qazal
6febd20fcf
set forced_realize for outputs [pr] ( #7469 )
2024-11-01 20:03:12 +02:00
Tobias Fischer
7c9a1d69f9
sdxl gen fix ( #7459 )
2024-11-01 13:57:01 -04:00
ignaciosica
9c832483f2
update shifts spec ( #7468 )
...
* update shifts spec
* hotfix: old style
2024-11-01 12:40:41 -04:00
ignaciosica
18bd98c203
Add shl and shr to llvmir ( #7449 )
...
* add shl and shr to llvmir
* hotfix: enforce type alignment for shr and shl in all backends
* hotfix: change shl and shr spec
* hotfix: typo
* hotfix: refactor shl and shr rules and add casting to ptx shl
* hotfix: bug
* hotfix: ptx shl and shr require buint32
* hotfix: cleanups
2024-11-01 23:49:34 +08:00
chenyu
18e159c9ac
comment about multi real and more tests [pr] ( #7467 )
2024-11-01 11:49:11 -04:00
chenyu
1f343aa40e
replace x.alu(BinaryOps.ADD, y) with add in multi [pr] ( #7466 )
2024-11-01 10:50:57 -04:00
geohotstan
6513690223
Add Tensor.hardsigmoid ( #7433 )
...
* move hardsigmoid to new branch
* add to test
* add NOTE to mention differing values for alpha and beta that match torch
* shift from relu6
* correct shift implementation
* or we just use relu? no more 666
2024-11-01 08:36:52 -04:00
George Hotz
fe78ed8cb7
improve match speed [pr] ( #7465 )
...
* improve match speed [pr]
* no sym in expand
* remove useless rule, sym back
* don't track that
2024-11-01 17:33:53 +08:00
George Hotz
a7ba3d2d91
move reduce to lowerer [pr] ( #7462 )
...
* move reduce to lowerer [pr]
* simpler
2024-11-01 16:39:20 +08:00
George Hotz
2cfca230b5
reduce collapse as a rule ( #7464 )
...
* reduce collapse as a rule
* better [pr]
* cleaner
2024-11-01 16:25:44 +08:00
George Hotz
4f6cf1f8cc
expand DEFINE_ACC [pr] ( #7461 )
2024-11-01 15:20:43 +08:00
qazal
d9f38f9518
group stores by UOp [pr] ( #7460 )
2024-11-01 15:09:16 +08:00
qazal
c1bd2d3f71
viz increment -1 kernel on enter [pr] ( #7448 )
...
* viz increment -1 kernel on enter [pr]
* two paths
* share
2024-11-01 14:14:54 +08:00
Tobias Fischer
1a9e145388
Tensor Clone Function ( #7154 )
...
* implemented clone function
* cleanup linting, single func
* added tests, cleaned up grad cloning
* fixed whitespace
2024-11-01 12:24:43 +08:00
chenyu
acd0fa1a7a
s/hasattr(self, '_buf')/self.is_allocated() [pr] ( #7458 )
...
use is_allocated helper in Buffer
2024-10-31 20:55:20 -04:00
chenyu
036409266d
clean up _prepare_jit_inputs [pr] ( #7457 )
...
removed an unnecessary cast and reordered a bit
2024-10-31 20:41:02 -04:00
chenyu
a21434504b
update payne_hanek_reduction [pr] ( #7455 )
2024-10-31 18:41:22 -04:00
chenyu
4f27862242
no more UPat._any [pr] ( #7454 )
2024-10-31 16:48:37 -04:00
chenyu
5777fca904
clean up cody_waite_reduction magic numbers ( #7452 )
2024-10-31 14:45:04 -04:00
chenyu
5648b9788e
more xlog2 cleanups ( #7451 )
...
following the notations in the paper closer
2024-10-31 13:52:31 -04:00
chenyu
4065c3dec8
remove special 0 case in frexp ( #7450 )
...
we can safely assume input is non-zero, also removed unneeded bitcast
2024-10-31 13:02:33 -04:00
chenyu
53db3478fe
cast to float32 for float16 xlog2 ( #7447 )
...
formula has 2X error with denormal floats
2024-10-31 10:36:29 -04:00
chenyu
5085b2fde7
cleanup xlog2 and remove unneeded functions ( #7446 )
...
denormal_map still looks wrong but a lot cleaner
2024-10-31 09:45:16 -04:00
chenyu
02636bc05e
simpler switch over in xsin ( #7426 )
2024-10-31 08:56:01 -04:00
qazal
c5a50465d1
big graph first [pr] ( #7443 )
...
* big graph first [pr]
* move things
2024-10-31 20:10:11 +08:00
qazal
38b1790575
move image dtype fixup [pr] ( #7444 )
...
* move image dtype fixup [pr]
* more work
* late dtype
* use base
2024-10-31 19:51:46 +08:00
George Hotz
f579693ec9
hotfix: casted nan/inf
2024-10-31 19:50:17 +08:00
George Hotz
a43b7a4b7c
less rewrite stages in matcher ( #7445 )
...
* less rewrite stages in matcher
* better name
2024-10-31 19:45:21 +08:00
George Hotz
5dd1ffd5d0
don't const rewrite in cstyle ( #7442 )
...
* don't const rewrite in cstyle
* Update cstyle.py
* simple_symbolic
* fix bfloat16 const on AMD
2024-10-31 19:16:49 +08:00
qazal
bdde795239
early filter sink buffers [pr] ( #7440 )
2024-10-31 18:50:36 +08:00
qazal
9905de3362
late append realizes [pr] ( #7439 )
...
* dont unbind in ops
* late append realizes [pr]
* Revert "dont unbind in ops"
This reverts commit e8d9da936d
2024-10-31 18:04:42 +08:00
George Hotz
50ddd11350
lil cleanup matchers [pr] ( #7437 )
...
* move delete_redundant_gates [pr]
* simpler uops test
* addr in delete_redundant_gates
* lines
* correct early delete gates
* shorter find_gate
2024-10-31 17:52:22 +08:00
qazal
a0bd385448
late uop_bufs [pr] ( #7438 )
2024-10-31 17:30:32 +08:00
qazal
7916d1f6ab
shorter UOps.BUFFER init [pr] ( #7436 )
2024-10-31 17:14:19 +08:00
George Hotz
2e3048fc57
Revert "improve full_graph_rewrite matchers for speed ( #7431 )" ( #7434 )
...
This reverts commit 996152d2de
2024-10-31 16:16:47 +08:00
George Hotz
996152d2de
improve full_graph_rewrite matchers for speed ( #7431 )
...
* remove finalize [pr]
* early transcendental
* fix tests
* load store indexing runs with devectorize
* move delete_redundant_gates
* ptx has to wait for the mask to move
2024-10-31 16:13:11 +08:00
qazal
5f49651360
verify assign pre astfying [pr] ( #7417 )
2024-10-31 16:02:07 +08:00
George Hotz
17c9a9fde4
pm_render [pr] ( #7430 )
...
* pm_render [pr]
* test fixes
* use gep, not src
* ptx only symbolic, not sym
* move cast rules
2024-10-31 15:04:50 +08:00
George Hotz
8fff8fc3e7
replace REDUCE and clean up arange ( #7429 )
...
* break apart arange [pr]
* fix missing
* cleanups to add/mul
* UOps.VECTORIZE
* don't vectorize const
2024-10-31 14:02:20 +08:00
George Hotz
fe2bc4c613
clean up arange/indexing matchers [pr] ( #7427 )
...
* clean up arange/indexing matchers [pr]
* syntax for assign
2024-10-31 12:12:44 +08:00
George Hotz
e446e95974
enforce ctx is called ctx [pr] ( #7424 )
...
* enforce ctx is called ctx [pr]
* fix bug and use has_ctx
* inspect signature
* assert
* no slow asserts
* now we can support contextual reduce
2024-10-31 11:39:19 +08:00
chenyu
9b08bb4c3e
fold the +x term in sine inside sin_poly ( #7425 )
2024-10-30 23:13:08 -04:00
chenyu
0739895b4d
tiny clena up pow2if and payne_hanek_reduction ( #7423 )
2024-10-30 22:22:48 -04:00
chenyu
118dd7721f
clean up transcendental.rintk [pr] ( #7422 )
...
added unit tests and updated the comment. it's rounding away from 0 for negatives
2024-10-30 20:37:28 -04:00
chenyu
fb694a63eb
Tensor.erf ( #7419 )
...
the same one used in onnx and the one in bert.
2024-10-30 18:12:28 -04:00
qazal
e955aa1bee
hotfix: process replay ( #7418 )
2024-10-30 22:45:40 +02:00
qazal
4c0ee32ef2
delete metadata from schedule ctx [pr] ( #7415 )
2024-10-31 01:49:49 +08:00
George Hotz
b4410545d8
hotfix: INDEX is yellow-green
2024-10-31 01:42:54 +08:00
qazal
d81e07e4fc
compare schedule len against group count [pr] ( #7414 )
2024-10-31 01:42:10 +08:00
qazal
1a2ee37dd3
hotfix: remove redundant test_schedules [pr] ( #7412 )
2024-10-31 01:10:31 +08:00
