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
10,980 commits 280 branches 15 tags 325 MiB
Commit graph

10,980 commits

This branch
This branch
All branches
Author SHA1 Message Date
George Hotz
db8c6d9a04 work 2025-11-10 15:26:16 -08:00
George Hotz
0647f87bf8 outer range runs in the scheduler 2025-11-10 14:49:42 -08:00
chenyu
829cdafccc
 		update openpilot slow conv uop ast (#13197) 
the two remaining slow ones
 2025-11-10 17:03:20 -05:00
George Hotz
0c978d45e6
 		stub attention (#13196) 
* stub attention

* name the kernels
 2025-11-10 13:48:38 -08:00
chenyu
58c30fc7ce
 		minor image_conv2d cleanup (#13193) 2025-11-10 16:05:40 -05:00
chenyu
60e55d9a2d
 		line count 18500 (#13191) 2025-11-10 13:52:13 -05:00
nimlgen
09a59c2203
 		qcom: support new chip versioning (#13185) 
* qcom: support new chip versioning

* ops

* nit

* fix

* f
 2025-11-10 23:57:29 +08:00
qazal
50934050bc
 		sqtt: append all wave execs (#13190) 2025-11-10 23:50:08 +08:00
qazal
38a24731a1
 		cleanup sqtt tooling (#13188) 
* cleanup viz/serve.py

* use latest profile in rgptool.py

* unwrap nullable in roc.py, fix disasms typing
 2025-11-10 20:52:57 +08:00
qazal
845a24dcc6
 		viz: group sqtt waves by program (#13187) 
* viz: group sqtt waves by program

* color the names
 2025-11-10 19:25:23 +08:00
George Hotz
fd6803000e
 		mutmut cfg (#13184) 
* mutmut cfg

* coveragerc
 2025-11-09 23:29:29 -08:00
wozeparrot
6252831ceb
 		feat: initial tk library (#13160) 2025-11-09 22:54:29 -08:00
George Hotz
925231aec1
 		repeat does less reshape for 1s (#13183) 2025-11-09 19:43:02 -08:00
George Hotz
d7369de048 hotfix: update weekly commits table 2025-11-09 19:37:06 -08:00
chenyu
6c48c87e51
 		improved ASSERT_MIN_STEP_TIME (#13182) 
* improved ASSERT_MIN_STEP_TIME

getting close, current time +1ms  then round up

* relax
 2025-11-09 16:41:12 -05:00
nimlgen
17715688c7
 		system: validate vendor for APLPCIIfaceBase (#13181) 2025-11-10 02:49:21 +08:00
nimlgen
614783693e
 		nv: remove hardcoded expansion_rom_off (#13180) 
* nv: remove hardcoded expansion_rom_off

* to max size
 2025-11-09 21:43:19 +08:00
chenyu
e1d46de8f8
 		update GROUPTOP heuristic more (#13178) 
reverts #13176
 2025-11-09 02:31:12 -05:00
chenyu
41e45c20ff
 		minor stuff reading the printed code [pr] (#13177) 2025-11-09 00:58:51 -05:00
chenyu
8e868dced8
 		only GROUPTOP one reduce kernel (#13176) 
* only GROUPTOP one reduce kernel

* ALLOWED_GATED_READ_IMAGE=148
 2025-11-08 22:38:44 -05:00
chenyu
834067d91f
 		move onnx import in compile3 (#13172) 
only used in test_vs_onnx
 2025-11-08 09:44:34 -08:00
nimlgen
7f3240dbfe
 		nv: cleanup alloc (#13170) 
* nv: cleanup alloc

* okay okay
 2025-11-09 00:14:46 +08:00
qazal
7250fc0354
 		viz: double click on kernel run goes to codegen (#13147) 2025-11-08 23:40:50 +08:00
qazal
8a7fa9e7b4
 		sqtt: show total cycles of kernel in viz (#13169) 2025-11-08 21:00:40 +08:00
chenyu
2ba8b4946f
 		external_benchmark_op_cat.py (#13168) 
* external_benchmark_op_cat.py

cat kernel that's 1ms on master and 50us with no GROUP and with NOLOCALS

* fix
 2025-11-08 01:54:10 -05:00
chenyu
a62496cb3d
 		clean up get_grouped_dims [pr] (#13159) 2025-11-08 01:53:54 -05:00
wozeparrot
eb0192b0bb
 		feat: print ranges that aren't ended (#13167) 2025-11-07 22:01:29 -08:00
George Hotz
b41541bc44
 		bounty: Remove Tensor._pool alternative implementation and verify kernels remain the same (#13164) 2025-11-07 16:59:48 -08:00
George Hotz
ffb9e8396f
 		fix indexing bug with convs 
* minimal difference for ONE_POOL=1

* fix indexing bug

* improve indexing debugger

* more debugger improvements

* always for reshape
 2025-11-07 16:45:19 -08:00
chenyu
6a509da7f3
 		Scheduler.reduceops helper [pr] (#13162) 2025-11-07 18:59:46 -05:00
George Hotz
2413311289
 		make _pool simpler (#13161) 
* make _pool simpler

* just syntax

* more correct and smaller

* try this now

* Revert "try this now"

This reverts commit 607cdc2164.

* ONE_POOL
 2025-11-07 15:58:44 -08:00
George Hotz
70054cdb14
 		move backward cast to broadcasted, expand to mixins (#13156) 
* shrink_to mixin

* move backward cast into _broadcasted

* expand to movement mixin

* move a few more

* fix spec issue
 2025-11-07 15:07:47 -08:00
George Hotz
f2519ea0ba
 		shrink_to mixin (#13155) 2025-11-07 11:46:24 -08:00
C T
0f9d7f650d
 		whisper: fix oob, explicit dtype (#13144) 
* fix dtype depending on numpy version

numpy v2 np.array returns int64 which Tensor passed through for the
first decode call, swallowing the <|notimestamps|> token and corrupting
the sequence

* fix whisper OOB

global limit on whisper's context length

* enforce whisper max_tokens_to_sample (match openai)

local limit on max tokens decoded
 2025-11-07 12:55:01 -05:00
Ahmed Harmouche
3ecff3a8da
 		Fix dim splitting bug for len(dim) == len(limited) case (#13142) 
* Fix gpudims bug on webgpu

* Fix split dim bug

* Remove webgpu_bug from examples

* Add test for shape correctness

* Fix 3D indexing

---------

Co-authored-by: chenyu <chenyu@fastmail.com>
 2025-11-07 12:31:06 -05:00
nimlgen
b8e48effcb
 		device: no compilers message with reasons (#13146) 
* device: no compilers message with reasons

* typings

* mypy
 2025-11-07 23:01:45 +08:00
nimlgen
35e461ef69
 		hcq: use exception group (#12616) 
* hcq: use exception group

* fix
 2025-11-07 21:23:12 +08:00
nimlgen
10dc8335d2
 		tinygpu: fix teardown crash (#13143) 
* tinygpu: fix crash

* um?

* double relase

* restore
 2025-11-07 19:52:54 +08:00
qazal
d4a216d7d9
 		viz: display compiler errors (#13141) 2025-11-07 18:09:50 +08:00
qazal
7e94369464
 		add helper for test_timing custom ops (#13140) 2025-11-07 17:13:55 +08:00
nimlgen
95620426d5
 		tinygpu: unmap dma when client closed (#13129) 
* tinygpu: unmap dma when client closed

* syn

* tiny fixes
 2025-11-07 16:08:43 +08:00
wozeparrot
500d7661fa
 		feat: show range len on index in viz (#13139) 2025-11-06 23:21:27 -08:00
George Hotz
bb6364d7c7
 		tuplize from linearizer behind flag (#13136) 
* remove tuplize from linearizer

* optional tuplize
 2025-11-06 20:15:03 -08:00
chenyu
bb8cf948f2
 		variation of (x%c)+(x//c)*c = x (#13135) 
when x is in the form of y//b, the idiv term might have combined
 2025-11-06 18:53:28 -05:00
George Hotz
42b34cf83d
 		bottom up linearizer (#13133) 
* bottom up linearizer

* late stores

* more complete

* remove broken heuristic

* upcast size

* opt

* more conservative

* it needs that

* disable opencl half on QCOM

* fix

* make that a real test

* cpu test okay

* ptx skip

* end is after the range
 2025-11-06 15:30:32 -08:00
George Hotz
e0d828dba8 little cleanups 2025-11-06 13:58:19 -08:00
chenyu
bfb0c0391f
 		test custom eye function (#13134) 
this version is also faster with NOOPT
 2025-11-06 14:51:55 -05:00
George Hotz
290441dd44
 		do loads early (#13131) 
* do loads early

* local and reg
 2025-11-06 09:57:09 -08:00
George Hotz
097264853d
 		very simple priority (#13130) 
* very simple priority

* still simple
 2025-11-06 09:25:28 -08:00
George Hotz
07b415e831
 		fixup op order (#13128) 
* fixup op order

* more order

* move a few more

* more

* DEBUG_LINEARIZE
 2025-11-06 08:50:04 -08:00