Commit graph

9,747 commits

Author SHA1 Message Date
George Hotz
48ca6d888d was dumb 2025-08-10 14:36:35 -07:00
George Hotz
b7ea16f161 localish fa 2025-08-10 14:30:25 -07:00
George Hotz
cc34518a52 RewriteNotReady 2025-08-10 13:56:38 -07:00
George Hotz
76a97e04b0 cleanups 2025-08-10 12:24:16 -07:00
George Hotz
0d64aa1f1e this does work...but with a global 2025-08-10 12:00:55 -07:00
George Hotz
9979730f3f children stuff that doesn't work 2025-08-10 10:57:54 -07:00
George Hotz
7ddcb8632f simpler 2025-08-09 08:14:02 -07:00
George Hotz
e268eb2d5c tform ffn 2025-08-08 18:29:48 -07:00
George Hotz
ee06481036 ranges 2025-08-08 18:18:27 -07:00
George Hotz
38c9b5ed2c conv hack 2025-08-08 14:36:48 -07:00
George Hotz
7249a711c2 half contig 2025-08-08 08:43:41 -07:00
George Hotz
efdf08f3e2 global rangeify 2025-08-07 15:12:00 -07:00
George Hotz
9a2f55425b global rangeify 2025-08-07 15:08:45 -07:00
George Hotz
9ed409d0f8
Merge branch 'master' into no_merge_views 2025-08-07 14:42:07 -07:00
George Hotz
82be8abfd2
move opt under codegen (#11569) 2025-08-07 14:19:17 -07:00
chenyu
702e38dc19
remove FUSE_ARANGE_UINT (#11567)
also add IGNORE_OOB=1 to bert runs. lowered BS on tinybox to 90 since 96 oom during eval without reset
2025-08-07 16:49:06 -04:00
George Hotz
6ed2dfd187
delete the arange dim mismatch restriction (#11568)
* delete the arange dim mismatch restriction

* skip that test race
2025-08-07 13:46:17 -07:00
wozeparrot
7ae4335127
feat: generate blend index (#11566) 2025-08-07 14:20:28 -04:00
chenyu
594cbdc66f
skip AM ResNet50 benchmark (#11565)
hanging with FUSE_ARANGE?
2025-08-07 14:07:01 -04:00
chenyu
aa1a6f2132
support threshold in Tensor.softplus (#11564)
fix gradient for large input
2025-08-07 13:43:18 -04:00
chenyu
7ee3770961
FUSE_ARANGE=1 (#11427)
* FUSE_ARANGE=1

* fix test

---------

Co-authored-by: George Hotz <72895+geohot@users.noreply.github.com>
2025-08-07 13:32:34 -04:00
George Hotz
4dfcfb1ae5 Revert "Revert "viz: align-center checkbox (#11555)""
This reverts commit c52facfd29.
2025-08-07 08:15:57 -07:00
George Hotz
7e42427a7b Revert "Revert "viz: remove color for unbind step (#11554)""
This reverts commit 5650c7b86c.
2025-08-07 08:15:51 -07:00
George Hotz
dc765fbeb7 Revert "viz: timeline perf (#11533)"
This reverts commit 031f26632b.
2025-08-07 08:08:51 -07:00
George Hotz
5650c7b86c Revert "viz: remove color for unbind step (#11554)"
This reverts commit 1e205775bd.
2025-08-07 08:08:50 -07:00
George Hotz
c52facfd29 Revert "viz: align-center checkbox (#11555)"
This reverts commit 91ec093464.
2025-08-07 08:08:50 -07:00
George Hotz
974cfbe76d Revert "viz: add support for colored tooltip text (#11556)"
This reverts commit b3f7ea6f93.
2025-08-07 08:08:49 -07:00
George Hotz
3bf0db80ef Revert "viz: pick the largest rect for proxy fillColor (#11558)"
This reverts commit 76079bc7f2.
2025-08-07 08:08:48 -07:00
George Hotz
9764c6cdee
fix mismatch reduce, try 2 (#11560)
* fix mismatch reduce, try 2

* fix heuristic

* delete that test

* don't start allowing ones
2025-08-07 07:57:58 -07:00
qazal
76079bc7f2
viz: pick the largest rect for proxy fillColor (#11558) 2025-08-07 16:40:17 +03:00
nimlgen
4f29a2c441
fix flaky test on macos (#11557) 2025-08-07 15:55:35 +03:00
qazal
b3f7ea6f93
viz: add support for colored tooltip text (#11556) 2025-08-07 15:04:43 +03:00
qazal
91ec093464
viz: align-center checkbox (#11555) 2025-08-07 14:22:02 +03:00
qazal
1e205775bd
viz: remove color for unbind step (#11554) 2025-08-07 14:16:21 +03:00
nimlgen
031f26632b
viz: timeline perf (#11533)
* viz: timeline perf

* progress

* fast

* less lines

* less lines

* less lines

* fix chrome
2025-08-07 13:16:17 +03:00
George Hotz
a1aa5670aa
Revert "fix mismatch reduce (#11547)" (#11549)
This reverts commit 49d21a9055.
2025-08-06 22:43:15 -07:00
George Hotz
49d21a9055
fix mismatch reduce (#11547)
* fix mismatch reduce

* cleanups

* fix shape

* fix mypy

* resolve
2025-08-06 21:12:51 -07:00
George Hotz
b8791e962c don't merge views, mops in kernel 2025-08-06 17:23:31 -07:00
George Hotz
21570545d3
move view pushing to codegen, try 2 (#11534)
* move view pushing to codegen, try 2

* fix up some linearizer tests

* fix test search

* fix test schedule

* delete that test

* fix test arange

* fix a few tests

* update tests

* push views

* ebs cleanup

* fix local/reg

* test and lint

* fix more tests

* test cleanups

* skipped that one
2025-08-06 15:58:38 -07:00
wozeparrot
2d5bdc939d
faster llama3 dataloader (#11540) 2025-08-06 18:25:57 -04:00
George Hotz
80d9cced07
more test cleanups (#11544)
* more test cleanups

* revert that
2025-08-06 15:05:21 -07:00
George Hotz
6fd1332763
update some tests for less Kernel (#11543)
* update some tests for less Kernel

* get_program update
2025-08-06 14:19:59 -07:00
George Hotz
09dc7af8e9
move bind to big graph (#11539)
* move bind to big graph

* fix tests

* unbind inside kernel only

* merge views

* fix multitensor

* failure text change
2025-08-06 13:27:51 -07:00
George Hotz
7c5e115747
test_mismatch_reduce (#11538) 2025-08-06 10:02:14 -07:00
George Hotz
4fe11725c6
pass through sink arg, update linearizer test (#11536)
* pass through sink arg, update linearizer test

* get_program help

* bump line count

* use new api
2025-08-06 09:48:48 -07:00
George Hotz
bfebb5c37b
do store in the replace_buffers (#11535) 2025-08-06 08:42:45 -07:00
geohotstan
1163292759
move onnx_parser into onnx (#11530) 2025-08-06 10:46:27 -04:00
George Hotz
7b16fadd87
load view late + simpler rewrite (#11525)
* add the load view later

* simpler replace buffers

* rewrite name
2025-08-06 06:55:11 -07:00
nimlgen
930d8dae0c
hcq: lazy prof signal allocation (#11531) 2025-08-06 15:28:11 +03:00
nimlgen
eafc7fda12
upd perfetto (#11528) 2025-08-06 14:00:34 +03:00