mirrors/tinygrad
2
0
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-06-24 02:14:17 +00:00
Code Issues Projects Releases Packages Wiki Activity
8,229 commits 280 branches 15 tags 325 MiB
Commit graph

3,493 commits

Author SHA1 Message Date
George Hotz
ff3438be4e fast 2025-03-21 13:04:18 +08:00
George Hotz
bc5e23061b diasm 2025-03-21 11:22:40 +08:00
George Hotz
5ce951fb34 l2 2025-03-21 11:14:12 +08:00
George Hotz
4a49d05a3f
 		Merge branch 'master' into dsp_search 2025-03-21 10:26:38 +08:00
George Hotz
d42350a401 simple test 2025-03-20 12:37:29 +08:00
George Hotz
3c5161b4cb
 		add validation of the bounds of Ops.INDEX (#9503) 
* add validation of the bounds of Ops.INDEX

* do mask properly

* more validation

* correct

* fix gated

* add CAST support to vmin/vmax

* fix ptx and image

* ptx no diff

* upat.index also stays

---------

Co-authored-by: qazal <qazal.software@gmail.com>
 2025-03-20 12:15:55 +08:00
qazal
0b20f91ce7
 		remove move_mask from the devectorizer (#9511) 
* remove move_mask from the devectorizer

* add (wrong) ptx

* reason

* enable index addition in PTX, we won't have the INDEX anyways

* space
 2025-03-20 11:53:12 +08:00
George Hotz
223feb2118
 		Merge branch 'master' into dsp_search 2025-03-20 10:52:30 +08:00
qazal
1839e8c9b3
 		place masks in INDEX for TestGatedStoreRewrite [pr] (#9512) 2025-03-20 09:46:53 +08:00
geohotstan
8c0d0a122c
 		Add return_indices to max_pool (#9506) 
* wow argmax is so good

* 1 less line

* clean up and better variable names

* is this torch thing right...?

* add more tests

* slap a TODO on it

* clean ups

* prettier looking code and fix ceil mode test

* add return types and some docs

* ok that was a bad example since indices == value, just no example
 2025-03-19 15:25:37 -04:00
chenyu
189f62d44f
 		add rounding to tqdm unit scale (#9507) 
fixed `AssertionError: ' 1.00/10.0  1000it/s]' != ' 1.00/10.0  1.00kit/s]'`
 2025-03-19 12:08:46 -04:00
b1tg
2c87a22cf2
 		fix prg size calculation when there are adjacent mapped ranges (#9498) 
Co-authored-by: b1tg <b1tg@users.noreply.github.com>
 2025-03-19 11:55:03 +08:00
chenyu
f8976dd2eb
 		enable more webgpu tests (#9502) 
OSX has larger buffer number limit, and it supports fp16 now
 2025-03-18 23:03:54 -04:00
qazal
ae688e4103
 		simple failing test for scheduling parallel reduce [pr] (#9501) 
* simple failing test for scheduling parallel reduce [pr]

* atol
 2025-03-19 10:52:13 +08:00
George Hotz
117b7a16ef
 		VALIDATE_WITH_CPU [pr] (#9488) 
* VALIDATE_WITH_CPU [pr]

* fix test
 2025-03-18 15:15:04 +08:00
qazal
935cd01f56
 		simple failing test for graph_rewrite children [pr] (#9489) 
* simple failing test for graph_rewrite children [pr]

* lint

* update too
 2025-03-18 13:07:21 +08:00
Anish Umale
5e58f4b65b
 		Tiny backend test_ops fix part 3 (#9483) 
* extract straightforward things from https://github.com/tinygrad/tinygrad/pull/9302

* pass dtype and device for ones_like
 2025-03-17 18:01:51 -04:00
TJ
9fcef4d009
 		add masked_select to tensor.py (#9468) 
* add masked_select to tensor.py

* fix tests

---------

Co-authored-by: George Hotz <72895+geohot@users.noreply.github.com>
 2025-03-17 16:05:36 -04:00
chenyu
4f8eac59ea
 		failed test case for threefry (#9469) 
* failed test case for threefry

not sure if it's always like this, but increment before _threefry_random_bits is incorrect. the counts should start with random numbers generated so far.

use jax to generate 20 + 20 + 10 random numbers, the first 20 + 20 matches and the last 10 are different. just moving increment after _threefry_random_bits matches the number but jit test failes

* workaround

* why is this different?

* revert those

* and that
 2025-03-17 14:52:10 -04:00
geohotstan
53d6f1e1bb
 		Add bitonic cat sort (#9422) 
* poc

* repeated values fail, sigh

* is this being timed out?

* fix up down names

* bitonic v2, does this run?

* bitonic v3, faster

* bitonic v3.1, faster

* bitonic v3.1.1, same speed unlucky

* support dim and indices

* bitonic v3.2, simpler code, TODO repeated indices

* bruv gimme green for once cmon

* cat (stack) implementation, slow but maybe one day when cat is fast meow

* revert to v3.2

* bitonic v4, who let the cats out edition

* clean up variable names

* figured out repeated indices :D

* ruff check --fix

* use sort for topk

* add Tensor.sort everywhere

* fix docs and add some types

* slightly better variable names

* am I doing torch inplace correctly?

* delegate sort to values_stable

* add a contig, faster first sort

* maybe don't test_inplace

---------

Co-authored-by: chenyu <chenyu@fastmail.com>
 2025-03-17 12:01:23 -04:00
qazal
e03c0aacf2
 		more explicit DONT_PUSH_VIEWS [pr] (#9479) 
* more explicit DONT_PUSH_VIEWS [pr]

* update tests to not handcode ast

* lint

* test_recursive_swizzle and test_simple_store_reshape
 2025-03-17 20:43:21 +08:00
qazal
3b00a778ba
 		fix view_left for unsafe pad ops [pr] (#9478) 2025-03-17 19:02:02 +08:00
George Hotz
31cd00e72f fix name get 2025-03-17 18:09:39 +08:00
qazal
813f713edc
 		merge_views for buffer ops + create valids last (#9472) 
* merge_views for buffer ops + create valids last

* view.arg

* pass
 2025-03-17 17:15:44 +08:00
qazal
bd1f71c1e2
 		simple failing test for extra ops in VALID [pr] (#9474) 
* simple failing test for extra valids [pr]

* this has DEBUG=4
 2025-03-17 17:02:40 +08:00
qazal
e26caf4c3a
 		hotfix: skip test_mean_half_precision_underflow on amd ci (#9476) 
The global size is very large (781250 gidx) and the emulated version takes more than 1
minute to execute the kernel.
 2025-03-17 16:47:48 +08:00
George Hotz
15ee742afa
 		add get_children_map to uop (#9470) 
* add get_children_map to uop

* update_children

* fix new children
 2025-03-17 14:36:13 +08:00
George Hotz
cb7a7f69c7
 		quantization preprocessor from DSP, should be universal (#9437) 
* quantization preprocessor from DSP, should be universal

* touchups

* fix tests
 2025-03-15 07:49:37 +08:00
chenyu
99b0287e4e
 		add GROUP and GROUPTOP to test_arange (#9432) 
it does not grow quadratically, but it's not 0 ops now
 2025-03-13 11:28:38 -04:00
qazal
90ffa9bd45
 		swizzle without buffer ops try 2 [pr] (#9427) 
* add DONT_PUSH_VIEWS to matchers

* swizzle without buffer ops try 2 [pr]

* swizzle reduceop

* simple failing test

* fix failing test

* s/on/for
 2025-03-13 10:00:40 +01:00
chenyu
22fc0a2e36
 		bert sum acc in half (#9412) 
also BS=96
 2025-03-11 23:03:15 -04:00
George Hotz
e174c6c3bc
 		new devectorizer (#9331) 
* new devectorizer

* lidx

* test linearizer passes

* fix images

* fix unfoldable image load

* delete unused

* improve fix_unfoldable_image_load

* working for image

* fixup types

* fixup transcendental

* cast_vec

* cleaner transcendental

* skip failing test

* err, flip that

* not devec

* sqrt
 2025-03-11 18:47:56 +08:00
George Hotz
2780e2027e
 		devectorize prereqs [pr] (#9404) 2025-03-11 12:33:29 +08:00
chenyu
01e8b60911
 		acc_dtype -> dtype (#9402) 
matched numpy and torch
 2025-03-10 16:05:30 -04:00
qazal
59dfb234eb
 		replace hardcoded ast with tensors in TestSwizzle [pr] (#9401) 2025-03-10 19:33:57 +01:00
geohotstan
1d64c12f2b
 		add Topk to tensor (#9343) 
* terrible but somewhat working impl

* linux behaves differently than macos?

* slightly better impl

* small clean up; haven't figured this out yet

* better

* torch has different behavior on linux and macos for duplicated values

* add sum docs

* fix test

* add torch return_type test

* add an exception test

* wrap_fxn instead, and move op lower in order

* better repeated values test

* rerun ci
 2025-03-09 20:01:42 -04:00
qazal
a1f41fadf6
 		test_schedule cleanups + add DONT_GROUP_REDUCES [pr] (#9392) 
* test_schedule cleanups + add DONT_GROUP_REDUCES [pr]

* replace with test_swizzle_reduceop

* delete duplicate tests

* test_allow_push_permutes

* one kernel tests
 2025-03-09 15:01:08 +01:00
qazal
286b480f82
 		do not replace assign with the offset buffer [pr] (#9387) 2025-03-08 11:57:44 +01:00
qazal
0d2762c010
 		prep refactor for adding buffer ops last [pr] (#9383) 
* prep refactor for adding buffer ops last [pr]

* freeze buffers

* add swizzle_reduceop

* shape for reduceop_view_right

* simpler elementwise_view_right

* add shapetracker to const

* only const

* from process replay
 2025-03-08 08:00:14 +01:00
nimlgen
243078dda9
 		am: optimize tlb usage (#9049) 
* am: optimize tlb usage

* fxies

* comments

* tiny
 2025-03-07 19:37:29 +03:00
geohotstan
088d86691b
 		fix onnx gather and onnx auto_pad VALID mode (#9375) 
* fix gather and auto_pad

* long -> int64
 2025-03-07 10:27:23 -05:00
hooved
136cf7b8b1
 		hotfix: load >2 GiB from disk on macOS (#9361) 
* enable loading >2 GiB buffer from disk on macOS

* handle None case raised by mypy

* add test

* revert fix to repro bug in CI

* tell CI to run a unit test for macOS

* reapply fix
 2025-03-07 14:51:58 +08:00
nimlgen
9bd13de44c
 		lower test_gemv_4096_16384 to 750 for red (#9367) 2025-03-05 22:44:48 +03:00
uuuvn
b75f307234
 		amd: autogen ip bases (#9360) 2025-03-05 22:30:38 +03:00
chenyu
2cb2fce8d9
 		lower test_gemm_8192 amd_tflops to 65 (#9364) 2025-03-05 14:06:11 -05:00
nimlgen
14c88abf27
 		add some options to allreduce bench (#9348) 2025-03-04 23:46:36 +03:00
Anish Umale
bafa40fe12
 		Tiny backend test_ops fix part1 (#9338) 
* extract name methods from https://github.com/tinygrad/tinygrad/pull/9302

* t.grad.numpy() -> t.grad.cpu().numpy()

* revert TORCH_DEBUG change

* revert dtype change in aten.sum
 2025-03-03 12:36:51 -05:00
George Hotz
0d4ba7dd87
 		import tinygrad.frontend.torch (#9337) 
* import tinygrad.frontend.torch

* type ignore
 2025-03-04 00:15:29 +08:00
qazal
23084fd850
 		merge merge_views and remove_movement_ops [pr] (#9333) 
* merge merge_views and remove_movement_ops [pr]

* fix that assert
 2025-03-03 12:38:59 +01:00
George Hotz
ece0a0f305
 		use empty for test instead of rand (#9332) 2025-03-03 16:19:06 +08:00