3 people 77e124e455 fix AMD WMMA emulation and test in CI (#17184) ...

* fix SPEC=1 test_tensor_cores

* implement i32 WMMA for RDNA3, add regression test to CI

* gfx950 scaled mfma llvmir fix

* detect VOP3PX2 in emu

* start cdna4 scaled mfam emu

* fix gfx12 llvmir signatures

* oops

* fix

* fix src2 const field extraction (gpt)

* scaled mfma fixes cdna4 (kimi)

* fp8 out breaks mfma (glm)

* fix const signature

* another

* fix f string for linter

* lint

* clean

* and a final lint

* .

* fix mypy

* skip slow tests on ci

* reduce unroll tensor shape -> 64x64

---------

Co-authored-by: George Hotz <[email protected]>

2026-08-04 23:25:32 -07:00

..

am

mockgpu: use autogen.libc (#16479)

2026-06-02 19:59:36 -04:00

amd

fix AMD WMMA emulation and test in CI (#17184)

2026-08-04 23:25:32 -07:00

cuda

ocelot: use c.DLL (#16540)

2026-06-08 21:27:28 -04:00

nv

64-bit UOp.variable support, try 2 (#17256)

2026-07-29 14:39:36 -04:00

driver.py

hw interface abstraction (#8524)

2025-01-07 18:18:28 +03:00

gpu.py

Very basic mock sqtt (#10512)

2025-05-26 14:38:28 -07:00

helpers.py

ocelot: use c.DLL (#16540)

2026-06-08 21:27:28 -04:00

mockgpu.py

llm: make tokenizer fast (kimi) (#17077)

2026-07-18 17:31:59 -07:00

usb.py

usb: keep only custom (#17029)

2026-07-15 22:40:52 +03:00