fix eviction bug, and update interval_tree_ via AP |
|
— |
2026-04-02 |
| PR |
0.00 |
[XLA:MSA] Process colored positions and uses with aliasing c |
|
— |
2026-04-02 |
| PR |
0.00 |
Support input coloring for sync copy and slice uses. |
|
— |
2026-04-02 |
| PR |
0.00 |
Change default for HloPrintOptions::syntax_sugar_async_ops_ |
|
— |
2026-04-02 |
| PR |
0.00 |
Use salted hashing for key_hash so that we have fewer collis |
|
— |
2026-04-01 |
| PR |
0.00 |
[XLA:CPU] Add Opt Preset FAST_COMPILE for interactive and te |
|
— |
2026-03-31 |
| PR |
0.00 |
Run `MakeArraysFromHostBuffer` and `CopyToHostBuffer` tests |
|
— |
2026-03-31 |
| PR |
0.00 |
Fix tf.math.cumulative_logsumexp producing NaN for +inf inpu |
|
Technical detail, concise, some typos, n |
2026-04-02 |
| PR |
0.00 |
[IFRT IR] Add compile option to set strict memory reservatio |
|
— |
2026-04-02 |
| PR |
0.00 |
[XLA:CPU] Add Pi approximation to fusion microbenchmark. |
|
— |
2026-04-02 |
| PR |
0.00 |
Implement non-descending layout support in `CpuRawBuffer::Co |
|
— |
2026-04-02 |
| PR |
0.00 |
Centralize PjRt Transpose Caching in CommonPjRtClient. |
|
— |
2026-04-02 |
| PR |
0.00 |
Integrate LLVM at llvm/llvm-project@293623ce99d4 |
|
— |
2026-04-01 |
| PR |
0.00 |
Validate num_reduction_axes against XNN_MAX_TENSOR_DIMS in V |
|
Very brief, domain-specific, and direct |
2026-03-23 |
| PR |
0.00 |
[XLA:MSA] Allow all async slice conversion candidates to be |
|
— |
2026-04-02 |
| PR |
0.00 |
Integrate LLVM at llvm/llvm-project@7ccd92e5e6e5 |
|
— |
2026-03-31 |
| PR |
0.00 |
Split outliner and unflattener by completely delegating dedu |
|
— |
2026-03-21 |
| PR |
0.00 |
[XLA:AlgebraicSimplifier] Simplify AR/RS(broadcast(constant) |
|
— |
2026-04-02 |
| PR |
0.00 |
introduce KernelCompiler interface |
|
— |
2026-04-02 |
| PR |
0.00 |
Change SparseCore offloading options for two test which were |
|
— |
2026-04-02 |
| PR |
0.00 |
PR #40117: Update XLA oneAPI presubmit build targets |
|
— |
2026-04-01 |
| PR |
0.00 |
Replace `MultiplyAndCheckOverflow` with `CheckedInt`. |
|
— |
2026-04-02 |
| PR |
0.00 |
Add a helper template to compute simple expressions and chec |
|
— |
2026-04-02 |
| PR |
0.00 |
Populate device_memory_bytes_limit attribute while reconstru |
|
— |
2026-03-31 |
| PR |
0.00 |
Reverts e6f36bd7a12f87aac6f21a5068719dfc91bed3eb |
|
— |
2026-04-02 |
| PR |
0.00 |
Remove unneeded condition variable from coordination service |
|
— |
2026-04-02 |
| PR |
0.00 |
Make `HloProgram::name()` return a unique and stable name ev |
|
— |
2026-04-02 |
| PR |
0.00 |
Reverts c6d844d439cf17d3173f4d557646d55b895298ba |
|
— |
2026-04-02 |
| PR |
0.00 |
Original change is causing presubmit tests failure. Rolling |
|
— |
2026-04-01 |
| PR |
0.00 |
HLO builder tuple_test is hardware-independent. |
|
— |
2026-04-01 |
| PR |
0.00 |
Reverts cd712c81dacec1d737af7e27dc01c8fc0a7f5020 |
|
— |
2026-04-02 |
| PR |
0.00 |
Consolidate sanitizer build checks for tfrt_session_python_t |
|
— |
2026-04-01 |
| PR |
0.00 |
PR #40311: [xla:gpu] Delete vestigial WaitForStreamsThunk |
|
— |
2026-04-02 |
| PR |
0.00 |
Serialize peak buffers to fix 100% fragmentation metrics on |
|
— |
2026-04-02 |
| PR |
0.00 |
Move shardy inliner into mlir sdy. Part 2. |
|
— |
2026-04-01 |
| PR |
0.00 |
[XLA] Put BufferAssignment dumps on a diet |
|
— |
2026-03-28 |
| PR |
0.00 |
Add element-wise boolean masks support to MLD select_v2 op |
|
— |
2026-04-02 |
| PR |
0.00 |
When updating the reachability from a Hlo instruction I1 tha |
|
— |
2026-03-27 |
| PR |
0.00 |
[XLA:GPU] Emit dynamic slice via the new tiling. |
|
— |
2026-04-02 |
| PR |
0.00 |
Rollforward for cl/799580282. It's not a pure rollforward be |
|
— |
2026-04-02 |
| PR |
0.00 |
Fix confusing MemoryError in tf.squeeze with Tensor axis |
|
Direct problem/solution writing, no AI s |
2026-02-11 |
| PR |
0.00 |
Fix tf.cond to validate both branches in eager mode |
|
Terse summary, direct and informal, no A |
2026-02-03 |
| PR |
0.00 |
Make XLA:GPU propagate all CPU HW features to FFI and XLA:CP |
|
— |
2026-04-02 |
| PR |
0.00 |
[mpmd] Use reserved_hbm_bytes attribute to reserve per-fragm |
|
— |
2026-04-01 |
| PR |
0.00 |
Bump rules_python 1.8.4 -> 1.8.5 to restore correct URL-base |
|
— |
2026-03-31 |
| PR |
0.00 |
Migrate codegen/tiling to SymbolicMap |
|
— |
2026-03-26 |
| PR |
0.00 |
PR #40310: [xla:gpu] Fix UndefinedBehaviorSanitizer for thun |
|
— |
2026-04-02 |
| PR |
0.00 |
Automated Code Change |
|
— |
2026-04-01 |
| PR |
0.00 |
Add TargetMachineOptions::Native() to create options for the |
|
— |
2026-04-02 |
| PR |
0.00 |
[XLA:GPU] Do not run large HLO fusions concurrently in comma |
|
— |
2026-04-02 |
| PR |
0.00 |
[XLA:GPU] Use XLA_VLOG_DEVICE where device id is useful in l |
|
— |
2026-04-02 |
| PR |
0.00 |
Stabilize SymbolicExpr and SymbolicMap string representation |
|
— |
2026-04-02 |
| PR |
0.00 |
Migrate IndexingMap internal representation to SymbolicMap |
|
— |
2026-04-02 |
| PR |
0.00 |
PR #34957: [ROCm] Enable matmul perf table implementation fo |
|
— |
2026-04-02 |
| PR |
0.00 |
PR #34957: [ROCm] Enable matmul perf table implementation fo |
|
— |
2026-03-25 |
| PR |
0.00 |
PR #39925: [ROCm] Remove deprecated hipCtx_t and HIP context |
|
— |
2026-04-02 |
| PR |
0.00 |
PR #39373: [ROCm] Fix issue with unsupported types combinati |
|
— |
2026-04-02 |
| PR |
0.00 |
Reverts 2b53c0051c18c4fe2dff6db5f7ade0e633cb6d99 |
|
— |
2026-04-02 |
| PR |
0.00 |
Automated Code Change |
|
— |
2026-04-02 |
| PR |
0.00 |
Migrate Tiling Expressions from mlir::AffineExpr to xla::Sym |
|
— |
2026-03-24 |
| PR |
0.00 |
PR #40232: Add missing kAllGatherStart/kAllGatherDone verifi |
|
— |
2026-04-02 |
| PR |
0.00 |
Migrate indexing analysis from MLIR Affine to SymbolicMap |
|
— |
2026-03-12 |
| PR |
0.00 |
Remove unused `options` parameter from `CompileTargetBinary` |
|
— |
2026-04-02 |
| PR |
0.00 |
Remove redundant dot_operation_single_threaded_runtime_test. |
|
— |
2026-03-30 |
| PR |
0.00 |
Remove test timeout long from prng_test. |
|
— |
2026-04-02 |
| PR |
0.00 |
[XLA:GPU] Accumulate TiledHloInstructions for the runtime va |
|
— |
2026-04-02 |
| PR |
0.00 |
mlir_hlo: add scan_dim_size to mhlo.scan and propagate from |
|
— |
2026-04-02 |
| PR |
0.00 |
[XLA:GPU] Introduce EmitterContext to group the params. |
|
— |
2026-04-02 |
| PR |
0.00 |
Avoid crashing in HloVerifier for unexpected shapes in AllGa |
|
— |
2026-04-02 |
| PR |
0.00 |
Fix eigh_tridiagonal crashing in tf.function when select='v' |
|
Technical description with bugs, uses sp |
2026-04-02 |
| PR |
0.00 |
[XLA:GPU] Added concat to the allowlist of block-level emitt |
|
— |
2026-03-30 |
| PR |
0.00 |
PR #40252: [xla:gpu] Use Global HangWatchdog in se_gpu_pjrt_ |
|
— |
2026-04-02 |
| PR |
0.00 |
[NFC] Annotate forwarding headers with `INLINER_FORWARD_TO`. |
|
— |
2026-04-02 |
| PR |
0.00 |
Make CpuTargetMachineOptions always available through GpuTop |
|
— |
2026-04-02 |
| PR |
0.00 |
[XLA:GPU] Disable split-k rewrite in Triton int4 device test |
|
— |
2026-04-02 |
| PR |
0.00 |
jpeg: replace manual new[]/delete[] with unique_ptr in Uncom |
|
Technical, domain-specific, no formal or |
2026-03-30 |
| PR |
0.00 |
Migrate XLA codegen attributes and operations to the new Sym |
|
— |
2026-02-11 |
| PR |
0.00 |
[XLA:GPU] Normalize two batch dimensions (which may appear a |
|
— |
2026-04-02 |
| PR |
0.00 |
1. Enable offloading bf16 reductions to ynnpack |
|
— |
2026-04-02 |
| PR |
0.00 |
PR #38198: [oneDNN] Upgrade to oneDNN v3.11 |
|
— |
2026-04-01 |
| PR |
0.00 |
Roll-forward Migrate XLA GPU emitters and HWIR to use Symbol |
|
— |
2026-03-31 |
