geohot
2e41472b02
support tc 2, all are pad
2025-08-28 14:40:50 -07:00
geohot
ac641f7b10
support tc padding
2025-08-28 14:34:42 -07:00
geohot
226c59fa5a
bugfix
2025-08-28 14:13:34 -07:00
geohot
3bbfcbccde
work
2025-08-28 13:55:07 -07:00
geohot
78a56b3461
fix some tests
2025-08-28 12:55:12 -07:00
George Hotz and GitHub
b5ac4501d4
Merge branch 'master' into working_postopt
2025-08-28 12:35:39 -07:00
George Hotz and GitHub
b9b438c516
small updates from postopt ( #11903 )
...
* tests from postopt
* modernize
* skip lin tests
* that's fixed?
* skip, not failure
2025-08-28 12:34:52 -07:00
nimlgen and GitHub
bb55a3001f
nv: flush reset message ( #11897 )
2025-08-28 22:17:20 +03:00
geohot
038d3bc295
clean up test
2025-08-28 11:39:02 -07:00
geohot
4b223c820a
fix test
2025-08-28 11:29:02 -07:00
nimlgen and GitHub
e8289c75b1
ci: do not reinstall existing pkgs in macos ( #11900 )
2025-08-28 21:20:15 +03:00
geohot
528e285d81
more tests
2025-08-28 11:15:47 -07:00
geohot
cd3dc67636
tensor cores need to pad
2025-08-28 11:04:30 -07:00
geohot
b19a8963c3
revert
2025-08-28 11:00:20 -07:00
George Hotz and GitHub
ec10e00cf5
Merge branch 'master' into working_postopt
2025-08-28 10:50:00 -07:00
geohot
d3aa38ad4a
work
2025-08-28 10:46:18 -07:00
chenyu and GitHub
134cf56904
update cache name for gpuocelot ( #11896 )
2025-08-28 13:11:10 -04:00
ea1be2e4cd
[bounty] Remove using reshape to register symbolic shape ( #11771 )
...
* Modify tests and start work towards removing symbolic reshape
* Refactor symbolic reshape
* fix small error
* much cleaner + fix more tests
* Can remove this now
* Update test_symbolic_ops and test_tiny
* Couple more tests
* Unused import
* More tests and add EXPAND to Tensor.empty
* Fix test beam search
* all int
* Fix rangeify by adding shrink
* Remove OOB check and so fix test_symbolic_jit
* test_symbolic_jit doesn't need OOB Context anymore either
* Should remove that test now
* Cleanups part 1
* fix linters
* Final cleanups
* Don't reassign inside for loop
---------
Co-authored-by: chenyu <[email protected] >
2025-08-28 12:30:49 -04:00
qazal and GitHub
53853ae49b
viz: switch to Path2D ( #11892 )
2025-08-28 18:58:16 +03:00
nimlgen and GitHub
874c1db4af
am: init support for aql ( #11888 )
2025-08-28 18:41:46 +03:00
17ecaf4682
Add test_variable_empty ( #11889 )
...
* Add test_variable_empty
* Move test and add TODO
---------
Co-authored-by: chenyu <[email protected] >
2025-08-28 11:38:27 -04:00
Nino Risteski and GitHub
54be477152
rope cache optim for jit prune in llm.py ( #11678 )
...
* rope cache optim for jit prune
* rope test
* tests in test attention
* Revert "rope test"
This reverts commit 69ede543d0 .
* lint
2025-08-28 08:31:29 -07:00
geohot
6e41040e91
get tc ranges
2025-08-28 08:24:33 -07:00
geohot
6b3d9d6663
fix some tests
2025-08-28 08:07:45 -07:00
geohot
0ece218588
fix uop test
2025-08-28 07:34:06 -07:00
quortus and GitHub
5f8fe9a331
Replace ASSIGN with STORE in test_linearizer ( #11821 )
2025-08-28 07:33:20 -07:00
geohot
feb12685f2
postopt using RKernel
2025-08-28 07:28:52 -07:00
4e8370309c
Support onnx If OP ( #11648 )
...
* start
* tiny clean up
* whoops, didn't mean to accidentally fix this
* fix .to(device), kinda hacky and this fix makes it slower?
* merge properly
* FINALLY figured out slowness, also hack pylint for now
* add DEBUGONNX print for subgraph
* oops
* WOOOOOOOO SHAPE CACHE 50% SPEED INCREASE
* small fix, but maybe all deterministic Tensor creation in fp should be cached
* cache condition
* sliiiightly cleaner
* better abstraction?
* remove sam from model_benchmark
* remove shape cache speed up for now
* less lines
* isinstance fix
---------
Co-authored-by: chenyu <[email protected] >
2025-08-28 10:17:35 -04:00
George Hotz and GitHub
6d6f0dada7
support for tuple ranges ( #11890 )
...
* support for tuple ranges
* breaks it
2025-08-28 07:02:31 -07:00
nimlgen and GitHub
60dd9a162c
memory: tiny tlsf cleanup ( #11887 )
2025-08-28 14:07:18 +03:00
chenyu and GitHub
beb5982165
FUSE_ATTENTION ( #11884 )
2025-08-27 19:59:17 -04:00
George Hotz and GitHub
cb5295168d
postrange boilerplate work ( #11881 )
2025-08-27 15:22:59 -07:00
George Hotz and GitHub
fd579433bc
pre expander shouldn't go in gpudims ( #11880 )
2025-08-27 14:52:24 -07:00
nimlgen and GitHub
44816218b5
memplan: fix large buffers planning ( #11878 )
...
* memplan: fix large buffers planning
* fix
* fix dsp
2025-08-27 23:54:27 +03:00
nimlgen and GitHub
4006366752
Revert "memplan: fix large buffers planning ( #11876 )" ( #11877 )
...
This reverts commit 7f90497efc .
2025-08-27 22:36:14 +03:00
nimlgen and GitHub
7f90497efc
memplan: fix large buffers planning ( #11876 )
...
* memplan: fix large buffers planning
* fix
2025-08-27 22:04:15 +03:00
George Hotz and GitHub
e4afdf9ea1
improve DEBUG=2 string with TB/s and TFLOPS [pr] ( #11875 )
2025-08-27 11:42:41 -07:00
Jordan Chalupka and GitHub
e9789d8a70
Add mxfp4 support ( #11873 )
...
* bump ggml url
* map mxfp4 to tensor
* tests
2025-08-27 10:56:56 -07:00
qazal and GitHub
884eb53e89
tracing: fix types ( #11871 )
...
* tracing: fix types
* /profiler isn't a thing
* return list
2025-08-27 15:50:43 +03:00
Sieds Lykles and GitHub
d39365809a
add ctx to z3_renderer arg ( #11867 )
...
* add ctx to z3_renderer arg
* update symbolic fuzzer
* rewrite u1,u2,u3
* update fuzz_fast_idiv
* remove imports
2025-08-27 03:38:15 +02:00
George Hotz and GitHub
24c00a4061
darken hex on viz ( #11865 )
...
* darken hex on viz
* more readable
2025-08-26 15:57:50 -07:00
qazal and GitHub
f38e4af226
viz: add custom zoom filter ( #11861 )
2025-08-27 01:30:29 +03:00
nimlgen and GitHub
62df6c39af
amd: correct handling of relocations ( #11863 )
...
* amd: correct handling of relocations
* ops
* add
2025-08-27 01:26:45 +03:00
George Hotz and GitHub
d261458ecd
add colors to range ( #11860 )
2025-08-26 14:32:12 -07:00
Sieds Lykles and GitHub
7dfc7e4abc
uops_to_z3 helper( #11859 )
2025-08-26 22:58:05 +02:00
chenyu and GitHub
1bbb578afd
named expression for POW and MAX gradient ( #11858 )
2025-08-26 16:03:03 -04:00
chenyu and GitHub
7028cb4167
clean up TestBitcastConstFolding ( #11856 )
2025-08-26 15:26:47 -04:00
George Hotz and GitHub
d4154e0349
split devectorizing of buf/index ( #11855 )
2025-08-26 12:05:48 -07:00
George Hotz and GitHub
b268755d51
small changes from postopt ( #11854 )
2025-08-26 11:56:16 -07:00
Sieds Lykles and GitHub
a3aeef45cc
associative variation of where branch-merging ( #11851 )
...
* add rule and test
* change comment
2025-08-26 19:27:05 +02:00