chenyu and GitHub
086f45cb81
dtype is not a UOp field anymore ( #17821 )
2026-08-28 16:17:29 -04:00
chenyu and GitHub
bd2711b7c1
put dtype to Ops.INS arg ( #17817 )
2026-08-28 15:05:17 -04:00
chenyu and GitHub
b3dad4fa0c
update X86 cast related stuff ( #17816 )
2026-08-28 14:30:34 -04:00
Teddy Tennant and GitHub
2eacd4fa68
fix Conv2d same padding with per axis dilation ( #17805 )
2026-08-28 12:44:43 -04:00
chenyu and GitHub
cb857a5a40
remove X86 ymm thingy ( #17803 )
2026-08-28 10:18:10 -04:00
chenyu and GitHub
6083de5cff
fix nan compares ( #17801 )
...
* fix nan compares
* fix
2026-08-28 09:17:18 -04:00
nimlgen and GitHub
55c8e18930
hcq2 spec2 ( #17798 )
2026-08-28 12:51:55 +03:00
nimlgen and GitHub
76dce1eb8d
tiny hcq2 changes ( #17797 )
...
* tiny hcq2 changes
* x
* x
2026-08-28 11:29:00 +03:00
Teddy Tennant and GitHub
26c4b6319e
fix interpolate with align_corners and an output size of 1 ( #17790 )
2026-08-27 19:17:05 -07:00
qazal and GitHub
53cad325c7
copy movement op tests construct copies ( #17792 )
...
Unit Tests / Models (push) Successful in 1m22s
Unit Tests / Linux (DSP) (push) Successful in 1m51s
Unit Tests / ONNX (CPU) Tests (push) Successful in 1m55s
Unit Tests / Linters (push) Successful in 1m59s
Unit Tests / Test LLM (push) Successful in 2m6s
Unit Tests / Fuzzing (push) Successful in 2m8s
Unit Tests / hcq2 (push) Successful in 2m43s
Unit Tests / Docs (push) Successful in 2m50s
Unit Tests / AMD ASM IDE (push) Successful in 2m47s
Unit Tests / Null Tests (push) Successful in 2m59s
Unit Tests / openpilot Compile Tests (push) Successful in 2m59s
Unit Tests / Torch Backend Training (push) Successful in 3m3s
Unit Tests / Unit Tests (push) Successful in 3m9s
Unit Tests / Python Backend (push) Successful in 3m16s
Unit Tests / Linux (amdllvm gfx1100) (push) Successful in 3m31s
Unit Tests / Linux (amdllvm gfx1201) (push) Successful in 3m31s
Unit Tests / SPEC=2 (2) (push) Successful in 3m39s
Unit Tests / Linux (DEV=CPU:LVP) (push) Successful in 3m36s
Unit Tests / Linux (DEV=CPU:LLVM) (push) Successful in 3m41s
Unit Tests / SPEC=2 (1) (push) Failing after 3m48s
Unit Tests / Compile-only (DEV=NULL:NAK:sm_120) (push) Successful in 1m43s
Unit Tests / Linux (DEV=CL) (push) Successful in 3m51s
Unit Tests / CL IMAGE Tests (push) Successful in 3m53s
Unit Tests / Optimization Tests (push) Successful in 3m53s
Unit Tests / Linux (DEV=WEBGPU) (push) Successful in 3m49s
Unit Tests / Linux (DEV=CPU:X86) (push) Successful in 3m58s
Unit Tests / Linux (am) (push) Successful in 3m59s
Unit Tests / Linux (amd gfx1100) (push) Successful in 4m0s
Unit Tests / Linux (amd gfx1201) (push) Successful in 4m0s
Unit Tests / Linux (DEV=CPU:CLANG) (push) Successful in 4m17s
Unit Tests / Torch Backend Tests (push) Successful in 4m29s
Deploy Docs / deploy (push) Successful in 4m34s
Unit Tests / Compile-only (DEV=NULL:IR3:a630) (push) Successful in 2m35s
Unit Tests / Linux (amdllvm gfx950) (push) Successful in 3m2s
Unit Tests / Linux (ptx) (push) Successful in 3m0s
Unit Tests / Linux (amd gfx950) (push) Successful in 4m12s
Unit Tests / Linux (nv) (push) Successful in 3m47s
Unit Tests / Compile-only (DEV=NULL:QCOMCL:a630) (push) Successful in 3m36s
Autogen / In-tree Autogen (push) Successful in 11m1s
Autogen / In-tree Autogen (macos) (push) Canceled after 0s
Benchmarks / Mac pytest (push) Canceled after 0s
Benchmarks / LLM (DEV=AMD) (push) Canceled after 0s
Benchmarks / LLM (DEV=METAL) (push) Canceled after 0s
Benchmarks / LLM (DEV=NV) (push) Canceled after 0s
Benchmarks / HLB-CIFAR10 (DEV=AMD) (push) Canceled after 0s
Benchmarks / HLB-CIFAR10 (DEV=METAL) (push) Canceled after 0s
Benchmarks / HLB-CIFAR10 (DEV=NV) (push) Canceled after 0s
Benchmarks / MLPerf (AMD) (push) Canceled after 0s
Benchmarks / MLPerf (NV) (push) Canceled after 0s
Benchmarks / Stable Diffusion (DEV=AMD) (push) Canceled after 0s
Benchmarks / Stable Diffusion (DEV=METAL) (push) Canceled after 0s
Benchmarks / Stable Diffusion (DEV=NV) (push) Canceled after 0s
Benchmarks / Multi-GPU Benchmarks (DEV=AMD) (push) Canceled after 0s
Benchmarks / Multi-GPU Benchmarks (DEV=NV) (push) Canceled after 0s
Benchmarks / Tests (DEV=AMD) (push) Canceled after 0s
Benchmarks / Tests (DEV=METAL) (push) Canceled after 0s
Benchmarks / Tests (DEV=NV) (push) Canceled after 0s
Benchmarks / UsbGPU Benchmark (push) Canceled after 0s
Benchmarks / openpilot 0.11.0 compile3 dmonitoring (DEV=QCOM) (push) Canceled after 0s
Benchmarks / openpilot 0.11.2 compile3 dmonitoring (DEV=QCOM) (push) Canceled after 0s
Benchmarks / openpilot 0.11.0 compile3 policy (DEV=QCOM) (push) Canceled after 0s
Benchmarks / openpilot 0.11.2 compile3 supercombo (DEV=QCOM) (push) Canceled after 0s
Benchmarks / openpilot 0.11.0 compile3 vision (DEV=QCOM) (push) Canceled after 0s
Benchmarks / openpilot 0.11.0 compile3 dmonitoring (DEV=QCOM:IR3) (push) Canceled after 0s
Benchmarks / openpilot 0.11.2 compile3 dmonitoring (DEV=QCOM:IR3) (push) Canceled after 0s
Benchmarks / openpilot 0.11.0 compile3 policy (DEV=QCOM:IR3) (push) Canceled after 0s
Benchmarks / openpilot 0.11.2 compile3 supercombo (DEV=QCOM:IR3) (push) Canceled after 0s
Benchmarks / openpilot 0.11.0 compile3 vision (DEV=QCOM:IR3) (push) Canceled after 0s
Benchmarks / DSP Benchmark (push) Canceled after 0s
Benchmarks / UsbGPU Benchmark (comma) (push) Canceled after 0s
Benchmarks / PCI Driver Benchmark (DEV=AMD) (push) Canceled after 0s
Benchmarks / PCI Driver Benchmark (DEV=NV) (push) Canceled after 0s
Benchmarks / LLVM Speed (push) Canceled after 0s
Platform Tests / MacOS (unit) (push) Canceled after 0s
Platform Tests / MacOS (unit, mock) (push) Canceled after 0s
Platform Tests / MacOS (DEV=METAL) (1) (push) Canceled after 0s
Platform Tests / MacOS (DEV=METAL) (2) (push) Canceled after 0s
Platform Tests / MacOS (DEV=CPU:CLANG) (push) Canceled after 0s
Platform Tests / MacOS (DEV=CPU:LLVM) (push) Canceled after 0s
Platform Tests / MacOS (DEV=CPU:LVP) (push) Canceled after 0s
Platform Tests / MacOS (DEV=WEBGPU) (push) Canceled after 0s
Platform Tests / Windows (DEV=CPU:CLANG) (push) Canceled after 0s
Platform Tests / Windows (DEV=CPU:LLVM) (push) Canceled after 0s
Platform Tests / Windows (DEV=CPU:X86) (push) Canceled after 0s
Platform Tests / Windows (DEV=WEBGPU) (push) Canceled after 0s
* copy movement op tests construct copies
* hop 3 gpus
* assert the amount of bytes moved
* work
* arange 4, everyone copies 16B first
2026-08-28 11:09:11 +09:00
chenyu and GitHub
bc485d271f
fix CMPLT between float16 and const 0 ( #17791 )
...
* fix CMPLT between float16 and const 0
* fix
2026-08-27 21:14:34 -04:00
chenyu and GitHub
38e0fe103b
fix where with float64 branches ( #17789 )
...
* fix where with float64 branches
* fix
2026-08-27 21:03:43 -04:00
chenyu and GitHub
fede358811
fix fancy indexing with uint8 index ( #17787 )
...
* fix fancy indexing with uint8 index
* fix
2026-08-27 19:59:27 -04:00
George Hotz and GitHub
f267a49639
UOp.param accepts single sint as shape (int,) ( #17784 )
...
* UOp.param accepts single sint as shape (int,)
Change UOp.param signature from shape:tuple[sint, ...]|None to shape:tuple[sint, ...]|sint|None.
A single sint means 1D shape (sint,). Update all callers from (n,) to n syntax.
No param_from_shape — multi-dim shapes stay as tuples.
* use single int syntax in hcq2 copy_with_kernel
2026-08-27 14:07:33 -07:00
chenyu and GitHub
54d69f439f
fix test_assign for no subbuffer backends ( #17757 )
...
Autogen / In-tree Autogen (comgr 2) (push) Failing after 45s
Unit Tests / Models (push) Successful in 1m37s
Unit Tests / Linux (DSP) (push) Successful in 1m39s
Unit Tests / Fuzzing (push) Successful in 1m58s
Unit Tests / Linters (push) Successful in 2m0s
Unit Tests / Test LLM (push) Failing after 2m22s
Unit Tests / Docs (push) Successful in 2m57s
Unit Tests / hcq2 (push) Successful in 3m7s
Unit Tests / AMD ASM IDE (push) Successful in 3m12s
Unit Tests / openpilot Compile Tests (push) Successful in 3m17s
Unit Tests / Null Tests (push) Successful in 3m20s
Unit Tests / Python Backend (push) Successful in 3m22s
Unit Tests / CL IMAGE Tests (push) Successful in 3m24s
Unit Tests / Torch Backend Training (push) Successful in 3m25s
Unit Tests / Linux (DEV=CPU:X86) (push) Successful in 3m28s
Unit Tests / Linux (DEV=CPU:LLVM) (push) Successful in 3m58s
Unit Tests / Unit Tests (push) Successful in 4m7s
Unit Tests / Linux (amdllvm gfx1100) (push) Successful in 4m3s
Unit Tests / Linux (DEV=CPU:LVP) (push) Successful in 4m5s
Unit Tests / SPEC=2 (2) (push) Successful in 4m8s
Unit Tests / Linux (DEV=WEBGPU) (push) Successful in 4m8s
Unit Tests / Linux (DEV=CL) (push) Successful in 4m13s
Unit Tests / Optimization Tests (push) Successful in 4m15s
Unit Tests / Linux (am) (push) Successful in 4m20s
Unit Tests / Linux (amdllvm gfx1201) (push) Successful in 3m41s
Unit Tests / SPEC=2 (1) (push) Successful in 4m28s
Unit Tests / Linux (amd gfx1100) (push) Successful in 4m27s
Unit Tests / Torch Backend Tests (push) Successful in 4m33s
Unit Tests / Linux (amd gfx1201) (push) Successful in 4m27s
Unit Tests / Compile-only (DEV=NULL:NAK:sm_120) (push) Successful in 1m38s
Deploy Docs / deploy (push) Successful in 4m39s
Unit Tests / Linux (DEV=CPU:CLANG) (push) Successful in 4m42s
Unit Tests / Compile-only (DEV=NULL:IR3:a630) (push) Successful in 2m22s
Unit Tests / Linux (amdllvm gfx950) (push) Successful in 3m20s
Unit Tests / Linux (ptx) (push) Successful in 3m12s
Unit Tests / ONNX (CPU) Tests (push) Successful in 5m38s
Unit Tests / Linux (amd gfx950) (push) Successful in 4m25s
Unit Tests / Linux (nv) (push) Successful in 4m13s
Unit Tests / Compile-only (DEV=NULL:QCOMCL:a630) (push) Successful in 3m44s
Autogen / In-tree Autogen (push) Successful in 11m23s
Autogen / In-tree Autogen (macos) (push) Canceled after 0s
Benchmarks / Mac pytest (push) Canceled after 0s
Benchmarks / LLM (DEV=AMD) (push) Canceled after 0s
Benchmarks / LLM (DEV=METAL) (push) Canceled after 0s
Benchmarks / LLM (DEV=NV) (push) Canceled after 0s
Benchmarks / HLB-CIFAR10 (DEV=AMD) (push) Canceled after 0s
Benchmarks / HLB-CIFAR10 (DEV=METAL) (push) Canceled after 0s
Benchmarks / HLB-CIFAR10 (DEV=NV) (push) Canceled after 0s
Benchmarks / MLPerf (AMD) (push) Canceled after 0s
Benchmarks / MLPerf (NV) (push) Canceled after 0s
Benchmarks / Stable Diffusion (DEV=AMD) (push) Canceled after 0s
Benchmarks / Stable Diffusion (DEV=METAL) (push) Canceled after 0s
Benchmarks / Stable Diffusion (DEV=NV) (push) Canceled after 0s
Benchmarks / Multi-GPU Benchmarks (DEV=AMD) (push) Canceled after 0s
Benchmarks / Multi-GPU Benchmarks (DEV=NV) (push) Canceled after 0s
Benchmarks / Tests (DEV=AMD) (push) Canceled after 0s
Benchmarks / Tests (DEV=METAL) (push) Canceled after 0s
Benchmarks / Tests (DEV=NV) (push) Canceled after 0s
Benchmarks / UsbGPU Benchmark (push) Canceled after 0s
Benchmarks / openpilot 0.11.0 compile3 dmonitoring (DEV=QCOM) (push) Canceled after 0s
Benchmarks / openpilot 0.11.2 compile3 dmonitoring (DEV=QCOM) (push) Canceled after 0s
Benchmarks / openpilot 0.11.0 compile3 policy (DEV=QCOM) (push) Canceled after 0s
Benchmarks / openpilot 0.11.2 compile3 supercombo (DEV=QCOM) (push) Canceled after 0s
Benchmarks / openpilot 0.11.0 compile3 vision (DEV=QCOM) (push) Canceled after 0s
Benchmarks / openpilot 0.11.0 compile3 dmonitoring (DEV=QCOM:IR3) (push) Canceled after 0s
Benchmarks / openpilot 0.11.2 compile3 dmonitoring (DEV=QCOM:IR3) (push) Canceled after 0s
Benchmarks / openpilot 0.11.0 compile3 policy (DEV=QCOM:IR3) (push) Canceled after 0s
Benchmarks / openpilot 0.11.2 compile3 supercombo (DEV=QCOM:IR3) (push) Canceled after 0s
Benchmarks / openpilot 0.11.0 compile3 vision (DEV=QCOM:IR3) (push) Canceled after 0s
Benchmarks / DSP Benchmark (push) Canceled after 0s
Benchmarks / UsbGPU Benchmark (comma) (push) Canceled after 0s
Benchmarks / PCI Driver Benchmark (DEV=AMD) (push) Canceled after 0s
Benchmarks / PCI Driver Benchmark (DEV=NV) (push) Canceled after 0s
Benchmarks / LLVM Speed (push) Canceled after 0s
Platform Tests / MacOS (unit) (push) Canceled after 0s
Platform Tests / MacOS (unit, mock) (push) Canceled after 0s
Platform Tests / MacOS (DEV=METAL) (1) (push) Canceled after 0s
Platform Tests / MacOS (DEV=METAL) (2) (push) Canceled after 0s
Platform Tests / MacOS (DEV=CPU:CLANG) (push) Canceled after 0s
Platform Tests / MacOS (DEV=CPU:LLVM) (push) Canceled after 0s
Platform Tests / MacOS (DEV=CPU:LVP) (push) Canceled after 0s
Platform Tests / MacOS (DEV=WEBGPU) (push) Canceled after 0s
Platform Tests / Windows (DEV=CPU:CLANG) (push) Canceled after 0s
Platform Tests / Windows (DEV=CPU:LLVM) (push) Canceled after 0s
Platform Tests / Windows (DEV=CPU:X86) (push) Canceled after 0s
Platform Tests / Windows (DEV=WEBGPU) (push) Canceled after 0s
* test_assign to test/backend
* fix
2026-08-27 10:47:21 -04:00
Teddy Tennant and GitHub
5c3d044465
fix asinh gradient at zero ( #17758 )
2026-08-27 09:51:17 -04:00
qazal and GitHub
d16a2510ad
remove some dead scheduler tests ( #17768 )
...
* remove tests that use .uop
* that test does nothing
2026-08-27 16:01:28 +09:00
George Hotz and GitHub
6042b87272
delete PCONTIG [PR] ( #17756 )
...
* delete PCONTIG
* cleanups
2026-08-26 12:07:28 -07:00
chenyu and GitHub
6a3b297548
fix PTX NIR SPEC=2 for bool [pr] ( #17753 )
...
* fix PTX NIR SPEC=2 for bool [pr]
storing bool with uint8 needs to pass SPEC
* the fix
2026-08-26 14:07:29 -04:00
George Hotz and GitHub
ea6c82f3be
small changes from new rangeify ( #17752 )
2026-08-26 10:33:31 -07:00
chenyu and GitHub
4bdc865131
delete unused rewrite rules [PR] ( #17748 )
2026-08-26 10:50:42 -04:00
Teddy Tennant and GitHub
4c20f1d357
fix asinh precision loss on negative inputs ( #17749 )
2026-08-26 10:08:04 -04:00
George Hotz and GitHub
625c05df1e
fix am_smi to respect dev ( #17742 )
2026-08-25 19:21:55 -07:00
chenyu and GitHub
7dc8b666e7
wgsl cast before load for packed [pr] ( #17731 )
...
instead of explicit dtype on load
2026-08-25 11:46:31 -04:00
chenyu and GitHub
9f01775cf4
dtype_from_uop for THREEFRY and FDIV ( #17727 )
2026-08-25 08:43:59 -04:00
chenyu and GitHub
9607787ce1
delete more explicitly set dtype to UOp [PR] ( #17726 )
...
* delete more explicitly set dtype to UOp [PR]
* not that
2026-08-25 08:32:45 -04:00
George Hotz and GitHub
5aabbb1991
fix Muon weight decay being a no-op ( #17709 )
...
* fix Muon weight decay being a no-op
LARS._step computed the post-momentum weight decayed param but only
used it for a dtype cast, so the decay was never applied. Fold the
decay into the update instead, matching torch's param.mul_(1 - lr*wd).
test_muon_wd passed anyway since lr*wd=1e-5 is far below atol, so
also bump the test's weight_decay to 10 to actually exercise it.
* muon: apply weight decay after lr scaling
keeps the decoupled decay independent of the LARS trust ratio r,
matching torch's param.mul_(1 - lr*wd). no behavior change today since
r is always 1.0 on the pre_wd=False path (Muon has tcoef=0).
2026-08-24 13:33:10 -07:00
chenyu and GitHub
8f9cbdf0cc
few more self folding [pr] ( #17657 )
2026-08-21 15:21:28 -04:00
nimlgen and GitHub
1cf8a2c7fe
hcq2: use shrink.bitcast ( #17653 )
...
* hcq2: shrink.bitcast
* x
* x
* x
* s
* x
* Dx
* Revert "hotfix: disable HCQ2"
This reverts commit a57188ea6d
2026-08-21 21:24:34 +03:00
chenyu and GitHub
3919ce8427
ceildiv in wgsl _packed_size [pr] ( #17662 )
2026-08-21 14:02:20 -04:00
qazal and GitHub
d716d0d927
hotfix: qkv bw kernel requires hipcc and cdna4 ( #17645 )
2026-08-21 12:38:45 +09:00
George Hotz and GitHub
9aa9e11301
compile kernels in parallel ( #17629 )
...
* compile kernels in parallel (slop)
* cleanups
* fixes
* hand cleanups
* no PARALLEL with VIZ
* ugh, we need to remove the n from names
* no hcq2 in that test
* main for abstractions3
* fix compile server to be fork safe
* fix num cpu threads in context
* just use a lock
* fix __main__ on spawn
* compileonce
* python3 is double
* xdist sizing
* parallel compile
* fetch supports extract
* fix issues
* revert compiler server to master, drop compileonce
* normal workflow
* PARALLEL=0 for QCOMCL compiletest
2026-08-20 16:41:45 -07:00
George Hotz and GitHub
3fdbb82bfe
add ansipad and PARALLEL contextvar ( #17642 )
2026-08-20 16:26:02 -07:00
sirhcm and GitHub
c773891e3f
skip INT_MIN % -1 on X86 and LLVM ( #17637 )
2026-08-20 17:26:26 -04:00
chenyu and GitHub
0e7ab863a0
x86 REX issue ( #17580 )
...
* failing test
* fix
2026-08-20 16:49:04 -04:00
George Hotz and GitHub
1707dca3b4
remove kernel_cnt, names are no longer unique ( #17633 )
...
* remove kernel_cnt, names are no longer unique
* check uops, not names
* fix hcq2
2026-08-20 13:28:25 -07:00
chenyu and GitHub
a8ecb73363
x64 imm uint64 ( #17636 )
...
* x64 imm uint64
* fix
2026-08-20 15:30:17 -04:00
qazal and GitHub
a1263fadf3
fused_qkv_rope in UOp try 2 ( #17619 )
...
* fused_qkv_rope in UOp try 2
* dont need that
* less
2026-08-20 12:28:34 +09:00
chenyu and GitHub
fc214da417
test updates for weak const change ( #17606 )
2026-08-19 15:56:01 -04:00
7064e76bc8
fix roll on zero-sized tensors ( #17603 )
...
Signed-off-by: Bennett <[email protected] [email protected]
2026-08-19 15:31:21 -04:00
chenyu and GitHub
9550378704
finish casted_consts migration [PR] ( #17595 )
2026-08-19 10:43:55 -04:00
chenyu and GitHub
e8ba214b56
casted CONST migration for x86 [pr] ( #17592 )
...
* casted CONST migration for x86 [pr]
* style
2026-08-19 09:38:09 -04:00
qazal and GitHub
d539aaf752
Revert "fused_qkv_rope in UOp ( #17591 )" ( #17593 )
...
This reverts commit 8c2bf02d17
2026-08-19 21:42:55 +09:00
qazal and GitHub
8c2bf02d17
fused_qkv_rope in UOp ( #17591 )
...
* llama: 4% faster fused_qkv_rope
* prep
* add uop kernel, has_hipcc is cached
* less
2026-08-19 18:07:30 +09:00
George Hotz and GitHub
c31038ff37
use KernelCountException when kernel count is being compared ( #17584 )
2026-08-18 16:06:03 -07:00
chenyu and GitHub
49778d9a48
start renderer casted const migration [pr] ( #17582 )
...
before rendering, rewrite strong typed const to casted weak const and have renderer adopt the new UOp. starting with PYTHON
2026-08-18 17:43:27 -04:00
chenyu and GitHub
138fb4a783
delete dead DType.scalar [PR] ( #17561 )
2026-08-16 21:12:17 -04:00
chenyu and GitHub
057a18a07c
fix emulated long cast to double ( #17559 )
2026-08-16 20:52:42 -04:00
George Hotz and GitHub
e688e07758
add max_shape/max_numel to mixins + pad_to ( #17553 )
2026-08-15 20:05:10 -07:00
chenyu and GitHub
417563ca20
fix webgpu is_nan [pr] ( #17551 )
2026-08-15 16:31:19 -04:00
sirhcmandGitHub