github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-04-29 03:00:14 -04:00
Code Issues Packages Projects Releases Wiki Activity
12,481 Commits 248 Branches 14 Tags
abd830b260affb9099eeeab3944b32aa0daafeee
Commit Graph

12481 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
nimlgen
abd830b260 am: setup_rinf returns only doorbell (#15112) 2026-03-03 19:27:41 +03:00
nimlgen
4b42bb54aa am: reset sdma to start from 0 (#15109) 2026-03-03 18:14:46 +03:00
George Hotz
01ddb4c267 add precompile to call (#15099) 
* add precompile to call

* put get back

* something

* after structure

* alt

* keep it call

* resolve call

* resolve linear call

* precompile works with llm

* revert rangeify

* color for debugging

* getenv PRECOMPILE

* clean up deco pattern

* fully recursive sink scheduling

* revert llama

* fix SPEC=2
 2026-03-03 22:32:42 +08:00
qazal
c7f908b788 sqtt: fix rdna4 structs (#15111) 
* work

* DEBUG=2
 2026-03-03 23:32:14 +09:00
qazal
8dd691761d sqtt: remove old files (#15108) 2026-03-03 22:43:24 +09:00
Christopher Milan
de043226ba benchmark comma usbgpu driving_vision step and load time (#15103) 
Co-authored-by: Comma Device <device@comma.ai>
 2026-03-03 06:08:03 -05:00
Christopher Milan
5f6b610da1 FLOAT16 logic for IMAGE==1 goes back to image_conv2d (#15105) 2026-03-03 05:37:57 -05:00
wozeparrot
529318259c fix: fix null tests to actually use null device (#15104) 2026-03-03 02:05:47 -08:00
George Hotz
7d025089e3 no after removal (#15102) 
* no after removal

* we are using walk

* null schedule test

* pytest deps

* Revert "pytest deps"

This reverts commit 5e1c5304ec.

* Revert "null schedule test"

This reverts commit 02da66053e.

* clean null tests
 2026-03-03 17:50:31 +08:00
wozeparrot
92c16810ac feat: per device mem_used (#15100) 2026-03-03 01:31:28 -08:00
qazal
e3a0598d0b viz: the whole pc should be in view (#15101) 2026-03-03 17:17:53 +09:00
b1tg
a9ea36de79 assembly/amd: v_cmp_lg_f32 is ordered not-equal (#14982) 2026-03-03 15:37:48 +08:00
wozeparrot
c35de9bd68 asm_gemm: support more sharding (#15002) 2026-03-02 23:16:37 -08:00
wozeparrot
824ba4386a llama3 dp fix (#15098) 2026-03-02 22:43:07 -08:00
chenyu
5dcf29b1a0 use clone in test_swap_slices (#15096) 2026-03-02 22:05:12 -05:00
Christopher Milan
c70e8af068 move IMAGE FLOAT16 logic to allocations (#15095) 
* FLOAT16 logic in allocations

* cleanup

* separate that

* only apply when IMAGE == 1

* test passing now

* create image buffers earlier
 2026-03-02 22:00:05 -05:00
George Hotz
d483e4153a buffer view is like buffer (#15082) 
* buffer view is like buffer

* fix

* swap_reshape_shrink

* contiguous on gguf, fix overlap

* revert that

* _device_supports_view

* this

* fix that test

* 0 buffers

* that test was wrong

* this

* check correct size

* contig BUFFER_VIEW

* this

* fix tests

* buffer view tests

* om

* fix torch

* no MOCKGPU

* skip
 2026-03-03 09:52:33 +08:00
qazal
62ee976c1b gemm/asm: cleanup repeated patterns to helper functions (#15094) 2026-03-03 08:14:47 +09:00
qazal
848f5cea96 viz: sqtt instruction packet trace (#15065) 2026-03-03 07:55:04 +09:00
chenyu
14d1c5fdfd assign fusion tests on detach and contiguous_backward (#15092) 2026-03-02 15:21:51 -05:00
nimlgen
dfa180413d tbgpu: sign nv (#15087) 2026-03-02 22:58:30 +03:00
chenyu
71f228f80f test exact kernel count in torch_backend/test_kernel_fusion (#15091) 2026-03-02 14:26:32 -05:00
chenyu
f80b1033c5 simpler Tensor.all (#15089) 
same generated kernel
 2026-03-02 11:08:55 -05:00
chenyu
4008f7d4e8 move Tensor.one_hot +1 to python (#15088) 2026-03-02 10:56:41 -05:00
nimlgen
dafbe9733a am: cleanup (#15086) 2026-03-02 17:06:21 +03:00
qazal
f7aeff6061 viz: cli.py cleanups, do not require PYTHONPATH (#15085) 
* cleanup the print

* sys.exit

* equal check

* cleanup unpacker

* cli doesn't need PYTHONPATH

* no semicolons

* %s/PYTHONPATH=. //g
 2026-03-02 19:24:38 +09:00
George Hotz
5ff278446c add contiguous_view_offset (#15084) 
* add contiguous_view_offset

* no int
 2026-03-02 18:05:04 +08:00
Christopher Milan
977c270774 IMAGE=1 kernel count failing tests (#15083) 2026-03-02 04:35:26 -05:00
George Hotz
3539693555 Support triu variable on diagonal + SDPA symbolic (#15081) 
* triu variable

* fails

* dumbbb

* no commutative in reshape

* real fix

* revert that

* sdpa symbolic tests
 2026-03-02 12:19:48 +08:00
wozeparrot
a4f6365929 llama3: fstep takes grads (#15069) 2026-03-01 20:05:07 -08:00
Nick
8e8e9f6ff6 assert removal for _tri() + tests (#15073) 
* assert removal for _tri() and tests

* removed import

* tests triu/tril like in prefill

---------

Co-authored-by: George Hotz <72895+geohot@users.noreply.github.com>
 2026-03-02 10:34:28 +08:00
nimlgen
ccbbca05ef beam: add dev_timeout for am (#15063) 
* beam: add dev_timeout for am

* all covered

* fk

* x

* fuzz

* reset

* f
 2026-03-01 16:57:29 +03:00
chenyu
8cb4368967 delete unused END NOOP rule [pr] (#15077) 2026-03-01 00:09:05 -05:00
chenyu
efce99adc9 skip isComposing key press in llm.py (#15076) 
for the CJK input user
 2026-02-28 20:31:53 -05:00
chenyu
103ea16ec0 add contiguous back to svd (#15074) 
can cause infinite loop
 2026-02-28 16:49:26 -05:00
chenyu
fe0fa8333b Revert "improve Tensor.sort indices (#15070)" (#15072) 
This reverts commit e3003631f2.
 2026-02-28 14:40:30 -05:00
chenyu
e3003631f2 improve Tensor.sort indices (#15070) 
* improve Tensor.sort indices

instead of N^2 match at the end, have an arange to start and go through the same N(logN)^2 path

* contiguous
 2026-02-28 14:16:16 -05:00
wozeparrot
cfc5cf65ad llama3: vocab padding fix + jit copies on fakedata (#15067) 2026-02-28 08:44:55 -08:00
chenyu
76170d035a relax atol for test_xlm_roberta_large (#15066) 2026-02-28 11:22:35 -05:00
qazal
cfb8e6922d viz: arrow keys move through time (#15064) 
* work

* automatic zoom, keeping scale

* the whole shape should be out of view
 2026-02-28 23:52:36 +09:00
nimlgen
9b3450c9da test gpu crash on cdna (#15062) 2026-02-28 13:17:59 +03:00
nimlgen
6bbf813dd3 ci: switch to tinygrad/amdcomgr_dylib (#15061) 2026-02-28 13:09:39 +03:00
nimlgen
77846300b2 am: reset vm fault (#15060) 2026-02-28 12:58:56 +03:00
George Hotz
dc54441e1f add better printing to tinygrad.apps.llm (#15059) 
* add better printing to tinygrad.apps.llm

* add gc.collect

* comment
 2026-02-28 16:38:50 +08:00
George Hotz
bb84e389cf functions for llama trainer (#15045) 
* functions for llama trainer

* function there

* axis match

* fix multi

* lil cleaner

* there's a bug with HK_FLASH_ATTENTION

* training functions

* for commit
 2026-02-28 12:15:18 +08:00
chenyu
9b4ba3f838 remove ReduceContext.range_to_ends [pr] (#15055) 
* remove ReduceContext.range_to_ends [pr]

make merge_reduce_ends pure. this state is causing issue when introducing more reduce merging rewrites

* tag
 2026-02-27 22:15:44 -05:00
chenyu
151608aa90 update test_multiple_to_single_device (#15056) 
follow up to #14482, add SCACHE=0 to the test
 2026-02-27 21:44:33 -05:00
chenyu
5fd06f4f02 differentiable setitem (#15054) 
* differentiable setitem

go through the where path for bw

* no return
 2026-02-27 17:25:15 -05:00
chenyu
db6b3e1edc fix mixed setitem with both basic and tensor indexing (#15050) 2026-02-27 15:35:48 -05:00
chenyu
c9f6d8751b don't remove_bufferize for Invalid (#15053) 
* don't remove_bufferize for Invalid

* replaced
 2026-02-27 15:16:09 -05:00