geohot
16beb20398
view_as helper, simpler resolve_function, spec update
2026-08-27 15:43:43 -07:00
geohot
5c81e5fa94
image is a in-graph view, folded into the param arg at render; drop dead multi param sharding
2026-08-27 15:36:35 -07:00
geohot
cde70d3c00
drop gratuitous syntax changes
2026-08-27 15:07:40 -07:00
geohot
0ff39ccfa4
no param_from_shape
2026-08-27 15:06:23 -07:00
geohot
ed78370367
move shape into arg for param/buffer
2026-08-27 15:06:23 -07:00
geohot
ee3161e924
hotfix: decrease dims in test_attention to get below the 90s limit
2026-08-27 14:44:06 -07:00
geohot
38fa0643cf
hotfix: update max test time from 300 -> 90
2026-08-27 14:27:22 -07:00
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
sirhcm and GitHub
14cbecef8f
autogen: use deb for comgr 2 ( #17782 )
2026-08-27 16:32:16 -04:00
chenyu and GitHub
54d69f439f
fix test_assign for no subbuffer backends ( #17757 )
...
* test_assign to test/backend
* fix
2026-08-27 10:47:21 -04:00
nimlgen and GitHub
980c42d611
hcq2 tiny speedups ( #17773 )
2026-08-27 17:42:37 +03:00
nimlgen and GitHub
7eb763a3c2
bnxt to extra ( #17772 )
...
* bnxt to extra
* x
* x
* les
2026-08-27 17:29:17 +03:00
Teddy Tennant and GitHub
5c3d044465
fix asinh gradient at zero ( #17758 )
2026-08-27 09:51:17 -04:00
nimlgen and GitHub
f19d89e29a
hcq2: a bit cleaner deps ( #17771 )
2026-08-27 14:24:59 +03:00
tonyfloatersu and GitHub
2d4e7afcbf
mockgpu/amd: use generated KFD ioctl metadata ( #17761 )
...
* init commit
* init commit
* minimize to single file scope
2026-08-27 12:24:37 +03:00
qazal and GitHub
0d66815849
set BEAM_PADTO=0 in llama ( #17769 )
2026-08-27 16:45:33 +09:00
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
qazal and GitHub
118e812b31
simplify more same device copies ( #17766 )
...
* simplify more same device copies
* stack of contigs or copies
* rename
2026-08-27 12:54:12 +09:00
sirhcm and GitHub
0820c7a424
benchmarks: openpilot matrix ( #17714 )
2026-08-26 18:53:55 -04:00
wozeparrot and GitHub
22ffff3258
gptoss: single jit ( #17525 )
2026-08-26 15:32:21 -07:00
George Hotz and GitHub
a0a901c8e4
faster qwen 3.8 ( #17720 )
...
* faster qwen
* test fix
* dead code
* fix gguf issue
* pretty nt loads
* lil
* use warp
2026-08-26 15:23:25 -07:00
sirhcm and GitHub
c015351ac5
fix _get_cpu_count for docker --cpus=N in python 3.13+ ( #17760 )
2026-08-26 16:53:30 -04:00
nimlgen and GitHub
6074c002e1
hcq2: fix jit ( #17747 )
...
* hcq2: fix reduce
* Dx
* inputs table
* emoty commit
2026-08-26 22:36:36 +03:00
George Hotz and GitHub
6042b87272
delete PCONTIG [PR] ( #17756 )
...
* delete PCONTIG
* cleanups
2026-08-26 12:07:28 -07:00
sirhcm and GitHub
cc72b9f7be
cleanup BENCHMARK_LOG ( #17754 )
2026-08-26 14:55:54 -04:00
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
0abcf09b74
never bufferize_to_store weak input [PR] ( #17751 )
...
github github
2026-08-26 13:25:52 -04:00
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
qazal and GitHub
ecf79e260d
better all2all schedule test ( #17746 )
...
* better all2all schedule test
* deconstruct those numbers
* reorder
2026-08-26 15:46:10 +09:00
b1tg and GitHub
9860e5d285
llm tokenizer: fix tekken, add gpt4o ( #17733 )
2026-08-25 23:30:35 -07:00
George Hotz and GitHub
625c05df1e
fix am_smi to respect dev ( #17742 )
2026-08-25 19:21:55 -07:00
sirhcm and GitHub
b49c03fb1c
benchmarks: don't use sudo on mac ( #17740 )
2026-08-25 21:06:04 -04:00
George Hotz and GitHub
dc04c7820e
lil fixes from new_rangeify ( #17741 )
...
* lil fixes from new_rangeify
* gpt sol review
2026-08-25 17:15:28 -07:00
chenyu and GitHub
6ece327cf3
CUSTOM arg is (str, dtype) [PR] ( #17737 )
2026-08-25 20:15:06 -04:00
wozeparrot and GitHub
07268b724f
fix: external_test_gpu_crash on python 3.14 ( #17739 )
2026-08-25 15:43:30 -07:00
sirhcm and GitHub
55032514ce
pin onnxruntime==1.24.1 ( #17738 )
2026-08-25 17:53:15 -04:00
YassineYousfi and GitHub
2824504f90
usb amd: yield between signal polls ( #17712 )
...
* AMD_USB_POLL_US
* com
* its micro
2026-08-25 14:11:53 -07:00
sirhcm and GitHub
7fef98c86e
benchmarks: remove usage of sudo from linux runners ( #17735 )
2026-08-25 16:21:08 -04:00
geohot
b831ca62d9
Reapply "disk cache: thread-local db conn ( #17694 )"
...
This reverts commit df528499ce .
2026-08-25 13:20:15 -07:00
nimlgen and GitHub
d851aca9ae
hcq2: fix copy/call in usb ( #17732 )
2026-08-25 19:03:19 +03:00
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
nimlgen and GitHub
023bfdb380
fix viz for hcq2 ( #17729 )
2026-08-25 18:23:22 +03:00
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
qazal and GitHub
ab79879613
bring back sqtt examples tests ( #17725 )
2026-08-25 17:00:18 +09:00
George Hotz and GitHub
a5678317c2
split rangeify to prepare.py ( #17722 )
2026-08-24 19:12:13 -07:00
chenyu and GitHub
1d694dd700
remove more explicitly set dtype [PR] ( #17721 )
2026-08-24 22:03:22 -04:00
chenyu and GitHub
65ca68567e
const cleanups [PR] ( #17719 )
2026-08-24 21:37:14 -04:00