## What's Changed * [Codegen] Test Cleanup 7/8: SPIRV tests by @qedawkins in https://github.com/iree-org/iree/pull/22750 * [Codegen] Test Cleanup 6/8: LLVMGPU tests by @qedawkins in https://github.com/iree-org/iree/pull/22749 * [Codegen] Test Cleanup 5/8: LLVMCPU tests by @qedawkins in https://github.com/iree-org/iree/pull/22748 * [Codegen] Test Cleanup 2/8: Common GPU tests by @qedawkins in https://github.com/iree-org/iree/pull/22745 * Use `scf::tileAndFuseConsumer` in `GPUFuseAndHoistParallelLoops` by @MaheshRavishankar in https://github.com/iree-org/iree/pull/22617 * Integrate llvm/llvm-project@c582688b by @yzhang93 in https://github.com/iree-org/iree/pull/22758 * [Codegen] Test Cleanup 3/8: Common tests by @qedawkins in https://github.com/iree-org/iree/pull/22746 * [DispatchCreation] Add FoldExtractSliceOfBroadcast Pattern by @bangtianliu in https://github.com/iree-org/iree/pull/22694 * Update gfx1250 LDS size by @Hardcode84 in https://github.com/iree-org/iree/pull/22760 * Bump version to 3.10 after 3.9 release. by @sa-faizal in https://github.com/iree-org/iree/pull/22759 * Integrate llvm/llvm-project@ebf5d9ef by @yzhang93 in https://github.com/iree-org/iree/pull/22761 * [Encoding] Implement compatibility check for packed_storage by @sommerlukas in https://github.com/iree-org/iree/pull/22757 * [Input] Register IREETensorExtDialect for Torch plugin by @IanWood1 in https://github.com/iree-org/iree/pull/22719 * Integrate LLVM @ 356479191ca0 by @lialan in https://github.com/iree-org/iree/pull/22772 * [Codegen][Tuner] Expose the python bindings for LinalgExt::inferScaledContractionDims and LinalgExt::isaScaledContractionOpInterface by @Muzammiluddin-Syed-ECE in https://github.com/iree-org/iree/pull/22763 * [tests][e2e] Add custom mxfp4 gemm tests to verify shape of interest. by @Muzammiluddin-Syed-ECE in https://github.com/iree-org/iree/pull/22775 * [Bazel] Migrate to bzlmod for LLVM compatibility by @maxbartel in https://github.com/iree-org/iree/pull/22771 * ScheduleExecution enhancements for timeline-aware scheduling and SCF. by @benvanik in https://github.com/iree-org/iree/pull/22483 * [Flow] Annotate scaled matmul dispatches by @Yu-Zhewen in https://github.com/iree-org/iree/pull/22773 * build_tools: fix: ensure that iree-flatcc-cli and iree-c-embed-data are build for the target during cross-compilation by @Manewing in https://github.com/iree-org/iree/pull/22755 * [Codegen][GPU] Add fusion barrier after result promotion by @qedawkins in https://github.com/iree-org/iree/pull/21709 * [CI] Update iree-org/iree-test-suites@132f91e4 by @efric in https://github.com/iree-org/iree/pull/22784 * [Codegen][GPU] Enable 3-stage pipelining with hipblaslt compute->write->read ordering by @jerryyin in https://github.com/iree-org/iree/pull/22788 * [LLVMGPU][Codegen] Reland "Emit packed chain FMA from select multi_reductions and contracts" by @efric in https://github.com/iree-org/iree/pull/22789 * Bump llvm-project to @a7c1f467339abd1942c89f2ef8b79083e89e7dad by @Max191 in https://github.com/iree-org/iree/pull/22787 * [Codegen] Support dynamic offsets in collapse_shape fusion to interface stores by @qedawkins in https://github.com/iree-org/iree/pull/22800 * [HAL] fix IREE_HAL_MAX_QUEUES to be number of bits in queue affinity type by @schuermans-roofline in https://github.com/iree-org/iree/pull/22702 * [Codegen][LLVMGPU] Replace TransposeSharedMem pipeline by @qedawkins in https://github.com/iree-org/iree/pull/21661 * Add iree_status_t stack trace support on Linux. by @benvanik in https://github.com/iree-org/iree/pull/22796 * [LLVMGPU] Fix lowering strategy for direct convolution by @yzhang93 in https://github.com/iree-org/iree/pull/22802 * [GPU] Add M dimension constraints for pingpong ukernel by @IanWood1 in https://github.com/iree-org/iree/pull/22801 * [Encoding] Remove unneeded command line option by @Muzammiluddin-Syed-ECE in https://github.com/iree-org/iree/pull/22816 * Add SCF support and fence coverage to ElideTimepointsPass. by @benvanik in https://github.com/iree-org/iree/pull/22611 * [GPU] Fix alignment check for scaled matmul by @Yu-Zhewen in https://github.com/iree-org/iree/pull/22737 * [Codegen][GPU] Replace prefetch_shared_memory with prefetch_num_stages in IREEGPUAttrs by @jerryyin in https://github.com/iree-org/iree/pull/22818 * [GlobalOpt] Fix rank-reduced permutation in SinkTransposeThroughExtractSlice by @ziliangzl in https://github.com/iree-org/iree/pull/22754 * [LLVMGPU] Update seeds for scaled gemm by @Muzammiluddin-Syed-ECE in https://github.com/iree-org/iree/pull/22798 * [Dispatch Creation] Don't fuse if there are no common parallel loops by @IanWood1 in https://github.com/iree-org/iree/pull/22819 * Reland "[LLVMGPU] Unroll elementwise operations [#21665]" by @lialan in https://github.com/iree-org/iree/pull/22828 * [Codegen][GPU] Preserve loop domain when collapsing dims in Conv to Matmul conversion by @yzhang93 in https://github.com/iree-org/iree/pull/22821 * Add jtuyls and Yu-Zhewen to CODEOWNERS for ROCM plugin and Encoding by @jtuyls in https://github.com/iree-org/iree/pull/22810 * [Encoding] Use struct directive for TestingAttr assembly format by @jtuyls in https://github.com/iree-org/iree/pull/22826 * Fix tests: `noubsan` was not being honored, and MXFP4 matmul tests are static-shape-only by @bjacob in https://github.com/iree-org/iree/pull/22836 * ElideAsyncCopiesPass refactoring for SCF/transfer support. by @benvanik in https://github.com/iree-org/iree/pull/22739 * [Runtime][HIP] Correct O(log n) bound search logic and eliminate O(n) loop(#22733) by @swote-git in https://github.com/iree-org/iree/pull/22734 * [e2e][ukernel] Remove dead/duplicate tests by @Yu-Zhewen in https://github.com/iree-org/iree/pull/22834 * [Codegen][GPU] Allow channel first layouts to lower through direct convolution path by @yzhang93 in https://github.com/iree-org/iree/pull/22840 * [Codegen] Add vector.to/from_elements to bf16 -> i16 conversion by @qedawkins in https://github.com/iree-org/iree/pull/22846 * Fix deadlock in `ROCMDialect::getMlirUKernels` by @bjacob in https://github.com/iree-org/iree/pull/22843 * [SPIRV][Codegen] Use single subgroup when reduction consumer has non-distributable broadcast by @efric in https://github.com/iree-org/iree/pull/22832 * [Encoding] Improve specialized encoding usage in lit test by @jtuyls in https://github.com/iree-org/iree/pull/22851 * [Stream] Encode packed_storage device and host tensors by @sommerlukas in https://github.com/iree-org/iree/pull/22722 * [Encoding] Add verifier for iree_encoding.layouts by @jtuyls in https://github.com/iree-org/iree/pull/22850 * [Codegen] Add WorkgroupCountHintOp to defer populating the workgroup count by @qedawkins in https://github.com/iree-org/iree/pull/22533 * Unify RISC-V toolchain environment variables and remove default path by @HanKuanChen in https://github.com/iree-org/iree/pull/22710 * [tests][e2e] Add more llama related shapes by @Muzammiluddin-Syed-ECE in https://github.com/iree-org/iree/pull/22831 * [Util] Implement InferIntDivisibilityOpInterface for affine ops by @Max191 in https://github.com/iree-org/iree/pull/22723 * Add CDNA3 test filtering to CI by @lialan in https://github.com/iree-org/iree/pull/22848 * [Torch Models] Fix SDXL golden dispatch counts by @IanWood1 in https://github.com/iree-org/iree/pull/22855 * [Dispatch Creation] Create more multi-use dispatches by @IanWood1 in https://github.com/iree-org/iree/pull/22011 * [LDS] Add AMDGPULowerCoalescedDMAToGatherLDS pass for direct global to LDS loads by @lialan in https://github.com/iree-org/iree/pull/22356 * Revert "[Util] Implement InferIntDivisibilityOpInterface for affine o… by @IanWood1 in https://github.com/iree-org/iree/pull/22860 * Integrates/llvm 20251208 by @bangtianliu in https://github.com/iree-org/iree/pull/22856 * [Codegen] Use workgroup_count_hint for most code paths by @qedawkins in https://github.com/iree-org/iree/pull/22549 * [Stream] Add UnifyEncodingForGlobals pass. 1/n by @hanhanW in https://github.com/iree-org/iree/pull/22767 * [Encoding] Add verifier for gpu/cpu/vmvx_encoding_resolver by @jtuyls in https://github.com/iree-org/iree/pull/22838 * [libcall] Update musl Makefile to include missing files. by @lialan in https://github.com/iree-org/iree/pull/22859 * Integrates/llvm 20251209 by @bangtianliu in https://github.com/iree-org/iree/pull/22864 * Bump the github-actions group across 1 directory with 3 updates by @dependabot[bot] in https://github.com/iree-org/iree/pull/22858 * [Codegen][GPU]Skip prologue pipeline barriers only for non-nested pipelined loops by @jerryyin in https://github.com/iree-org/iree/pull/22868 * [Pkgci] Update golden dispatch counts by @IanWood1 in https://github.com/iree-org/iree/pull/22869 * [DT][GPU] Redefine intrinsics M/N interleaving by @Yu-Zhewen in https://github.com/iree-org/iree/pull/22812 * [bindings] Add `iree_tensor_ext` to python bindings by @rkayaith in https://github.com/iree-org/iree/pull/22872 * [DT][SVE] adjust tile sizes for mmt4d & disable transposition of narrow-N matmuls by @egebeysel in https://github.com/iree-org/iree/pull/21701 * [Encoding] Add resolver swizzle verification by @jtuyls in https://github.com/iree-org/iree/pull/22867 * [Codegen] Do not swap extract_slice and collapse_shape for a special case by @yzhang93 in https://github.com/iree-org/iree/pull/22870 * Integrates/llvm 20251210 by @bangtianliu in https://github.com/iree-org/iree/pull/22876 * [LDS] Improve multiple transfers per lane by @lialan in https://github.com/iree-org/iree/pull/22879 * Integrates/llvm 20251211 by @bangtianliu in https://github.com/iree-org/iree/pull/22885 * [Stream] Handle AsyncCloneOp in UnifyEncodingForGlobals tracing. by @hanhanW in https://github.com/iree-org/iree/pull/22895 * Integrates/llvm 20251212 by @bangtianliu in https://github.com/iree-org/iree/pull/22897 * Update shark/SHARK to amd-shark/AMD-SHARK in documentation and URLs in IREE by @bangtianliu in https://github.com/iree-org/iree/pull/22883 * [Codegen] Add PCF dialect by @qedawkins in https://github.com/iree-org/iree/pull/22804 * Bump torch-mlir by @rkayaith in https://github.com/iree-org/iree/pull/22894 * [ROCM][DT] Add architecture matching to ukernel_info attribute by @Yu-Zhewen in https://github.com/iree-org/iree/pull/22899 * [Codegen] Add PCF bufferization interfaces by @qedawkins in https://github.com/iree-org/iree/pull/22805 * Bump to llvm/torch-mlir@3cebce2 by @zjgarvey in https://github.com/iree-org/iree/pull/22902 * [Stream] Update dispatch sites encodings in UnifyEncodingForGlobals pass 2/n by @hanhanW in https://github.com/iree-org/iree/pull/22886 * [LDS] Remove `GPULowerToGlobalLoadsPass` and `GlobalLoadDMAOp` by @lialan in https://github.com/iree-org/iree/pull/22862 * Integrate LLVM at b3ec8be by @bjacob in https://github.com/iree-org/iree/pull/22908 * [Dispatch Creation] Ignore unfusable consumer by @IanWood1 in https://github.com/iree-org/iree/pull/22874 * Add Codegen/Dialect CODEOWNERS entries for myself by @qedawkins in https://github.com/iree-org/iree/pull/22918 * [Stream] Handle tied operand result encodings in UnifyEncodingForGlobals 3/n by @hanhanW in https://github.com/iree-org/iree/pull/22891 * [Codegen] Drop certain encoding attributes after type propagation by @sommerlukas in https://github.com/iree-org/iree/pull/22740 * [DT][SVE] adjust tile sizes for unpack by @egebeysel in https://github.com/iree-org/iree/pull/21728 * [Stream] Selecting unified encodings from encoding resolvers. by @hanhanW in https://github.com/iree-org/iree/pull/22898 * [Encoding] Add EncodingProperties retrieval to SerializableAttr by @jtuyls in https://github.com/iree-org/iree/pull/22905 * [GlobalOpt] Fuse transpose into matmul-looking linalg.generic by @IanWood1 in https://github.com/iree-org/iree/pull/22901 * Bump the github-actions group with 3 updates by @dependabot[bot] in https://github.com/iree-org/iree/pull/22913 * Bump stablehlo to 6fabd27 by @kuhar in https://github.com/iree-org/iree/pull/22927 * [Encoding] Test i1 mask attention with packed_storage by @sommerlukas in https://github.com/iree-org/iree/pull/22756 * [bindings] Link CAPIAMDGPU into libIREECompiler by @tgymnich in https://github.com/iree-org/iree/pull/22929 * Integrate LLVM at 0729a74 by @bjacob in https://github.com/iree-org/iree/pull/22925 * [LinalgExt] Added support for LinalgExt::IndexOp(s) inside of attention regions by @keshavvinayak01 in https://github.com/iree-org/iree/pull/22768 * [CI][torch models] Increase acceptable tolerance in punet by @efric in https://github.com/iree-org/iree/pull/22935 * [DispatchCreation] Sink bit-extend producers into split-reduction `forall` loop by @rkayaith in https://github.com/iree-org/iree/pull/21850 * Integrate LLVM at 5f15fee by @bjacob in https://github.com/iree-org/iree/pull/22933 * [Stream] Prioritize NamedParameterAttr in UnifyEncodingForGlobals source tracing. 4/n by @hanhanW in https://github.com/iree-org/iree/pull/22922 * Bump torch-mlir to ac7b5f5 by @zjgarvey in https://github.com/iree-org/iree/pull/22934 * Add myself to CODEOWNERS for tests/e2e by @bjacob in https://github.com/iree-org/iree/pull/22944 * e2e matmul tests: fix some recent CDNA3-only tests by @bjacob in https://github.com/iree-org/iree/pull/22945 * [Input][Torch] Direct lowering of torch convolution_backward to linalg by @a-sidorova in https://github.com/iree-org/iree/pull/22904 * [CI][Torch Models] Tighten golden times by @efric in https://github.com/iree-org/iree/pull/22936 * [LinalgExt] Added DecomposeAggregatedOp Pass by @hhkit in https://github.com/iree-org/iree/pull/22779 * [LinalgExt] Added Decomposition of ExpReductionOp to LinAlg by @hhkit in https://github.com/iree-org/iree/pull/22561 * [Codegen][GPU] Improve lowering strategy for direct convolution by @yzhang93 in https://github.com/iree-org/iree/pull/22937 * Integrate LLVM at 1c023cb by @bjacob in https://github.com/iree-org/iree/pull/22943 * [GPU] Implement dummy getUnifiedEncoding method for GPU encoding resolver. by @hanhanW in https://github.com/iree-org/iree/pull/22923 * Integrate LLVM at 60e7c47 by @bjacob in https://github.com/iree-org/iree/pull/22952 * [CI] Update golden dispatch counts by @efric in https://github.com/iree-org/iree/pull/22950 * [CI][Torch Models] Update golden binary size by @efric in https://github.com/iree-org/iree/pull/22951 * Disable simplify regions after scf-to-cf by @nirvedhmeshram in https://github.com/iree-org/iree/pull/22946 * [GPU][Codegen] Expand iteration space based on new `expand_dims` attribute by @efric in https://github.com/iree-org/iree/pull/22342 * Revert "[GPU][Codegen] Expand iteration space based on new `expand_dims` attribute (#22342) by @efric in https://github.com/iree-org/iree/pull/22982 * Integrate LLVM at c0f4a8a by @jtuyls in https://github.com/iree-org/iree/pull/22979 * Integrate LLVM at 75a0347 by @jtuyls in https://github.com/iree-org/iree/pull/22985 * [Codegen][ROCDL] Improve dynamic dimension bounds handling in ROCDLConfigureBufferInstructions by @Yu-Zhewen in https://github.com/iree-org/iree/pull/22892 * [Codegen] add FoldExtractSliceOfBroadcast pattern to TileAndDistributeToWorkgroups by @bangtianliu in https://github.com/iree-org/iree/pull/22953 * [LDS] CoalescedGatherDMAOp support even more transfer sizes. by @lialan in https://github.com/iree-org/iree/pull/22890 * Integrate llvm-project/llvm@66c65f0132d05 by @kuhar in https://github.com/iree-org/iree/pull/22990 * [Codegen] Fix ConvertAccGemmToGemm to handle read-write arguments correctly. by @MaheshRavishankar in https://github.com/iree-org/iree/pull/22975 * [CI] Enable split dwarf and debug info in CI builds by @kuhar in https://github.com/iree-org/iree/pull/22993 * Bump dawidd6/action-download-artifact from 11 to 12 in the github-actions group by @dependabot[bot] in https://github.com/iree-org/iree/pull/22989 * [CI] Bump MacOS x64 version used in CI by @kuhar in https://github.com/iree-org/iree/pull/22991 * [CI] Reclaim disk space on macos arm runners by @kuhar in https://github.com/iree-org/iree/pull/22998 * [CI] Use `build_all.sh` in clang checks by @kuhar in https://github.com/iree-org/iree/pull/22999 * Integrate llvm/llvm-project@292c9e3d198249 by @kuhar in https://github.com/iree-org/iree/pull/23000 * [CI] Allow macos checks to be triggered with ci-extra by @kuhar in https://github.com/iree-org/iree/pull/23001 * [Codegen][ROCDL] Fix int64 overflow in `getSpannedBytes` by @Yu-Zhewen in https://github.com/iree-org/iree/pull/23002 * [CI] Drop macos job concurrency block by @kuhar in https://github.com/iree-org/iree/pull/23004 * Integrate llvm-project/llvm@0db04963d34be by @kuhar in https://github.com/iree-org/iree/pull/23006 * [CI] Do not run schedule-only jobs on integrates by @kuhar in https://github.com/iree-org/iree/pull/23009 * [TensorExt] Add Interface methods for lowering to loops and estimating bounds. by @MaheshRavishankar in https://github.com/iree-org/iree/pull/22268 * Integrate llvm/llvm-project@7976ac990000 by @kuhar in https://github.com/iree-org/iree/pull/23010 * Bump Torch-MLIR to [9a1f70] by @vivekkhandelwal1 in https://github.com/iree-org/iree/pull/22980 * [NFC] Adding iree_vm_bytecode_module_create flag enum arg. by @benvanik in https://github.com/iree-org/iree/pull/23017 * Support float type conversion between f32 and f64 in VM lists. by @willghatch in https://github.com/iree-org/iree/pull/22893 * Adding VMFB disassembly and cleaning up move handling. by @benvanik in https://github.com/iree-org/iree/pull/23018 * Fixing potential stack unwinding leak in VM. by @benvanik in https://github.com/iree-org/iree/pull/23019 * [AMDGPU][LDS] Unify DMA transfer linearization based on destination contiguity by @lialan in https://github.com/iree-org/iree/pull/22995 * [CI] Upgrade to GCC 11 to fix stablehlo template instantiation error by @Yu-Zhewen in https://github.com/iree-org/iree/pull/23008 * Un-packing VM ABI shims and adding proper alignment in C and C++ modules. by @benvanik in https://github.com/iree-org/iree/pull/22845 * Cleaning up move bit handling for disassembly/verifier/dispatch. by @benvanik in https://github.com/iree-org/iree/pull/23020 * Adding DiscardRefs and AssignRef VM ops. by @benvanik in https://github.com/iree-org/iree/pull/23021 * Adding --iree-vm-materialize-ref-discards pass. by @benvanik in https://github.com/iree-org/iree/pull/23026 * Integrate llvm/llvm-project@10a245bd02 by @Abhishek-Varma in https://github.com/iree-org/iree/pull/23022 * [Codegen] Add ConvertForallToLoopsPass by @qedawkins in https://github.com/iree-org/iree/pull/22911 * [AMDGPU][LDS] Prefer contiguous subviews in `GPUConvertToCoalescedDMA` by @lialan in https://github.com/iree-org/iree/pull/23011 * Adding new linear-scan register allocator. by @benvanik in https://github.com/iree-org/iree/pull/23027 * [CI] Add ci-extra flags for scheduled builds. by @amd-eochoalo in https://github.com/iree-org/iree/pull/23033 * [Codegen] Add FuseConsumersPass by @qedawkins in https://github.com/iree-org/iree/pull/22912 * [NFC] Cleanup of bytecode ISA decoding macros. by @benvanik in https://github.com/iree-org/iree/pull/23042 * [Codegen] Add FusePCFWritesPass by @qedawkins in https://github.com/iree-org/iree/pull/22914 * Adding iree_vm_list_get_ref_retain_or_move. by @benvanik in https://github.com/iree-org/iree/pull/23043