Commit Graph
14374 Commits
Author SHA1 Message Date
chenyuandGitHub 0a0b6cb596 fix TestDevCopySpeeds command (#17607) 2026-08-19 15:54:41 -04:00
chenyuandGitHub b8cc74ecf8 no float in tensor shape [pr] (#17605) 2026-08-19 15:40:16 -04:00
7064e76bc8 fix roll on zero-sized tensors (#17603)
Signed-off-by: Bennett <[email protected]>
Co-authored-by: Bennett <[email protected]>
2026-08-19 15:31:21 -04:00
nimlgenandGitHub 0c5307b4f3 realize: fast stat (#17600)
* hcq2: fast stat

* x

* Dx
2026-08-19 22:04:06 +03:00
chenyuandGitHub a4fadcf606 fix TestDevCopySpeeds SIZE (#17602)
SIZE should be int
2026-08-19 14:53:17 -04:00
chenyuandGitHub c218b4842d fold_bitcast should truncate its input [pr] (#17601) 2026-08-19 14:39:36 -04:00
chenyuandGitHub bd6e70ac15 delete stale tests (#17596) 2026-08-19 11:18:05 -04:00
chenyuandGitHub 9550378704 finish casted_consts migration [PR] (#17595) 2026-08-19 10:43:55 -04:00
chenyuandGitHub b3e2f17b24 update NULL tests that depends on strong dtype CONST (#17594) 2026-08-19 10:26:38 -04:00
chenyuandGitHub e8ba214b56 casted CONST migration for x86 [pr] (#17592)
* casted CONST migration for x86 [pr]

* style
2026-08-19 09:38:09 -04:00
chenyuandGitHub 68b4407fe3 casted CONST migration for cstyle [pr] (#17587) 2026-08-19 09:01:40 -04:00
qazalandGitHub d539aaf752 Revert "fused_qkv_rope in UOp (#17591)" (#17593)
This reverts commit 8c2bf02d17.
2026-08-19 21:42:55 +09:00
qazalandGitHub 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
chenyuandGitHub ca86a42703 casted CONST migration for nir [pr] (#17588) 2026-08-18 23:04:07 -04:00
sirhcmandGitHub df3b114fbc ci: standardize all ubuntu runs-on to ubuntu-24.04 (#17586) 2026-08-18 21:49:07 -04:00
chenyuandGitHub e37b44d048 casted CONST migration for LLVM and PTX [pr] (#17585) 2026-08-18 21:07:43 -04:00
sirhcmandGitHub 2cfb421a81 ci: cleanup deps (#17583) 2026-08-18 19:51:03 -04:00
George HotzandGitHub c31038ff37 use KernelCountException when kernel count is being compared (#17584) 2026-08-18 16:06:03 -07:00
chenyuandGitHub 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
wozeparrotandGitHub 72280bb218 gptoss: zero-2 optim (#17581) 2026-08-18 14:28:57 -07:00
nimlgenandGitHub af2a43c850 hcq2: 64bit addresses (#17576) 2026-08-18 16:52:18 +03:00
chenyuandGitHub a1366e2f6c alu(long, weakint) can do math in int too [pr] (#17579)
* alu(long, weakint) can do math in int too [pr]

* remove
2026-08-18 09:08:35 -04:00
nimlgenandGitHub 0b757bb9bc Revert "disk: neable polling (#17538)" (#17578)
This reverts commit c17849a1f8.
2026-08-18 15:31:25 +03:00
qazalandGitHub 7cbe8e0d15 viz: expanding srcs should not override history (#17577) 2026-08-18 17:45:43 +09:00
sirhcmandGitHub a746861ac0 compile server for cuda on mac (#17574) 2026-08-17 22:55:59 -04:00
George HotzandGitHub 8d2cc64b69 llm: refactor delta attention (#17564)
* refactor delta attention

* cleanups

* bugfixes

* stack

* recurrent w chunk_size 1

* revert that

* extra test
2026-08-17 19:24:03 -07:00
chenyuandGitHub cb892e1b92 base_rewrite reorder [PR] (#17575)
put const before cast, const will become casted const later
2026-08-17 20:55:44 -04:00
chenyuandGitHub d4a1f39038 clean up cstyle render inf and nan [PR] (#17573) 2026-08-17 18:01:11 -04:00
chenyuandGitHub 34c9b9d434 add back cast where rule [pr] (#17572) 2026-08-17 15:27:20 -04:00
chenyuandGitHub 901d257a26 some more torch backend cleanups (#17571)
* some more torch backend cleanups

* fix
2026-08-17 15:19:57 -04:00
b1tgandGitHub b757437f64 llm: respect expert_gating_func (#17458)
* llm: respect expert_gating_func

* test

* enum

* clean
2026-08-17 12:19:33 -07:00
nimlgenandGitHub 00d6eed43c hcq2: speed (#17570)
* hcq2: speed

* x

x
2026-08-17 21:03:44 +03:00
b1tgandGitHub 2776c5b369 fix call arg indexing in shard scheduling (#17519) 2026-08-17 09:57:50 -07:00
nimlgenandGitHub 58edff61d9 hcq2: one submitter (#17556)
* hcq2: c submitter

* x

* x

* x

* simpler

* simpler

* x

* x

* Dx

* revrt

* Dx

* x

* fst

* fix
2026-08-17 16:08:19 +03:00
chenyuandGitHub 954d4f7797 add back beautiful_mnist_torch in CI (#17569) 2026-08-17 08:23:34 -04:00
chenyuandGitHub 7fe8e350c5 delete bad torch backend function override (#17568) 2026-08-17 07:56:05 -04:00
chenyuandGitHub 42714e1399 update a few is CONST check to check device None [pr] (#17563)
* update a few is CONST check to check device None [pr]

* clone
2026-08-17 07:33:16 -04:00
chenyuandGitHub 821e80ff9a remove torch backend detach hack (#17565) 2026-08-17 07:33:05 -04:00
George HotzandGitHub 37a54dc7cf add some dels to jit for OOM fixes (#17566) 2026-08-16 23:39:37 -07:00
chenyuandGitHub e25f86721d more torch backend fixups (#17562) 2026-08-16 21:31:53 -04:00
chenyuandGitHub 138fb4a783 delete dead DType.scalar [PR] (#17561) 2026-08-16 21:12:17 -04:00
chenyuandGitHub bfd4048abf no dtype in vconst_like [PR] (#17560) 2026-08-16 21:06:38 -04:00
chenyuandGitHub 057a18a07c fix emulated long cast to double (#17559) 2026-08-16 20:52:42 -04:00
chenyuandGitHub c30bf116b7 few torch_backend fix (#17558)
* few torch_backend fix

* fix
2026-08-16 20:07:25 -04:00
nimlgenandGitHub e7bf2a811d iface in device (#17554)
* iface in device

* drop

* move

* sorry
2026-08-16 12:09:24 +03:00
George HotzandGitHub e688e07758 add max_shape/max_numel to mixins + pad_to (#17553) 2026-08-15 20:05:10 -07:00
nimlgenandGitHub 97022960ae device: fix remap (#17549) 2026-08-16 01:01:36 +03:00
chenyuandGitHub 417563ca20 fix webgpu is_nan [pr] (#17551) 2026-08-15 16:31:19 -04:00
chenyuandGitHub 5ca87f1bac fix cast to weak twice [pr] (#17548)
also no gradient for weak target
2026-08-15 12:39:54 -04:00
chenyuandGitHub 26cbadd69a no pm_fold_cast_const in full_rewrite_to_sink [PR] (#17547) 2026-08-15 10:18:59 -04:00