ba7314c26b
cleanup lbs ( #4163 )
2024-04-12 22:32:16 -07:00
a7c6864260
remove CAST_BEFORE_VIEW ( #4152 )
...
* remove CAST_BEFORE_VIEW
testing perf, also this might have issue with assign?
* remove all
2024-04-13 01:05:08 -04:00
ebc94c9d6c
rewrite the jit in the context of new schedule ( #4162 )
...
* rewrite the jit in the context of new schedule
* mypy better
* fix placeholder
* tests
* all functionality should work
* fix tests
* no CacheCollector
2024-04-12 21:54:36 -07:00
b67f759780
abstractions3 is currently wishful thinking ( #4124 )
...
* abstractions3 is currently wishful thinking
* a3
* work
* minor
* progress on a3
* more
* update abstractions3
* cleaner
2024-04-12 16:46:01 -07:00
27a98aaecc
Rewritten SVG Logos ( #4150 )
...
* rewrote the svg logos to use polygons and render better
* changed self-closing tags' style to better conform to the original
2024-04-12 14:09:57 -07:00
63eb0a68af
fix return dtype of gather ( #4159 )
2024-04-12 16:25:12 -04:00
d9c5a2b1bb
fix return dtype of getitem Tensor indexing ( #4158 )
...
the use of sum can auto-upcast the result. fixed by using the data dtype as the acc_dtype
2024-04-12 15:55:02 -04:00
f6c8032e5d
assert if expr_idxs return might be outside of int32 ( #4157 )
2024-04-12 14:18:35 -04:00
24a27a01a9
hotfix: CUDA_P2P works ( #4155 )
2024-04-12 18:20:12 +03:00
5a57b48134
cuda p2p enable when available ( #4153 )
2024-04-12 16:21:54 +03:00
380f27d629
move sum acc_dtype into lazy so it applies to backward ( #4149 )
...
* move sum acc_dtype into lazy so it applies to backward
* unit test
2024-04-11 14:43:56 -04:00
bbda20c0db
CompiledASTRunner -> CompiledRunner ( #4148 )
2024-04-11 08:49:52 -07:00
0f16709c00
hotfix: remove test speed vs torch
2024-04-11 08:37:57 -07:00
c0796374e4
refactor membufs ( #4147 )
2024-04-11 08:30:44 -07:00
b7e281cf10
JitItem -> ExecItem ( #4146 )
...
* JitItem -> ExecItem
* execitem in realize
* cleaner
* JITRunner -> Runner
2024-04-11 08:24:57 -07:00
e79a11b99c
hotfix: revert llama change
2024-04-10 20:13:15 -07:00
2e6c39b0b2
Do less realizes ( #4141 )
...
* less realize
* corealize jit inputs
* prints
* print before we run
2024-04-10 19:50:50 -07:00
06bcae13b4
PADTO SUM if parents of sum are all zero-preserving ( #4140 )
...
* PADTO SUM if parents of sum are all zero-preserving
* test case unsafe ops after sum is fine
* reuse UNSAFE_PAD_OPS
* update db version
2024-04-10 22:16:12 -04:00
081dd1573f
hotfix: keep CUDA D2D copy behind the CUDA_P2P flag
2024-04-10 21:36:48 +00:00
af5984df43
cudagraph memcpy through host ( #4137 )
2024-04-10 13:17:17 -07:00
5e6d2155e4
Add driving monitoring model to benchmarks ( #4134 )
...
* add driving monitoring model to benchmarks
* handle crash
2024-04-10 14:27:03 -04:00
bf3583f9b2
use Buffer.ensure_allocated in search _ensure_buffer_alloc ( #4132 )
2024-04-10 13:11:50 -04:00
a35375df85
run_schedule is so simple now ( #4130 )
2024-04-10 09:49:30 -07:00
86bd2eb500
hotfix: update copy_from_fd for new DiskBuffer
2024-04-10 15:41:06 +00:00
ee457a4b20
no more underlying diskbuffer, that's just the device ( #4129 )
2024-04-10 08:32:25 -07:00
fe88591890
update onnx to 1.16.0 ( #4127 )
...
* update
* pass tests and skip tests
2024-04-10 11:19:13 -04:00
6bbbeb93ac
skip a few clang test that took > 30 seconds in CI ( #4126 )
...
* skip slow CLANG test test_train_cifar
* skip those too
* and that
* only CI
* one more
2024-04-10 02:00:34 -04:00
08ddeb5685
create schedule has global vars ( #4125 )
...
* abstractions3 is currently wishful thinking
* create_schedule_with_vars
2024-04-09 21:42:16 -07:00
216eb235e5
hotfix: cast mnist to float
2024-04-09 19:30:03 -07:00
fea774f669
spend 5 lines to bring mnist into the repo ( #4122 )
2024-04-09 19:24:57 -07:00
42edae8935
pickle schedules ( #4114 )
...
* pickle schedules
* Update test_pickle.py
* Update test_pickle.py
---------
Co-authored-by: George Hotz <72895+geohot@users.noreply.github.com>
2024-04-09 13:47:25 -07:00
38ae4194a6
Fixes for ops_kfd ( #4105 )
...
* kfd_ops: Fix GPU node discovery on NUMA systems
Ignore potentially multiple CPU NUMA nodes and any GPU nodes that are
not accessible because of device cgroups.
Signed-off-by: Felix Kuehling <felix.kuehling@amd.com>
* kfd_ops: Format the GFX arch target name correctly
The target version in sysfs properties is a decimal representation with
two digits per component.
The format for LLVM GFX target names is a bit quirky for historical
reasons. It uses one digit for the minor version and stepping. When it
ran out of decimal digits for the stepping on gfx90X it started using
hexadecimal there. But the major version is still decimal and went
double digit in GFX10.
Make sure to parse and format it accordingly for all supported GPUs.
Signed-off-by: Felix Kuehling <felix.kuehling@amd.com>
---------
Signed-off-by: Felix Kuehling <felix.kuehling@amd.com>
2024-04-09 13:21:21 -07:00
10dbf90b2c
hotfix: test speed
2024-04-09 13:20:39 -07:00
ae849d12d7
numpy device + pickle it ( #4120 )
2024-04-09 13:19:30 -07:00
1ef9c50fd7
Update ssa input order and annotate types in cstyle and assembly ( #4117 )
...
variable prefix is never optional (removed the default "t") and UOp can be optional (added the default None).
2024-04-09 13:10:29 -04:00
15f2f39658
conceptually simpler fancy index ( #3335 )
...
* init
* add failed case
* fix: temp comment out MULACC cast
* is this right?
* add test case
* oops, forgot to get rid of temp test
* WOOOOOO TOOK OUT 2 TRANSPOSES IN GATHER YAY
* cleaner
* comment cleanup
* update docs
* resolve conflict
* oops
* SUPA FAST
* comment out a test
* del some print statements
* use new broadcast stuff
* more clean up
* move try except
* skip fancy indexing for python backend test_ops
2024-04-09 11:18:04 -04:00
980124a605
add lerp operation to tensor ( #4102 )
...
* feat: add lerp operation to tensor
* fix
* style: fit in one line:
* tests: test backward for lerp
2024-04-08 17:03:27 -07:00
46850a0269
search: add a BEAM_COMPARE env to optionally not compare to hc/tc ( #4107 )
...
* search: add a BEAM_COMPARE env to optionally not compare to hc/tc
setting BEAM_COMPARE=0 will prevent additional memory allocation
needed to do the timing tests assuming the BEAM result is in
the diskcache.
* change to optionally use Buffer.allocate
2024-04-08 18:54:01 -04:00
c390828f61
refactor outbufs ( #4112 )
2024-04-08 14:54:10 -07:00
7fd12aba85
graph remove input buffer references ( #4100 )
...
Co-authored-by: chenyu <chenyu@fastmail.com>
2024-04-08 16:49:16 -04:00
078d841479
add SPLIT_REDUCEOP to disable reduce split ( #4115 )
...
verify with `SPLIT_REDUCEOP=0 BIG=2 MPS=1 python3 -m pytest -rA test/test_speed_v_torch.py -k sum`. 10X slower on mac
2024-04-08 16:31:08 -04:00
eea42d864f
account for all outputs ( #4113 )
2024-04-08 10:04:19 -07:00
dbd39ab78a
setitem support setting python const ( #4111 )
2024-04-08 11:37:50 -04:00
f8dc82a8a7
use single tensor for llama kv chache ( #4108 )
...
similar to optimization in gpt2
2024-04-08 00:38:32 -04:00
92c0675ccf
setitem initial support ( #4093 )
...
* wip setitem
it's an eager assign to output shapetracker view
* cleanups and tests
* more cleanups
2024-04-07 20:35:22 -04:00
183708b3fd
broadcast expand to match torch ( #4085 )
...
* initial version
* heh gimme grrrreen
* version 2
* clean ups
* some test confusion
* fix onnx
* rename to _broadcast_tensors
* improved errors and test
* fixed?
* some test fixup
* version 3 lol
* comments
* cleaner
* add failure test for expand to 0 test
* 1 more assertRaises test
* make err msg better
* also rewrite the expand onnx op? :s
2024-04-07 16:23:13 -04:00
2b81d9b334
Fix broken test ( #4104 )
2024-04-07 12:02:12 -04:00
9a95d87366
metal CI run llama with 4 shards ( #4103 )
...
this can catch multi tensor issue on mac.
2024-04-07 11:04:08 -04:00
444d2a7487
hotfix: fix SDMA read_pointer_address in KFD
2024-04-07 13:13:15 +00:00
bb7567b365
Fix metal ( #4101 )
2024-04-07 05:21:19 -07:00
George Hotz
chenyu
MaximilianEmel
nimlgen
George Hotz
qazal
terafo
geohotstan
Felix Kuehling
David González Martínez
Francis Lam
andresgit
uuuvn