NXP Backend: Adding the loader for eIQ Neutron SDK converter module.#16326

Closed

jirioc wants to merge 1243 commits intopytorch:mainfrom

nxp-upstream:nxf96559/feature/EIEX-657-implement-loading-mechanism-for-eiq-neutron-sdk-converter-module

This pull request is big! We're only showing the most recent 250 commits

Commits on Dec 19, 2025

Arm backend: Add docstrings to compile specs (#15886)

Sebastian-Larsson
authored and
jirioc
committed
Arm backend: Make PassPipeline test valid inputs (#15892)

oscarandersson8218
authored and
jirioc
committed
Update ReplaceSqueezeAndUnsqueezeWithViewPass to use new pass interface

DrJessop
authored and
jirioc
committed
Security: Update glob to 10.5.0 to fix CVE-2025-64756 (#15904)

abhinaykukkadapu
authored and
jirioc
committed
Fix circular dep in flat tensor import (#15902)

lucylq
authored and
jirioc
committed
Arm backend: Support quantized cond and while (#15849)

Erik-Lundell
authored and
jirioc
committed
Arm backend: Rename passes for consistency (#15888)

martinlsm
authored and
jirioc
committed
Arm backend: Add support for ubsan in executor_runner (#15894)

perheld
authored and
jirioc
committed
Cortex_M backend: Add conv op (#15896)

AdrianLundell
authored and
jirioc
committed
Arm backend: Update pass name in arm_aot_compiler (#15916)

AdrianLundell
authored and
jirioc
committed
Remove no-op clones in xnnpack (#15884)

GregoryComer
authored and
jirioc
committed
Fix UnboundLocalError: local variable 'qnn_compile_spec_buffer' refer… (#15881)

cccclai
authored and
jirioc
committed
Fix extract model script (#15924)

metascroy
authored and
jirioc
committed
Arm backend: Fix get_module_name_filter (#15910)

oscarandersson8218
authored and
jirioc
committed
Arm backend: Add docstrings to vgf/backend.py and ethosu/backend.py (#15887)

authored and
jirioc
committed
Arm backend: Move support_extension to base class (#15909)

oscarandersson8218
authored and
jirioc
committed
Arm backend: test without output re-order workaround (#15826)

authored and
jirioc
committed
Fix Neutron backend API compatibility and multiprocessing fallback

apullin
authored and
jirioc
committed
Refactor download function for robustness and retries (#15933)

cccclai
authored and
jirioc
committed
[aoti-backend-consolidation 2/3] backend.py (#15528)

authored and
jirioc
committed
add cuda benchmark ci (#15883)

Gasoonjia
authored and
jirioc
committed
Add bundled IO support to portable executor runner (#15368)

authored and
jirioc
committed
Rename the qnn demo backend (#15930)

cccclai
authored and
jirioc
committed
Quantize lora linears (#15935)

lucylq
authored and
jirioc
committed
Add mypy ignore for snakeviz (#15806)

authored and
jirioc
committed
Tag scales for external data

lucylq
authored and
jirioc
committed
Save external constant tensors to custom filename

lucylq
authored and
jirioc
committed
make partial export failure not stop cuda benchmark (#15939)

Gasoonjia
authored and
jirioc
committed
Remove download progress reporting (#15946)

cccclai
authored and
jirioc
committed
Fix build isolation and editable install issues (#15941)

larryliu0820
authored and
jirioc
committed
Create a button to manually trigger all cuda benchmarks (#15938)

Gasoonjia
authored and
jirioc
committed
only download benchmark data for further uploading (#15945)

Gasoonjia
authored and
jirioc
committed
Remove AWS Device Farm benchmark jobs (#15433)

GregoryComer
authored and
jirioc
committed
Disable QNN build option for the default x86 build (#15949)

cccclai
authored and
jirioc
committed
Expose get_num_threads via pybind

kimishpatel
authored and
jirioc
committed
Move user-defined passes after SpecPropPass

lucylq
authored and
jirioc
committed
Arm backend: Add missing default case in executor_runner (#15895)

perheld
authored and
jirioc
committed
Arm backend: Minimal example of pruning (#15851)

gggekov
authored and
jirioc
committed
Arm backend: Update bias quantization range (#15918)

oscarandersson8218
authored and
jirioc
committed
Added more users to exclude (#15781)

nil-is-all
authored and
jirioc
committed
forward fix PR 15368 (#15964)

cccclai
authored and
jirioc
committed
fix eval_llama_qnn custom annotation (#15953)

DannyYuyang-quic
authored and
jirioc
committed
add 12.9 cuda support (#15818)

JacobSzwejbka
authored and
jirioc
committed
Adding compile option warmup_execute_after_compile to optionally run execute rigtht after compilation to create command buffers.

trivedivivek
authored and
jirioc
committed
Fix cpp/c compilation warnings.

hsharma35
authored and
jirioc
committed
Add ETDump event tracer support to LLaMa runner

navsud
authored and
jirioc
committed
Qualcomm AI Engine Direct - GLM1.5B (#15691)

winskuo-quic
authored and
jirioc
committed
Fix relu test + use xt macros.

hsharma35
authored and
jirioc
committed
Cortex-M backend: Fuse Relu, Hardtanh and Hardsigmoid (#15917)

AdrianLundell
authored and
jirioc
committed
Switch to pnpm for dependency management (#16019)

mergennachin
authored and
jirioc
committed
Update ReplaceAddMMWithLinearPass to use new pass interface

DrJessop
authored and
jirioc
committed
Update QCOM llama hardware support (#15965)

cccclai
authored and
jirioc
committed
Add 16a4w_block QAT config

sxu
authored and
jirioc
committed
Do not include cortex_m_lib for unit tests. Include only specific ops (#16027)

psiddh
authored and
jirioc
committed
custom comparator support in calculate_numeric_gap

Gasoonjia
authored and
jirioc
committed
Fix index out of bounds error for 1d convs

DrJessop
authored and
jirioc
committed
Use 4x3 tiled shader for linear mat mul which performs slightly better.

trivedivivek
authored and
jirioc
committed
Add CentOS to the supported host OS (#16021)

cccclai
authored and
jirioc
committed
QNN Backend: Enable QualComm SM8850 targets (#15987)

authored and
jirioc
committed
Qualcomm AI Engine Direct - Remove legacy code related to the shared buffer (#16000)

shewu-quic
authored and
jirioc
committed
Qualcomm AI Engine Direct - Improve CLI tools (#16012)

chenweng-quic
authored and
jirioc
committed
More Passes updated to use new interface

DrJessop
authored and
jirioc
committed
Qualcomm AI Engine Direct - Support triu op and linear op with non-constant weights (#16014)

shewu-quic
authored and
jirioc
committed
Fix -Wimplicit-const-int-float-conversion error.

hsharma35
authored and
jirioc
committed
Add generic operator implementations.

hsharma35
authored and
jirioc
committed
Rename generic operator cpp/headers.

hsharma35
authored and
jirioc
committed
use *args and **kwargs as signature of monkey patched function

Gasoonjia
authored and
jirioc
committed
Arm backend: Add test for BatchNorm1D QAT folding (#16001)

oscarandersson8218
authored and
jirioc
committed
Arm backend: Add int16x8 LayerNorm test cases (#16015)

SaoirseARM
authored and
jirioc
committed
Arm backend: Update output reorder workaround (#15981)

authored and
jirioc
committed
Call prepare_traced_pt2 instead of prepare_pt2 in get_fake_quant_model

ethansfng
authored and
jirioc
committed
Arm backend: Align docstrings in arm_quantizer.py with backend standard (#15975)

authored and
jirioc
committed
Some minor performance improvements to buffer 4b mat mul.

trivedivivek
authored and
jirioc
committed
Adding a payload version to become compatible with extra output tensors in 2.2.1
jirioc
committed
Arm backend: Add missing bool case in prepare_input_tensors

perheld
authored and
jirioc
committed
Arm backend: Pass location as string instead of TosaOpLocation (#15994)

tom-arm
authored and
jirioc
committed
Cortex_M backend: Fuse clamp + hardswish decompostion (#16016)

AdrianLundell
authored and
jirioc
committed
Arm backend: Add support for int32 clamp (#15977)

ArmRyan
authored and
jirioc
committed
docs: Add logging guidance for debug partition (#15979)

yujiaoliang
authored and
jirioc
committed
make triton kernel usage user controlable

Gasoonjia
authored and
jirioc
committed
Remove sharded ckpt from export_llama (#15968)

lucylq
authored and
jirioc
committed
Make ExecuTorch CoreML asset/trash/databse folders configurable.

metascroy
authored and
jirioc
committed
Using buffer for weight tensors for quantized mat mul op.

trivedivivek
authored and
jirioc
committed
Remove fairseq from export_llama

lucylq
authored and
jirioc
committed
More passes updated to be more efficient and correctly set their modified bit

DrJessop
authored and
jirioc
committed
Arm Backend: improve non-persistent placeholder and bool handling (#15992)

authored and
jirioc
committed
Cortex_M backend: Remove int8 requirement for zero_points (#16061)

AdrianLundell
authored and
jirioc
committed
Arm backend: Fixes and ignores for bandit

perheld
authored and
jirioc
committed
NXP Backend: Add pass to remove unnecessary Quantize/Dequantize nodes. (#15148)

MartinPavella
authored and
jirioc
committed
Fix fake mode detection

angelayi
authored and
jirioc
committed
[OpenVINO] Fix Quantizer for PTQ (#15891)

anzr299
authored and
jirioc
committed
Using generic implementation for 16-bit activations and 8 bit weights for linear in backends

RahulC7
authored and
jirioc
committed
[etrecord] Implement generic fallback for `GraphModuleSerializer.handle_call_function` (#16069)

pytorchbot
authored and
jirioc
committed
Add Run Android LlamaDemo with QNN backend (#16011)

authored and
jirioc
committed
Using generic implemntation for 16-bit activations and 8 bit weights for Conv2D in Backends

RahulC7
authored and
jirioc
committed
Arm backend: Add support for quant. decomposition (#15993)

oscarandersson8218
authored and
jirioc
committed
Arm backend: Add support for asan in executor_runner (#16009)

perheld
authored and
jirioc
committed
Arm backend: Use %zu instead of %lu (#15974)

gggekov
authored and
jirioc
committed
Arm backend: Make SupportedTOSAOperatorChecks work for INT+FP (#16072)

authored and
jirioc
committed
Arm backend: Improve dtype validation (#15871)

oscarandersson8218
authored and
jirioc
committed
Arm backend: Make sure DW-conv weights are reshaped once (#16071)

oscarandersson8218
authored and
jirioc
committed
Arm backend: Improve quantizer configuration in tests (#16073)

Erik-Lundell
authored and
jirioc
committed
Arm Backend: Add support for select_scatter.default (#15972)

agrima1304
authored and
jirioc
committed
Arm backend: Move down a few passes to after fold Q/DQ pass (#16035)

martinlsm
authored and
jirioc
committed
Cortex_M backend: Remove scalar-scalar tests (#16078)

oscarandersson8218
authored and
jirioc
committed
Add view_copy/static_reshape support to XNNPACK delegate (#7959)

GregoryComer
authored and
jirioc
committed
Arm backend: Move FuseConstantArgsPass (#16076)

oscarandersson8218
authored and
jirioc
committed
[ET-VK][ez] Address regressed conv2d perf numbers on main (#16084)

pytorchbot
authored and
jirioc
committed
grid_sampler_2d_out Portable Kernel Implementation (#16051)

authored and
jirioc
committed
Using generic implementation for 16-bit activations and 8-bit weights for matmul in backends

RahulC7
authored and
jirioc
committed
Fix OSS import for ref_implementations

DrJessop
authored and
jirioc
committed
[ET-VK] Enable and test texture IO for quantized convolution ops (#16082)

authored and
jirioc
committed
Support eliminate_quant_dequant_pairs flag.

eigen-k
authored and
jirioc
committed
Update more replace ops passes to new interface

DrJessop
authored and
jirioc
committed
Fix duplication bug in serialize_for_executorch

lucylq
authored and
jirioc
committed
Arm backend: Replace ±inf and FP limit values with ±255.0 (#15976)

YufengShi-dudu
authored and
jirioc
committed
Re-land cve security fix (#16103)

lucylq
authored and
jirioc
committed
Fix data pointer const-correctness in executorch kernels

maxjing
authored and
jirioc
committed
Using push constants for softmax op.

trivedivivek
authored and
jirioc
committed
Arm backend: Update epsilon to correct default of 2**-12 (#16096)

SaoirseARM
authored and
jirioc
committed
Arm backend: Increase tolerance on Tanh int16 test (#16094)

SaoirseARM
authored and
jirioc
committed
Arm backend: Rename build_reshape_tosa_1_0 to build_reshape_tosa (#16095)

Sebastian-Larsson
authored and
jirioc
committed
Arm backend: Handle all types in prepare_input_tensors() (#16092)

perheld
authored and
jirioc
committed
Reduce allocation overhead in quantized sdpa (#16119)

authored and
jirioc
committed
[Executorch] Introduce caching cpu memory allocator (#16120)

authored and
jirioc
committed
Using push constants for clone op parameters.

trivedivivek
authored and
jirioc
committed
[Executorch] Use temp allocator for allocating scratch memory (#16121)

authored and
jirioc
committed
[Executorch] Make module constructors uniform across (#16122)

authored and
jirioc
committed
NXP backend: Improve `view_copy` delegation (#15270)

MartinPavella
authored and
jirioc
committed
Support negative dimensions on "aten.split_with_sizes_copy.default" (#16006)

jgibson2
authored and
jirioc
committed
[Executorch] make slice_copy parallel (#16125)

pytorchbot
authored and
jirioc
committed
Ensure modified flag is returned correctly for RemovePermutesAroundElementwiseOps.

eigen-k
authored and
jirioc
committed
Metal backend: SDPA metal implementation (#16086)

manuelcandales
authored and
jirioc
committed
Add bitwise left/right (#15893)

mergennachin
authored and
jirioc
committed
bump arm cortex size test (#16134)

lucylq
authored and
jirioc
committed
reorganize VOLK_IMPLEMENTATION usage

SamGondelman
authored and
jirioc
committed
Using push constants for binary scalar op parameter.

trivedivivek
authored and
jirioc
committed
support skip atten in export

jackzhxng
authored and
jirioc
committed
add ET_BUNDLE_IO_ENABLED in executor runner

zonglinpeng
authored and
jirioc
committed
Fix double-tracing in SpecPropPass (#15485)

GregoryComer
authored and
jirioc
committed
Qualcomm AI Engine Direct - support SW6100 (#16003)

authored and
jirioc
committed
Remove op_linear_scratch_example

lucylq
authored and
jirioc
committed
Adding Test for CadenceWith16BitMatmulActivationsQuantizer

RahulC7
authored and
jirioc
committed
Enforce src and dst do not overlap in memcpy

lucylq
authored and
jirioc
committed
Add RmsNormNopQuantizer and Pattern

ethansfng
authored and
jirioc
committed
Using push constants for add mm native op.

trivedivivek
authored and
jirioc
committed
Handle same tensor appearing multiple times in the cat input

ethansfng
authored and
jirioc
committed
Migrate more generic cadence operators to oss.

hsharma35
authored and
jirioc
committed
Decompose after export in export_llama

lucylq
authored and
jirioc
committed
NXP backend: added aten.mul support (#15971)

authored and
jirioc
committed
Arm backend: Add U55 and U85 tests for deit_tiny (#16145)

oscarandersson8218
authored and
jirioc
committed
NXP backend: added aten.slice support (#15889)

novak-vaclav
authored and
jirioc
committed
Fix ReplaceConvolutionOptionalArgsWithConcreteArgsPass (#16143)

authored and
jirioc
committed
[ET-VK][ez] Improve quantized convolution test case naming (#16141)

authored and
jirioc
committed
Annotate oncall for frequent signals

Ben0mega
authored and
jirioc
committed
Prevent signed integer overflow in pixel_shuffle size calculation (#16138)

GregoryComer
authored and
jirioc
committed
Rm platform args in executorch

jaejunku
authored and
jirioc
committed
Qualcomm AI Engine Direct - gpu support part1 (#12165)

authored and
jirioc
committed
Fix CoreML iOS26 numerics in static attention (#16144)

metascroy
authored and
jirioc
committed
Fix rope ref implementation.

hsharma35
authored and
jirioc
committed
Modify depthwise int8 conv2d to reduce register/memory pressure

GregoryComer
authored and
jirioc
committed
Implement guardrail to check for existing branches (#16132)

authored and
jirioc
committed
Cortex_m backend: Support channels-broadcasting for ADD/MUL (#16131)

AdrianLundell
authored and
jirioc
committed
Arm backend: Make INT+FP default for vgf-backend (#16176)

oscarandersson8218
authored and
jirioc
committed
Arm backend: Add Quantization test pipeline (#16151)

oscarandersson8218
authored and
jirioc
committed
Qualcomm AI Engine Direct - Add MHA2SHA pass (#15438)

shewu-quic
authored and
jirioc
committed
Qualcomm AI Engine Direct - enable operator max_pool3d by decomposition (#15897)

jethroqti
authored and
jirioc
committed
Adding Test To Ensure All Future Quantizers Are Tested

RahulC7
authored and
jirioc
committed
Fix off-by-on error

lucylq
authored and
jirioc
committed
Add lora test using qwen (#16161)

lucylq
authored and
jirioc
committed
Add test for CadenceRmsNormNopQuantizer

ethansfng
authored and
jirioc
committed
Expose static llama in OSS

metascroy
authored and
jirioc
committed
Migrate generic cadence operators to oss.

hsharma35
authored and
jirioc
committed
NXP backend: Add QAT support for NeutronQuantizer (#15692)

StrycekSimon
authored and
jirioc
committed
Arm backend: Add function to help linker force include backend (#16177)

authored and
jirioc
committed
Update replace ops to correctly set modified bit

DrJessop
authored and
jirioc
committed
Qwen lora quantize test (#16188)

lucylq
authored and
jirioc
committed
Scan support (#16028)

JacobSzwejbka
authored and
jirioc
committed
Add null pointer check for evalues (#14749)

rascani
authored and
jirioc
committed
Minor Refactor of ExecutorchRuntimeException (#16193)

authored and
jirioc
committed
Summary: Add context when LLMModule load fails to aid in debugging (#16174)

authored and
jirioc
committed
[ez][ET-VK] Small fix for choose_qparams_affine_impl (#16186)

authored and
jirioc
committed
[ez][ET-VK] Update Vulkan runtime application name (#16187)

authored and
jirioc
committed
Update torch pin to 11-20 (#16198)

mergennachin
authored and
jirioc
committed
Fix tests after torch pin PR (#16209)

mergennachin
authored and
jirioc
committed
Fix test-torchao-huggingface-checkpoints (#16211)

mergennachin
authored and
jirioc
committed
Add resume for download function (#15948)

cccclai
authored and
jirioc
committed
Increase docker time limit (#16212)

mergennachin
authored and
jirioc
committed
StaticAttetnion runtime support for generate_full_logits=False

sxu
authored and
jirioc
committed
Remove internal dependencies for cadence ops.

hsharma35
authored and
jirioc
committed
optimum pin bump

JacobSzwejbka
authored and
jirioc
committed
Revert "Add resume for download function" (#16215)

cccclai
authored and
jirioc
committed
Update trunk.yml (#16216)

metascroy
authored and
jirioc
committed
Update examples/models/qwen2_5/README.md (#16150)

DamonFool
authored and
jirioc
committed
Update more ops to use the new pass interface, update some ref implementations

DrJessop
authored and
jirioc
committed
Portable aoti (#15967)

JacobSzwejbka
authored and
jirioc
committed
NXP Backend: Update documentation to the new scheme (#15219)

robert-kalmar
authored and
jirioc
committed
Arm backend: Fix broken index_tensor tests (#16220)

oscarandersson8218
authored and
jirioc
committed
Arm backend: Relax pre-push header checks (#16195)

Sebastian-Larsson
authored and
jirioc
committed
Update ReplaceConvWithChannelLastConvPass and MakeSliceAndCatDimOutermostPass to correctly set modified bit

DrJessop
authored and
jirioc
committed
Arm backend: Add conv3d support to Tosa/Vgf backends (#16093)

ArmRyan
authored and
jirioc
committed
Arm backend: Refactor pass skipping mechanisms (#16153)

ArmRyan
authored and
jirioc
committed
Arm backend: Update vgf model tests (#16199)

oscarandersson8218
authored and
jirioc
committed
Add Python 3.13 support (#16004)

PetarTerziev-UL
authored and
jirioc
committed
Replace duplicated RemoveCloneOpPass with RemoveCloneOpsTransformImported.

eigen-k
authored and
jirioc
committed
Temporarily disable view-copy partitioning in XNNPACK (#16226)

GregoryComer
authored and
jirioc
committed
Add ability to specify CoreML pipeline passes (#16118)

authored and
jirioc
committed
Upgrade Docker container from gcc9 to gcc11 and add gcc9-nopytorch variant (#16227)

authored and
jirioc
committed
Arm backend: Update vgf ops tests [Part 1] (#16245)

oscarandersson8218
authored and
jirioc
committed
NXP backend: Add dim order support to NeutronBackend. (#16146)

MartinPavella
authored and
jirioc
committed
Arm backend: Fix broken SD3Transformer2DModel test (#16247)

oscarandersson8218
authored and
jirioc
committed
Fix typo in tutorial for exporting llm (#15906)

liu-mengyang
authored and
jirioc
committed
Merging gh/abhinaykukkadapu/15/base to main (#16252)

authored and
jirioc
committed
Remove unused exception parameter from executorch/backends/vulkan/test/custom_ops/utils.cpp (#16253)

authored and
jirioc
committed
Fix fbcode/oss sync (#16254)

manuelcandales
authored and
jirioc
committed
Qualcomm AI Engine Direct - Merge the two pybind libraries into a single library (#15999)

shewu-quic
authored and
jirioc
committed
Arm backend: Update vgf ops tests [Part 2] (#16251)

oscarandersson8218
authored and
jirioc
committed
Append to cflag and honor CMAKE_POSITION_INDEPENDENT_CODE OFF (#16074)

zingo
authored and
jirioc
committed
Update supported Python versions in windows build (#16232)

kirklandsign
authored and
jirioc
committed
Fix slice quant pattern match bug (#15624)

jackzhxng
authored and
jirioc
committed
More robust PTAX finding (#16229)

JacobSzwejbka
authored and
jirioc
committed
Support cosine operator on XNNPACK (#15431)

GregoryComer
authored and
jirioc
committed
Generalize quantized input wrapper

ethansfng
authored and
jirioc
committed
Update staging buffer allocation flags by transfer direction

GregoryComer
authored and
jirioc
committed
Arm backend: Test partial quantization of models (#16249)

authored and
jirioc
committed
Arm backend: Match arg ranks for min/max ops (#16181)

authored and
jirioc
committed
Arm backend: Add support for masked_fill_.Scalar (#16272)

oscarandersson8218
authored and
jirioc
committed
Arm backend: Fix two control flow bugs (#16274)

Erik-Lundell
authored and
jirioc
committed
Rename ethous-u-scratch to arm-scratch (#16224)

mansnils
authored and
jirioc
committed
Cover tokenizer build when running install_executorch.sh --clean (#16264)

cccclai
authored and
jirioc
committed
[ET-VK] Pass detailed op information to event tracer (#16266)

authored and
jirioc
committed
[ET-VK][test-utils] Implement submodule extraction utilities (#16267)

authored and
jirioc
committed
Add Python 3.13 to windows wheel build CI (#16284)

larryliu0820
authored and
jirioc
committed
Qualcomm AI Engine Direct - QNN ExecuTorch Intermediate Output Debugger (#15735)

winskuo-quic
authored and
jirioc
committed
Qualcomm AI Engine Direct - Refactor QNN Manager (#15903)

shewu-quic
authored and
jirioc
committed
Arm backend: Add processed module to global scope in aot_arm_compiler (#16275)

Erik-Lundell
authored and
jirioc
committed
Arm backend: Add T5 model tests (#16278)

YufengShi-dudu
authored and
jirioc
committed
Lint fix: qnnpreprocess (#16295)

manuelcandales
authored and
jirioc
committed
Fix openvino torchao build (#16296)

manuelcandales
authored and
jirioc
committed
Arm backend: Add support for clamp.Tensor (#16273)

authored and
jirioc
committed
Arm backend: Improve while support (#16287)

Erik-Lundell
authored and
jirioc
committed
Arm backend: Add TOSA-dtype validation to tester (#16293)

oscarandersson8218
authored and
jirioc
committed
Arm backend: Disable relu-fusing for sym. quant (#16286)

oscarandersson8218
authored and
jirioc
committed
Add broadcast semantics to quantized ops.

hsharma35
authored and
jirioc
committed
Fix a typo on build-wheels-windows.yml (#16300)

larryliu0820
authored and
jirioc
committed
Clean up various leftover platform arg usage around the repo that never hit validation

8Keep
authored and
jirioc
committed
Add 'docker-builds' to viable/strict requirement (#16282)

kirklandsign
authored and
jirioc
committed
forward fix

manuelcandales
authored and
jirioc
committed
triton sdpa kernel with better perf (#16167)

Gasoonjia
authored and
jirioc
committed
NXP Backend: Adding the loader for eIQ Neutron SDK converter module.
jirioc
committed