github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-01-09 15:08:02 -05:00
Code Issues Packages Projects Releases Wiki Activity
11,299 Commits 203 Branches 13 Tags
572dfd5506ac262fcf903e108e88e6b2edc66ce6
Commit Graph

11299 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
qazal
572dfd5506 add static amd program info to viz (#13594) 
* llvm-readelf

* amd_readelf + soft_err

* cleanup

* multiple metadata

* max wgp size, may be less
 2025-12-08 04:08:14 +08:00
qazal
73093314bd viz: support list of sidebar info (#13612) 2025-12-08 03:09:43 +08:00
chenyu
b981b6f89e remove old llama grad_acc (#13611) 
* remove old llama grad_acc

* GRADIENT_ACC_STEPS=1
 2025-12-07 13:03:47 -05:00
Christopher Milan
94d7646bdc fix anonymous struct fields (#13610) 2025-12-07 12:56:38 -05:00
nimlgen
dcd50baca4 amd/nv: cleanup (#13608) 2025-12-07 17:05:26 +03:00
nimlgen
ac5f1e115d autogen: repro for the bug (#13607) 
* autogen: repro for the test

* mute
 2025-12-07 15:51:03 +03:00
Christopher Milan
4eae4b0ce6 unify adreno autogen with mesa (#13604) 
* unify adreno autogen with mesa

* gen pm4

* TestTiny::test_plus works

* add a6xx enums

* IMAGE=2 TestTiny::test_gemm works

* remove adreno from CI

* cleanup
 2025-12-06 15:17:36 -05:00
kamilisjon
e20bc0b9b5 remove unused function parameter in beam search (#13602) 2025-12-06 11:40:47 -05:00
nimlgen
abafb96441 hcq: check all subbufs are free (#13599) 
* hcq: check all subbufs are free

* fix

* Update ops_amd.py
 2025-12-06 17:43:18 +03:00
nimlgen
f2b549d921 amd: refactor scratch calc (#13595) 
* amd: refactor scratch calc

* fix
 2025-12-06 16:41:35 +03:00
chenyu
4562f217e1 more bert updates (#13597) 
prep split jit
also lower BS to 72
 2025-12-06 08:32:43 -05:00
wozeparrot
93f1baca77 feat: tk fa in tensor (#13580) 2025-12-05 14:36:29 -08:00
chenyu
cb4c6324ef revert bert grad accumulation (#13596) 
prep for the new split jit style
 2025-12-05 17:30:08 -05:00
qazal
f20212e1ec refactor viz error handler (#13593) 2025-12-06 02:37:39 +08:00
Christopher Milan
dec2f50aee reenable process replay for lvp (#13592) 2025-12-05 12:36:35 -05:00
chenyu
0977206b1c Revert am (#13591) 
* Revert "hotfix: amd: tmpring (#13589)"

This reverts commit 4d8b283b36.

* Revert "amd: use correct structs (#13583)"

This reverts commit d8b09eda57.
 2025-12-05 11:03:12 -05:00
chenyu
ac1227575f IMAGE=1 driving_vision in benchmark (#13587) 2025-12-05 10:20:54 -05:00
nimlgen
4d8b283b36 hotfix: amd: tmpring (#13589) 
* hotfix: amd: tmpring

* more
 2025-12-05 18:19:05 +03:00
qazal
8c332219f9 viz: remove x86asm highlighter (#13586) 
* viz: remove x86asm highlighter

* formatting
 2025-12-05 21:05:50 +08:00
qazal
5d8726d8d2 viz: refactor to generic sidebar (#13584) 2025-12-05 20:09:41 +08:00
nimlgen
d8b09eda57 amd: use correct structs (#13583) 2025-12-05 14:46:38 +03:00
qazal
6d92e9ffbf hotfix: skip process replay on lvp (#13585) 2025-12-05 19:25:23 +08:00
Christopher Milan
8011b953c9 mesa: remove glsl type hack (#13578) 
* mesa: remove glsl type hack

* lazy type access

* save a line

* fix windows?

* mypy happy
 2025-12-04 21:18:56 -05:00
George Hotz
c5bd28e21d start work on schedule cache (#13529) 
* start work on schedule cache

* local unique

* schedule cache works

* schedule cache cleanup

* fix tests

* preserve metadata

* oops, fix cache

* put that there

* fix spec

* always miss

* why is that broken?

* src[0].op

* fix process replay

* delete abstractions2

* reenable the actual schedule cache

* metadata is best effort

* fix JIT in examples/gradaccum_mnist.py

* full jit

* fixed and test is real
 2025-12-04 17:24:49 -08:00
wozeparrot
62e2fc5108 tk: global load/store rv (#13577) 2025-12-04 17:23:48 -08:00
Christopher Milan
5cfe1698e8 autogen: strip function parameter qualifiers (#13576) 
* autogen: strip function parameter qualifiers

* regen hip

* re-regen hip
 2025-12-04 19:54:34 -05:00
qazal
f21c9dbf4b enable PMC with VIZ=2 (#13575) 2025-12-05 03:09:53 +08:00
qazal
d7caae5f61 viz: tabulate pmc (#13574) 
* viz: tabulate pmc

* linter

* enable nesting

* pmc comes before waves
 2025-12-05 03:08:39 +08:00
chenyu
42f6cf3a90 tighter test_real_world mem and kernel count bounds (#13573) 
also check if actual usage is within 20% of set limit, the old limits are too big to be useful
 2025-12-04 13:35:39 -05:00
chenyu
89f9e1dcd5 add SGD to beautiful_mnist (#13571) 2025-12-04 12:17:29 -05:00
qazal
512a8f3dd4 viz: start global memory PMC tests (#13569) 2025-12-05 00:40:27 +08:00
chenyu
7df56d3b99 Optimizer.device is a property (#13568) 2025-12-04 09:25:15 -05:00
nimlgen
db99a61fad qcom: support cpu mappings (#13565) 
* test

* qcom: support cpu mappings

* clean

* msg
 2025-12-04 14:50:46 +03:00
George Hotz
bd6a068ef7 move track_rewrites to outer schedule cache (#13556) 
Co-authored-by: qazal <qazal.software@gmail.com>
 2025-12-04 19:13:45 +08:00
qazal
3eae146139 faster process replay [pr] (#13564) 2025-12-04 18:52:07 +08:00
Rory Clear
6eab756578 fix and test loading num_batches_tracked (#13538) 
* fix and test loading num_batches_tracked

* add failing reverse case

* try reshape state dict if mismatch

* reshape for () and (1,)

---------

Co-authored-by: George Hotz <72895+geohot@users.noreply.github.com>
 2025-12-04 01:22:49 -08:00
nimlgen
877a7fdd61 jit: support encdec (#13563) 
* jit: support encdec

* fix
 2025-12-04 11:58:34 +03:00
Douglas Nyberg
a8a62bc08e add max/min reduction support to ScatterND (#13562) 2025-12-04 00:53:47 -08:00
ayanhan
edf929ec9d fix: add __delitem__ to Tensor with proper TypeError (#13561) 2025-12-04 00:53:08 -08:00
Douglas Nyberg
9411ecedc4 fix CUDA half-precision trunc() type mismatch (#13559) 2025-12-03 21:53:16 -05:00
ayanhan
92b40290c7 fix: add test_sum_int and remove outdated TODO in test_custom_kernel (#13560) 2025-12-03 21:51:58 -05:00
Christopher Milan
0a54434b15 mitigate ctypes c_bool bitfield bug (#13558) 
* mitigate ctypes c_bool bitfield bug

* don't delete old test
 2025-12-03 20:46:04 -05:00
George Hotz
96d16675fe update examples/gradaccum_mnist.py to use the JIT 2025-12-03 16:11:42 -08:00
George Hotz
24ca8eeaa7 small fixups from schedule_cache (#13557) 2025-12-03 15:41:16 -08:00
Douglas Nyberg
f5abd38132 remove tfa dependency: use keras.optimizers.Lamb and tf.raw_ops for LARS (#13555) 2025-12-03 17:48:27 -05:00
George Hotz
a4c4e48385 add LUNIQUE op (#13554) 2025-12-03 14:34:34 -08:00
George Hotz
a909cd4581 faster HEVC decode (#13552) 
* faster HEVC decode

* bind to variables

* cleanups

* more cleanups
 2025-12-03 11:33:05 -08:00
chenyu
22777a89ea minor test_uop_symbolic updates (#13551) 2025-12-03 13:17:44 -05:00
chenyu
a205f98ef4 tighter bound for MOD (#13550) 2025-12-03 11:24:29 -05:00
nimlgen
fcdb01abe7 hip: fix ioctl (#13548) 2025-12-03 16:40:43 +03:00