nimlgen/tinygrad
1
0
Fork 0
forked from tinygrad/tinygrad
Code Pull Requests Activity
10,729 Commits 334 Branches 15 Tags
914defd55d6909b746554b4dbf5aaa641abd2cc0
Commit Graph
10729 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Sieds LyklesandGitHub 914defd55d give endrange priority (#12870) 
* uncomment line

* try giving endrange priority
 2025-10-23 05:19:13 +02:00
qazalandGitHub 2f95c10702 remu new instructions / use volatile in emulator tests (#12862) 
* remu new instructions

* start moving to volatile

* test_simple works

* test_exec_mov works and lid is still here

* test_exec_cmp_vopc

* clang did s_mov_b32 exec_lo, 1

* don't hardcode v1

* support volatile in tests

* hw_test passes

* only the volatile version

* subrev saturating behavior
 2025-10-23 11:13:43 +08:00
George HotzandGitHub e718254004 simpler end (#12879) 
* simpler

* fix that
 2025-10-23 10:35:58 +08:00
wozeparrotandGitHub 6e00dec95d feat: pin openpilot 0.10.1 models (#12878) 2025-10-22 14:57:54 -07:00
wozeparrotandGitHub 3a9aa05359 feat: extra nvcc options (#12876) 2025-10-22 13:21:11 -07:00
chenyuandGitHub f0831c8c30 add 0.10.0 to comma benchmark (#12875) 
* add 0.10.0 to comma benchmark

disabled the 0.10.1 ones which are pinned to master. it does not work because benchmark uses the cached old version

* that's pinned
 2025-10-22 15:18:21 -04:00
nimlgenandGitHub e7e535cd53 amd: sqtt for gfx9 (#12844) 
* amd: start sqtt for gfx9

* writes something, but sometimes zeroes

* HEADER!

* w

* tiny

* mypy
 2025-10-23 02:31:07 +08:00
b1tgandGitHub 81108f91ee amd tc: 16x16x32 (#12874) 
* amd tc: 16x16x32

* test

* clean, test amd_cdna4
 2025-10-22 13:48:01 -04:00
George HotzandGitHub bf173c0a37 we don't support multi end yet (#12869) 2025-10-22 23:43:32 +08:00
nimlgenandGitHub a7bc0104c2 amd: clean up sqtt_stop (#12872) 2025-10-22 22:17:03 +08:00
nimlgenandGitHub b6eb9172ea amd: fix ip offsets (#12867) 2025-10-22 20:50:18 +08:00
geohot 174811fc0f hotfix: slightly looser load spec for AMD bfloat16 2025-10-22 19:55:59 +08:00
George HotzandGitHub 7762b3558b clean up the spec (#12868) 
* tighten up the spec

* move validate into a different file

* that moved to validate

* after(barr)
 2025-10-22 19:50:42 +08:00
George HotzandGitHub 726988fa4b late ifs try 2 (#12865) 
* late ifs try 2

* fix image

* fix that test

* panic

* ptx fixups

* preserve toposort

* those pass locally

* Revert "those pass locally"

This reverts commit 063409f828.

* no ls

* make that explicit
 2025-10-22 18:49:27 +08:00
George HotzandGitHub 6abe90fb7c fix linearizer non-determinism (#12866) 2025-10-22 17:51:35 +08:00
qazalandGitHub cebc2b5721 cleanup viz profiler metadata ui (#12860) 
* cleanup viz profiler metadata ui

* text

* select over .args

* space
 2025-10-22 17:31:12 +08:00
Sieds LyklesandGitHub 8d0256c46b Move gate to load for loaded index (#12861) 
* change condition

* change test to better represent how the uop looks irl
 2025-10-22 09:53:07 +02:00
chenyuandGitHub 6d86e962c7 update ASSERT_MIN_STEP_TIME (#12857) 
0.10.1 driving_policy is good now, still need driving_vision and dmonitoring to be fast
 2025-10-21 22:46:07 -04:00
George HotzandGitHub 92778c7a8b rename opts to ren, add store ranges back (#12856) 
* rename opts to ren

* fix docs and bring store back
 2025-10-22 09:15:38 +08:00
chenyuandGitHub c5cee74706 remove BLOCK_REORDER (#12854) 
not used
 2025-10-21 19:10:14 -04:00
chenyuandGitHub 0b673eddec simpler newton_schulz transpose (#12853) 2025-10-21 17:21:45 -04:00
b1tgandGitHub 60d7e232f2 cuda fp8 (#12782) 
* cuda fp8

* tensor core

* tc test

* clean

* clean pm
 2025-10-21 15:05:25 -04:00
Harald SchäferandGitHub 587ccc0e5c compile3: make selftests opt-in (#12851) 2025-10-21 11:32:27 -07:00
wozeparrotandGitHub c3149c618a feat: nvcc compiler (#12852) 2025-10-21 11:31:23 -07:00
chenyuandGitHub 8baa61bd67 use torch 2.9 and its Muon in test (#12773) 
* use torch 2.9 and its Muon in test

* relax and disable
 2025-10-21 13:35:17 -04:00
chenyuandGitHub f51f9aaa16 muon ns_params -> ns_coefficients (#12850) 
match the official torch one
 2025-10-21 12:35:52 -04:00
wozeparrotandGitHub 62e7b8b870 feat: just use compile3 (#12849) 2025-10-21 07:56:50 -07:00
nimlgenandGitHub c7336c3e31 amd: sqtt for aql (#12846) 2025-10-21 22:35:01 +08:00
George HotzandGitHub 8960ac54f3 remove RewriteStep premature optimization (#12840) 
* remove RewriteStep premature optimization

* fix ebs

* core line count
 2025-10-21 21:45:20 +08:00
Sieds LyklesandGitHub 7f798a9630 Cleanup const buffers (#12829) 
* split pm_cleanups

* update test_schedule

* shrink when we remove bufferize

* dont do shrink if shape is empty

* update tests

* remove *1 from metadata

* deal with the noop bufferize

* only noop on cvar

* cleanup

* fix if

* rename
 2025-10-21 14:53:49 +02:00
nimlgenandGitHub 1ad6598963 amd: trace all instructions (#12831) 2025-10-21 20:52:24 +08:00
sirhcmandGitHub cdc72556a1 no more brew (#12839) 2025-10-21 08:12:46 -04:00
George HotzandGitHub 20a232f1c5 bugfixes from multioutput + PCONTIG=3 for fa bw memory fix (#12837) 
* bugfixes from multioutput

* PCONTIG=3 fixes fa memory usage

* that's base
 2025-10-21 19:21:02 +08:00
qazalandGitHub 0435d31f1c viz: generic back button functionality (#12838) 2025-10-21 18:52:00 +08:00
George HotzandGitHub 7d9551ce2e move to late/control_flow.py (#12835) 2025-10-21 18:15:06 +08:00
George HotzandGitHub d711a4b933 delete old linearizer (#12834) 
* new linearizer with early endrange

* cleanups

* second stage removal

* not store

* do that later

* end cleanup

* fix globals

* end

* multi end

* fix ends earlier

* work

* do_merge_ends

* mini change

* range_gate

* fix cpu

* test fixups

* ranges on index

* not for ptx

* delete linearizer

* remove more junk

* delete that test

* we insert endif

* all ends
 2025-10-21 17:52:18 +08:00
qazalandGitHub 40633ab34d list buffer args to kernel in profiler (#12826) 
* list buffer args to kernel in profiler

* stable order

* back button works

* deselect also works
 2025-10-21 17:51:36 +08:00
George HotzandGitHub c780cd9abb new linearizer with early endrange (#12823) 
* new linearizer with early endrange

* cleanups

* second stage removal

* not store

* do that later

* end cleanup

* fix globals

* end

* multi end

* fix ends earlier

* work

* do_merge_ends

* mini change

* range_gate

* fix cpu

* test fixups

* ranges on index

* not for ptx
 2025-10-21 17:37:48 +08:00
geohot d59d4cdbe4 lil less is okay 2025-10-21 17:09:44 +08:00
qazalandGitHub 32af1ff84b viz graph drawing small cleanups (#12830) 
* viz graph drawing small cleanups

* str literal
 2025-10-21 15:51:32 +08:00
Sieds LyklesandGitHub 367fbabc30 remove Ops.SUBSTITUTE (#12827) 
* remove Ops.SUBSTITUTE

* remove from viz
 2025-10-21 08:19:42 +02:00
qazalandGitHub 57f6b6f229 style view codegen like a link in profiler (#12825) 2025-10-21 12:15:13 +08:00
qazalandGitHub 154cdfe46d viz state cleanups (#12821) 
* viz state cleanups

* more generic
 2025-10-21 11:44:51 +08:00
George HotzandGitHub a71a41f6d1 rename Ops.ENDRANGE -> Ops.END (#12824) 2025-10-21 11:32:18 +08:00
qazalandGitHub 8521fd5263 viz: hierarchical rewrites (#12805) 
* viz: hierarchical rewrites

* count of subrewrites

* arrows

* better keyboard things

* add select and deselect utils

* works

* diff

* event stopPropagation

* work

* don't change the rewrite

* walk tree back
 2025-10-21 10:55:41 +08:00
George HotzandGitHub df2f8b9295 use after on locals (#12815) 
* use after on locals

* fix estimates

* too much compute

* correct for both ptx and normal

* err, that

* tighter spec

* keep that
 2025-10-21 10:29:12 +08:00
sirhcmandGitHub 68c045bf0a NIR: Check for brew packages tinymesa and tinymesa_cpu (#12739) 
* brew install tinymesa_cpu

* brew --prefix tinygrad_cpu too

* fix brew paths

* check both brew paths

* better errors

* handle failure
 2025-10-21 09:38:43 +08:00
wozeparrotandGitHub 990e8b97ee feat: log openpilot 0.10.1 times (#12816) 2025-10-20 18:30:34 -07:00
George HotzandGitHub 565a7a6218 num_batches_tracked has shape () (#12820) 2025-10-21 09:22:39 +08:00
geohot 25beea5769 hotfix: suppress_finalizing on device __del__ 2025-10-21 09:04:36 +08:00