nimlgen
6838b35cff
mockgpu: hevc ( #14606 )
...
* mockgpu: hevc
* eng
2026-02-07 12:27:55 +03:00
chenyu
884592f6c8
pin z3-solver version ( #14605 )
...
found exact input that crashes z3 4.15.4
2026-02-06 22:49:31 -05:00
George Hotz
7a2a3b5c71
Remove Ops.KERNEL, it's all Ops.CALL now ( #14603 )
2026-02-07 10:21:54 +08:00
George Hotz
ca6604eae2
kernel is call ( #14577 )
...
* call is kernel
* closer
* fix bugs
* dedup
* pm_gate_kernel_sink
* better
* Revert "better"
This reverts commit b4c799b810e53f094ce7
2026-02-07 10:10:14 +08:00
Christopher Milan
ad9e2f0de7
decompose bf16 ( #14601 )
2026-02-06 19:24:09 -05:00
Christopher Milan
7bb45e7df0
decompose fp8 to bigger floats [skip_process_replay] ( #14554 )
...
* decompose fp8 also
* it works
* cleanup
* no shift required
* default to float
* cleanup
* fixes
* fp8e5m2
* don't rely on behavior comparing nans
* cleanup
2026-02-06 19:05:40 -05:00
chenyu
7d193a6e26
fix wgsl bitcast ( #14600 )
...
was wrong for signed int
2026-02-06 16:57:36 -05:00
chenyu
b9fe8b7591
fix opt in process replay [pr] ( #14599 )
2026-02-06 16:49:56 -05:00
chenyu
197ebcbbbc
log seed with flush=True in fuzz_symbolic ( #14597 )
...
* log seed with flush=True in fuzz_symbolic
i think z3 can crash. added reading seed from argv to see if we repro later
* fuzz_symbolic_symbolic_div
2026-02-06 15:03:57 -05:00
qazal
a80fb4e641
viz: better ordering of device engines in profiler ( #14590 )
2026-02-06 23:08:09 +09:00
nimlgen
fbeb978170
diff devices for sdma ( #14589 )
...
* start
* x
* fix
* sdma
* c
* clean
* x
* hm
* cleaer
2026-02-06 16:39:12 +03:00
George Hotz
03af2404e2
small changes and test fixes from kernel is call ( #14586 )
2026-02-06 17:08:33 +08:00
George Hotz
3c26ce29b2
make disk tensor tests process safe ( #14584 )
2026-02-06 15:39:55 +08:00
qazal
cf73d7e2a7
hotfix: disable slower asm gemm shape from llama seqlen 8192 ( #14582 )
2026-02-06 15:05:19 +09:00
chenyu
15d3344d9e
use int inputs in test_assign ( #14580 )
...
int is less flaky
2026-02-06 00:07:31 -05:00
chenyu
b09dc646f5
revert some late_buffer_view change ( #14578 )
...
revert #14478 which breaks tinyfs
2026-02-05 22:51:40 -05:00
George Hotz
6cbcf98627
KernelInfo is required on get_program ( #14571 )
...
* rangeify always adds KernelInfo
* fix tests
* skip flaky test
2026-02-06 10:49:27 +08:00
chenyu
79b7799dba
clean up linearize schedule [pr] ( #14565 )
...
* clean up linearize schedule [pr]
don't mix ScheduleItem and UOp in schedule queue
* ok
2026-02-05 15:24:09 -05:00
chenyu
41a179f542
fix test_xlm_roberta_large ( #14564 )
...
onnxruntime does not allow symlink that's outside model dir. update snapshot_download to use local_dir instead of cache_dir. some ad hoc migration step to copy the existing model too
2026-02-05 14:56:06 -05:00
Christopher Milan
aa9dc50577
dtype decomps don't require bitshifts ( #14542 )
...
* dtype decomps don't require bitshifts
* simplify shr/shl
* ruff
2026-02-05 14:42:30 -05:00
chenyu
2b47a9a1b5
skip test_xlm_roberta_large ( #14563 )
...
symlink model not allowed in latest onnxruntime
2026-02-05 14:00:24 -05:00
qazal
190042358f
llama: faster bf16 matmul / rope backward ( #14558 )
2026-02-05 23:57:25 +09:00
George Hotz
43e7eda4e7
grad_b uses custom gemm ( #14550 )
...
* grad_b uses custom gemm
* fix multi backward, acc is in float32
* test_gemm_batched
* square gemm
---------
Co-authored-by: qazal <77887910+Qazalin@users.noreply.github.com >
Co-authored-by: qazal <qazal.software@gmail.com >
2026-02-05 15:22:27 +09:00
qazal
f9cfb64cd9
test asm_gemm in CI ( #14551 )
...
* test asm_gemm in CI
* default float16
* use a smaller shape for multi
* smaller size
* smaller for CI
* smaller for ci
* need half
2026-02-05 13:32:22 +09:00
chenyu
03d0fa9c3f
merge as_buf into buf_uop [pr] ( #14541 )
2026-02-04 16:32:23 -05:00
chenyu
d57d24c7d4
Buffer.as_buffer -> Buffer.as_memoryview [pr] ( #14535 )
...
it casts to memoryview. also inline the as_typed_buffer checks to Tensor._data
2026-02-04 11:31:11 -05:00
chenyu
67f91e897b
UOp.is_contiguous -> UOp.has_buffer_identity [pr] ( #14530 )
...
one more confusing buffer related method, but it's definitely not is_contiguous
2026-02-04 09:21:26 -05:00
Christopher Milan
8c3c026d86
decomp float16 to float32 ( #14417 )
...
* decomp float16 to float32
* denormals arent zero
* add test
* denormals are zero
* fix
* oops
* bitcast works
* fix LOADs
* test_dtype passing
* cleanup
* mypy
* debug print
* only emulate if EMULATED
* very ugly, but passes spec
* add test_dtype_alu tests
* Revert "very ugly, but passes spec"
This reverts commit fdc3999b65
2026-02-04 01:37:47 -05:00
chenyu
9c2fc118ef
relax setitem target check ( #14526 )
...
old check was too conservative
2026-02-03 22:32:49 -05:00
nimlgen
2f55005ad9
qcom: sync cpu cache when from_blob ( #14518 )
...
* um
* fx
* d
* x
* x
* x
* x
* f
* ren
2026-02-03 21:51:03 +03:00
George Hotz
d59e6e7a37
move more tests to test/null, split some existing ones ( #14512 )
...
* move more tests to test/null, split some existing ones
* null work
* null work
* move more
* fixes
* move PIL
* PIL in CLIP
* don't move that
2026-02-03 20:20:20 +08:00
qazal
5c1d21349e
viz: profiler command line tool ( #14515 )
2026-02-03 19:51:25 +09:00
George Hotz
dd2de4f838
rename all DEFINE_GLOBAL to PARAM ( #14511 )
2026-02-03 15:09:38 +08:00
George Hotz
dc77b3318b
move files that pass with NULL=1 to test/null ( #14508 )
...
* move files that pass with NULL=1 to test/null
* fix windows
* cpu 0
* bugfix + durations
2026-02-03 13:52:36 +08:00
George Hotz
888819ee09
call autodiff gradient ( #14510 )
2026-02-03 13:51:02 +08:00
wozeparrot
bbcd3d67a3
fa: faster ( #14453 )
2026-02-02 21:34:17 -08:00
chenyu
3c5845e8a5
remove cut_store_range ( #14505 )
...
special scheduling for CPU
2026-02-02 21:58:36 -05:00
chenyu
4f2e7aed24
fix multiple REDUCE on same RANGE ( #14504 )
...
each RANGE maps to one END, but reduce_to_acc is local and would not know this
2026-02-02 20:42:09 -05:00
chenyu
66d2b02f11
delete files that depends on extra.optimization.helpers ( #14499 )
2026-02-02 13:33:33 -05:00
George Hotz
ec0398fceb
test amd gpu crashes ( #14459 )
...
* test amd gpu crashes
* cleanup
* less sketch tests
2026-02-02 18:57:47 +03:00
George Hotz
6e958dbfd4
assembly/amd: add RDNA4 support to emulator ( #14341 )
...
* start new rdna4
* work
* plus works
* more pass
* rdna4
* assembly/amd: fix RDNA4 emulator for float16 and VOP3 clamp
* stale
* rev
* rr
* rdna4 emu tests
* cleanup
* cleanup
* simp
* works
* better factorizaion
* hacks
* fix mockgpu
* guard both
* cleaner
* gate
* bug fix and a few tests
* all test_tiny
2026-02-02 21:35:59 +08:00
qazal
1746d1f997
remove SPEC=0 context in custom_kernel tests, pyrender always skips it ( #14489 )
2026-02-02 16:32:01 +09:00
George Hotz
d4007f36e0
remove DEFINE_GLOBAL (it is PARAM now) ( #14488 )
2026-02-02 14:56:37 +08:00
Christopher Milan
2931b52875
skip autogen if MTLCompiler is loaded ( #14466 )
2026-02-01 22:12:27 -05:00
chenyu
ea1f1d2b9d
test_assign_to_bitcast_view ( #14483 )
...
currently disk allows assign same size dtype into a bitcasted view
2026-02-01 16:46:04 -05:00
chenyu
6deeccc192
fix RING with single dest ( #14482 )
2026-02-01 12:14:46 -05:00
chenyu
3ff390159b
don't implicitly change dtype in assign ( #14481 )
...
broadcast shape is fine, but implicitly cast dtype is hard to find
2026-02-01 11:48:54 -05:00
imaolo
2111762a48
failed test case for RING output device ( #14191 )
...
* Add enable/disable scheduler cache ContextVar
* add allreduce ring and naive to() tests
* clearer test comparing native vs ring allreduce
* split tests, add helper
* removing trailing whitespace
---------
Co-authored-by: chenyu <chenyu@fastmail.com >
2026-02-01 11:48:43 -05:00
chenyu
02afae04f4
atol in test_call_gemm ( #14480 )
...
flaky
2026-02-01 11:24:58 -05:00
chenyu
5705398a1f
assign cleanup [pr] ( #14479 )
...
share more code path between disk and non-disk. also raise RuntimeError instead of Assert for mismatches
2026-02-01 09:10:22 -05:00
