Arun Thangamani 8e18cdcb15

[mlir][x86vector] AVX10 I8 Dot Op (#178807) ...

Adds AVX10 i8 dot-product operation and defines lowering to LLVM
intrinsics.

Target assembly instruction: `llvm.x86.avx10.vpdpbssd.512`

2026-02-02 16:49:00 +05:30

..

cvt-packed-f32-to-bf16.mlir

…

dot-bf16.mlir

…

legalize-for-llvm.mlir

[mlir][x86vector] AVX10 I8 Dot Op (#178807)

2026-02-02 16:49:00 +05:30

roundtrip.mlir

[mlir][x86vector] AVX10 I8 Dot Op (#178807)

2026-02-02 16:49:00 +05:30

shuffle-vector-fmas.mlir

[mlir][x86vector] Shuffle FMAs (#172823)

2026-01-20 17:21:26 +05:30

sink-vector-producer-ops.mlir

[mlir][x86vector] Sink Vector.transfer_reads and vector.load before the consumer (#169333)

2025-12-11 16:59:55 +00:00

vector-contract-bf16-to-fma.mlir

[mlir][x86vector] Lower BF16 vector.contract to FMA using AVX2 BF16 packed ops. (#170267)

2025-12-17 14:41:58 +01:00

vector-contract-to-fma.mlir

[mlir][x86vector] Lower vector.contract to FMA or packed type dot-product (#168074)

2025-11-24 06:33:07 +01:00

vector-contract-to-packed-type-dotproduct.mlir

[mlir][x86vector] AVX10 I8 Dot Op (#178807)

2026-02-02 16:49:00 +05:30