Test Cortex-M Backend

test-backend-cortex-m.yml

2500+ workflow runs

Fix export of zero-stride constants and inputs, and of constants that delegates share
completed successfully:
Test Cortex-M Backend#10665:Pull request#24404synchronizebyshoumikhin
shoumikhin:constprop-zero-stride
Nov 3, 202610s
exir and MLX: fix folded-output clones, shared constants, and constant boolean masks
completed successfully:
Test Cortex-M Backend#10664:Pull request#24481synchronizebyshoumikhin
shoumikhin:exir-mlx-constant-fixes
Nov 3, 20269s
Avoid copying the whole file when reading .pte and .ptd headers
requires action with the application:
Test Cortex-M Backend#10663:Pull request#23561synchronizebyHiroForYou
HiroForYou:avoid-header-copy
Nov 3, 2026Action required
Keep max_pool2d and avg_pool2d test inputs within the exported shape …
completed successfully:
Test Cortex-M Backend#10662:Commit2799a79pushedbymeta-codesync[bot]Bot
main
Nov 3, 202614m 25s
Reinstall pip from PyPI in the native wheel pre-build script (#24461)
completed successfully:
Test Cortex-M Backend#10661:Commit6d903f4pushedbyJakeStevens
main
Nov 3, 202615m 19s
[executorch][native] Judge activation-quantized engines against an int8-activation golden
completed successfully:
Test Cortex-M Backend#10660:Pull request#24362synchronizebydigantdesai
gh/digantdesai/189/head
Nov 3, 202614s
[executorch][vulkan] Fix dq8ca q4gsw weight sums stride when N % 8 is 1 to 4
completed successfully:
Test Cortex-M Backend#10659:Pull request#24370synchronizebydigantdesai
gh/digantdesai/197/head
Nov 3, 202619s
[executorch][native] Pin eager kernels when generating op samples
completed successfully:
Test Cortex-M Backend#10658:Pull request#24358synchronizebydigantdesai
gh/digantdesai/185/head
Nov 3, 202614s
[executorch][vulkan] Round f16 image stores to nearest even
completed successfully:
Test Cortex-M Backend#10657:Pull request#24372synchronizebydigantdesai
gh/digantdesai/199/head
Nov 3, 202615s
[executorch][native-ggml] Reject weight-only quantized linears unless activation quant is allowed
completed successfully:
Test Cortex-M Backend#10656:Pull request#24361synchronizebydigantdesai
gh/digantdesai/188/head
Nov 3, 202612s
[executorch][native-vk] Note the int4 weight-only N % 4 and rank-4 linear gaps
completed successfully:
Test Cortex-M Backend#10655:Pull request#24371synchronizebydigantdesai
gh/digantdesai/198/head
Nov 3, 202613s
[executorch][native] Record op packages with bundled golden and precise samples
completed successfully:
Test Cortex-M Backend#10654:Pull request#24357synchronizebydigantdesai
gh/digantdesai/184/head
Nov 3, 202624s
[executorch][native] Add GGUF q4_k and q6_k packages to the op samples
completed successfully:
Test Cortex-M Backend#10653:Pull request#24366synchronizebydigantdesai
gh/digantdesai/193/head
Nov 3, 202613s
[executorch][native-ggml] Compute cuBLAS GEMMs in f32 in the CUDA sample tests
completed successfully:
Test Cortex-M Backend#10652:Pull request#24363synchronizebydigantdesai
gh/digantdesai/190/head
Nov 3, 202610s
[executorch][native] Split engine host factories out of the PTN runners
completed successfully:
Test Cortex-M Backend#10651:Pull request#24356synchronizebydigantdesai
gh/digantdesai/183/head
Nov 3, 202617s
[executorch][native] Cover negative, rounded and subnormal GGUF super-block scales in the op samples
completed successfully:
Test Cortex-M Backend#10650:Pull request#24368synchronizebydigantdesai
gh/digantdesai/195/head
Nov 3, 202612s
[executorch][native] Suggest rounding to nearest when engines narrow values
completed successfully:
Test Cortex-M Backend#10649:Pull request#24364synchronizebydigantdesai
gh/digantdesai/191/head
Nov 3, 202613s
[executorch][native] Run the ggml CPU act_quant sample test in CI
completed successfully:
Test Cortex-M Backend#10648:Pull request#24373synchronizebydigantdesai
gh/digantdesai/200/head
Nov 3, 202612s
[executorch][native] Decode quantized weights in float32 for precise samples
completed successfully:
Test Cortex-M Backend#10647:Pull request#24367synchronizebydigantdesai
gh/digantdesai/194/head
Nov 3, 202613s
[executorch][test-suite] Feed the Linear tests inputs with the model's 67 input features
completed successfully:
Test Cortex-M Backend#10646:Pull request#24355synchronizebydigantdesai
gh/digantdesai/182/head
Nov 3, 202615m 55s
[executorch][native] Run engines on bundled op samples with gtest
completed successfully:
Test Cortex-M Backend#10645:Pull request#24360synchronizebydigantdesai
gh/digantdesai/187/head
Nov 3, 202623s
[executorch][vulkan] Accept 1-D activations in 8da4w linear
completed successfully:
Test Cortex-M Backend#10644:Pull request#24369synchronizebydigantdesai
gh/digantdesai/196/head
Nov 3, 202611s
[executorch][native] Record dynamic-shape linears as chained bundled samples
completed successfully:
Test Cortex-M Backend#10643:Pull request#24405synchronizebydigantdesai
gh/digantdesai/201/head
Nov 3, 202611s
[executorch][native] Judge engine outputs against golden and precise samples
completed successfully:
Test Cortex-M Backend#10642:Pull request#24359synchronizebydigantdesai
gh/digantdesai/186/head
Nov 3, 202610s
[executorch][native] Add FACTO-drawn linear cases to the op samples
completed successfully:
Test Cortex-M Backend#10641:Pull request#24365synchronizebydigantdesai
gh/digantdesai/192/head
Nov 3, 202613s