George Hotz
b796bbae87
fix valid in indexing tests ( #16087 )
2026-05-07 14:11:28 -07:00
wozeparrot
4d1a9dca41
fix: don't copy precompiled custom kernel outputs ( #16084 )
2026-05-07 14:02:38 -07:00
qazal
f9083cf901
use subactions for benchmark.yml process replay [pr] ( #13396 )
2026-05-08 03:46:25 +09:00
nimlgen
2f0aa884d5
tinygpu: minimal is macos13 for resets ( #16075 )
2026-05-07 21:25:56 +03:00
chenyu
072db9924c
div to mixin ( #16078 )
...
also deleted idiv method
2026-05-07 12:52:37 -04:00
chenyu
516b00e286
mod and fmod to mixin ( #16077 )
2026-05-07 12:13:39 -04:00
qazal
a9a87ad8fd
viz/cli: less flags ( #16076 )
...
* viz/cli: merge -s and -i flags
* only -t
* merge parser
* fix
2026-05-08 00:22:40 +09:00
qazal
f813a04b3f
viz: pickle path in str ( #16073 )
2026-05-07 18:49:21 +09:00
wozeparrot
730fa66bf3
llama speed 6 ( #16071 )
2026-05-06 20:51:03 -07:00
Christopher Milan
7b91f7c90c
nv: look for gsp firmware in /lib/firmware ( #16068 )
2026-05-06 21:35:47 -04:00
George Hotz
8e84317743
the renderer part of gate moving from index to load/store ( #16064 )
...
* the renderer part of gate moving from index to load/store
* fixed
* fix gated stores
* fix spec
* better?
* Where after gated load becomes alt value
* cleaner expression
* fix python backend
* remove dead code
2026-05-06 13:47:04 -07:00
chenyu
ef085304bc
stronger divmod_recombine ( #16066 )
2026-05-06 15:41:54 -04:00
qazal
d7d32d82ee
viz/cli: print first uop with DEBUG=6 ( #16065 )
...
* viz/cli: print first uop with DEBUG=6
* rename fmt to emit
* define inst
2026-05-07 03:39:34 +09:00
chenyu
af4140f3be
fix divmod recombine for floordiv ( #16062 )
2026-05-06 14:22:42 -04:00
chenyu
c6ad3d3ac2
better divmod late rewrite ( #16061 )
...
better order
2026-05-06 11:31:48 -04:00
chenyu
aaabe42373
relax fold_divmod_general ( #16058 )
2026-05-05 21:37:56 -04:00
Christopher Milan
1de14cf33a
am: autogen soc ( #16055 )
2026-05-05 20:39:43 -04:00
chenyu
869eae6b37
fix double div rewrites ( #16054 )
2026-05-05 19:34:35 -04:00
Christopher Milan
bd06ea9f97
am: simplify import_module ( #16046 )
2026-05-05 19:25:53 -04:00
qazal
795501e1da
fix device in null graph events ( #16053 )
...
* failing test
* fix compute
* fix sdma
2026-05-06 07:44:08 +09:00
wozeparrot
ab6218bc92
llama mp fixes ( #16050 )
2026-05-05 15:35:32 -07:00
chenyu
34fe37d64e
use FLOORDIV and FLOORMOD ( #16048 )
...
* use FLOORDIV and FLOORMOD
also removed CORRECT_DIVMOD_FOLDING
* fix
* Revert "fix"
This reverts commit 86af33b88ef31943c61e67189b072eca4896409a.
* fix
* fix
2026-05-05 18:32:54 -04:00
Christopher Milan
76ff378007
autogen: fewer apt dependencies ( #16049 )
2026-05-05 17:22:41 -04:00
nimlgen
5fa0016ffc
supports_exec_item -> supports_uop ( #16033 )
2026-05-05 22:41:13 +03:00
qazal
cee17e0d2f
viz: fix diff color ( #16045 )
2026-05-06 03:40:53 +09:00
chenyu
9c37a0c75d
Ops.FLOORDIV and Ops.FLOORMOD ( #16038 )
...
* Ops.FLOORDIV and Ops.FLOORMOD
lowered into IDIV and MOD in get_late_rewrite_patterns
* still need this
* exclude
* like that?
2026-05-05 11:42:14 -04:00
qazal
d79bf356c2
viz: add CALL -> codegen link ( #16044 )
...
* work
* cleaner
* details
* rm
2026-05-05 23:34:44 +09:00
Christopher Milan
1c8cb0769a
am: autogen asic_regs ( #16004 )
2026-05-04 22:52:07 -04:00
George Hotz
26406bed83
amd uses .valid, not index src valid ( #16042 )
2026-05-04 18:35:15 -07:00
chenyu
a357a0449a
Tensor.div cleanup ( #16041 )
2026-05-04 19:27:36 -04:00
nimlgen
5b4f62519d
cache buffer_views as well ( #16039 )
...
* cache buffer_views as well
* reuse
* back
* x
2026-05-05 00:00:09 +03:00
Christopher Milan
8e99c4f097
fetch checks sha256 ( #16037 )
2026-05-04 16:08:38 -04:00
George Hotz
1884f67a39
simplify full_rewrite_to_sink spec ( #16035 )
...
* simplify full_rewrite_to_sink spec
* test cleanups
2026-05-04 11:44:13 -07:00
chenyu
a4fccd23b2
remove kwargs in UOp.vectorize [pr] ( #16034 )
2026-05-04 12:46:38 -04:00
qazal
b1d88ebf02
viz/cli: aggregate flops in -t ( #16031 )
...
* 38
* plumbing
* more flops
* flop/s and bytes/s
* arithmetic mean
* tests
* harmonic mean
* range
* better
* simplify
* fix prints
* no string parsing needed
2026-05-04 17:35:02 +03:00
qazal
c02e390c2b
viz: encode flops, mem and metadata in json ( #16032 )
...
* gate print
* update everywhere to check path
* server encodes json
* ui changes
* cli changes
* tests never need regex
* no str replace
* update test_pipes
* remove that
2026-05-04 23:06:18 +09:00
bigyoshi
4024d8438f
runtime/graph: avoid core_id runtimevar merge conflicts ( #16026 )
...
Co-authored-by: bigyoshi51 <269989564+bigyoshi51@users.noreply.github.com >
2026-05-03 19:16:02 +03:00
qazal
9684334dfe
viz: fix flops in graph, add null graph tracing ( #16024 )
...
* min repro, todos
* null graph tracing
* work
* work
* work
* only test_flops
* exec points back
* first
* better
* integral timestamps maybe
* cleanup
* simpler, update NULL to use SDMA naming
* integration test
* sdma
2026-05-03 22:32:44 +09:00
wozeparrot
419d525553
feat: handle multioutput kernel grads ( #16028 )
2026-05-02 22:31:45 -07:00
mefengl
9717d3a3a2
hotfix: prepend LD_LIBRARY_PATH to DLL posix search dirs ( #16023 )
2026-05-02 20:45:19 +03:00
qazal
7daf4b7d52
viz: split cli test ( #16015 )
...
* viz: split cli test
* arg3 is msg
2026-05-03 01:47:11 +09:00
nimlgen
d65b8ca25f
jit: remove *input_list from the graph sources ( #16021 )
2026-05-02 14:42:47 +03:00
qazal
7dae9e6f7f
viz: keep VIZ.value = 0 during python shutdown, cleanup launch ( #16022 )
...
* viz: keep VIZ.value = 0 during python shutdown, cleaner execv
* rm
2026-05-02 20:35:53 +09:00
Christopher Milan
637bdd5530
am: only support CDNA3/4 and RDNA3/4 ( #16017 )
2026-05-02 00:02:14 -04:00
George Hotz
4a2e1f1076
STORE doesn't have ranges anymore ( #16019 )
...
* STORE doesn't have ranges anymore
* fix
2026-05-01 15:00:27 -07:00
chenyu
0bffbc5f8a
onnx fmod uses fmod ( #16018 )
2026-05-01 16:47:11 -04:00
chenyu
782d1ff80f
Tensor.fmod ( #16014 )
...
c-style mod matches torch
2026-05-01 16:02:18 -04:00
nimlgen
1079441332
revoke bus master ( #16007 )
2026-05-01 18:00:01 +03:00
qazal
8b147a9ed5
minimal repro for llama copies 2 ( #16011 )
2026-05-01 22:23:47 +09:00
qazal
a29dd7b19b
Revert "cleanup: untrack wait Metal buffers ( #15954 )" ( #16010 )
...
* Revert "cleanup: untrack wait Metal buffers (#15954 )"
This reverts commit 5eb1fd5d3c
2026-05-01 21:18:19 +09:00
