commaai
/
tinygrad
mirror of https://github.com/commaai/tinygrad.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
tinygrad/test
History
CaltropHungerton 38fb1e14a2
Intel XMX Tensor Core Support (#5622) 
* fixed xmx demo

* i think i'm invoking the DPAS but it's slow

* compiler build arg to stop register spilling, indicated where to fix flop counter

* don't mind this

* do NOT mind me

* do not mind me

* do not view

* i will add bf16 later

* in process of figuring out tc fields

* we figured out the fields!!!

* added check for cl device vendor, added seperate IntelRenderer

* remove tc thread_local_aliases

* cleaning debris before draft pr

* edits for linter

* deduping and checking device extensions

* i will find more line reductions in other places

* before merge upstream

* double grf size in compiler to fix register spilling (bandaid), device checking changes

* tc python emulation

* fixed emulation

* tests for emulated intel tensor core

* TC=0, 1 working on upstream, fixed perf

* test

* debris

* check for specialized cl device when we canonicalize device

* bf16 support, tc=3 test added

* address tests

* revert half2 loads on intel tc, cleanup

* linter

* fold_expanded revert

* lint, whitespace fix

* cuda bf16 (only one with bf16) is skipped in test tensor cores, so i will skip for intel bf16 too

* make line shorter, no need for noqa E501

* removed device intel

* fix python emulation

---------

Co-authored-by: George Hotz <72895+geohot@users.noreply.github.com>
 		2024-08-16 09:19:21 -07:00
..
external fix qcom process_replay for kernel diff (#6079) 2024-08-14 15:05:49 -04:00
imported update pylint path to check indent/space for all (#6022) 2024-08-10 14:41:09 -04:00
models add failing regression test for image (#5540) 2024-07-17 17:27:18 -07:00
testextra names shadowing builtins (#5179) 2024-06-27 08:15:01 -04:00
unit UOp pattern DEFINE_VAR with min==max is also CONST (#6095) 2024-08-15 12:09:44 -04:00
Dockerfile
…
__init__.py
…
helpers.py minor test fixups from the AST is UOp diff (#6081) 2024-08-14 23:58:04 +03:00
test_arange.py fix tests locally on metal (#6025) 2024-08-10 12:36:22 -07:00
test_assign.py test masked assign views (#4599) 2024-05-15 15:06:48 +03:00
test_compile_failures.py new test for e2e compile failures (#6101) 2024-08-15 18:56:22 -07:00
test_const_folding.py MetaOps.KERNEL (#5543) 2024-07-17 19:41:23 -07:00
test_conv.py db in wal mode (#5388) 2024-07-12 20:43:36 -07:00
test_conv_shapetracker.py rename lazyops to parents [run_process_replay] (#6091) 2024-08-15 17:27:32 +03:00
test_copy_speed.py
…
test_custom_function.py s/loadops/metaops [run_process_replay] (#5421) 2024-07-12 13:26:50 -07:00
test_device_speed.py move uopgraph to file [run_process_replay] (#5364) 2024-07-10 17:34:50 -07:00
test_dtype.py Shape change bitcast for more dtypes (#6047) 2024-08-14 10:03:34 -07:00
test_dtype_alu.py rename lazyops to parents [run_process_replay] (#6091) 2024-08-15 17:27:32 +03:00
test_fusion_op.py increase test_recursive_add time from 1s to 2s (#6078) 2024-08-14 13:52:02 -04:00
test_fuzz_shape_ops.py fix typing for test to run in py38 (#4930) 2024-06-12 13:22:30 -04:00
test_gc.py threefry again (#3785) 2024-03-18 16:47:07 -04:00
test_graph.py fix hcq sync (#5062) 2024-06-26 17:50:37 +03:00
test_hcq.py fix non-jitted transfers in profile (#5980) 2024-08-08 17:58:08 +03:00
test_image_dtype.py add failing regression test for image (#5540) 2024-07-17 17:27:18 -07:00
test_jit.py remove realize from threefry (#5969) 2024-08-07 15:08:49 -07:00
test_kernel_cache.py
…
test_lazybuffer.py use assertIs in test_schedule (#6035) 2024-08-11 19:19:18 +03:00
test_lazyop.py scheduleitem is not Tuple [run_process_replay] (#5425) 2024-07-12 15:13:19 -07:00
test_linearizer.py Intel XMX Tensor Core Support (#5622) 2024-08-16 09:19:21 -07:00
test_linearizer_dumb.py spec for in order LOAD/STORE indexing (#6073) 2024-08-14 19:18:00 +03:00
test_linearizer_failures.py rewrite bool ADD to OR and MUL to AND (#6084) 2024-08-15 10:11:57 -04:00
test_linearizer_overflows.py lowerer is kernel [run_process_replay] (#5437) 2024-07-12 18:50:55 -07:00
test_masked_st.py
…
test_method_cache.py simple LoadOps.ASSIGN (#3745) 2024-03-14 20:44:34 -07:00
test_multitensor.py added fix and reg tests (#6060) 2024-08-12 21:00:48 -04:00
test_net_speed.py nv mockgpu (#4600) 2024-05-15 23:46:08 +03:00
test_nn.py add LSTMCell to nn (#6080) 2024-08-14 12:08:42 -07:00
test_ocl.py touchup cl_errors (#6058) 2024-08-13 13:06:59 -04:00
test_ops.py bilinear interp uint8 fails (#6103) 2024-08-15 19:34:39 -07:00
test_optim.py improve test_dropout_on_shard (#4912) 2024-06-11 11:36:02 -04:00
test_pattern_matcher.py revert UOps eq, this needs to be isolated in realize.py (#6063) 2024-08-13 18:02:34 +03:00
test_pickle.py some TestPickleJIT tests (#5860) 2024-08-01 12:39:59 -07:00
test_profiler.py hcq profiler support args (#5989) 2024-08-09 00:18:36 +03:00
test_randomness.py jit sampling functionn in test_randomness.test_multinomial (#5034) 2024-06-18 14:21:05 -04:00
test_renderer_failures.py Fix compile error for max with inline const (#5840) 2024-08-12 23:40:39 +08:00
test_sample.py
…
test_schedule.py rename lazyops to parents [run_process_replay] (#6091) 2024-08-15 17:27:32 +03:00
test_search.py rename lazyops to parents [run_process_replay] (#6091) 2024-08-15 17:27:32 +03:00
test_setitem.py setitem in-place operator tests (#4577) 2024-05-14 01:28:02 -04:00
test_specific_conv.py nv mockgpu (#4600) 2024-05-15 23:46:08 +03:00
test_speed_v_torch.py remove CUDACPU flag in tests [run_process_replay] (#5902) 2024-08-04 16:06:38 -04:00
test_subbuffer.py remove CUDACPU flag in tests [run_process_replay] (#5902) 2024-08-04 16:06:38 -04:00
test_symbolic_jit.py sort vars in jit when building expected input args (#4990) 2024-06-16 15:55:51 -04:00
test_symbolic_ops.py symbolic Tensor.var (#4843) 2024-06-05 12:55:54 -04:00
test_symbolic_shapetracker.py support symbolic reshape with non-contiguous (#4844) 2024-06-05 16:01:19 -04:00
test_tensor.py Ignore non-computational backends when overwriting the default (#5770) 2024-08-10 09:23:29 -07:00
test_tensor_data.py BEAM_COMPARE=2 validates the correctness of BEAM kernels (#5458) 2024-07-13 13:53:43 -07:00
test_tensor_variable.py Should this symbolic test fail? (#4501) 2024-06-18 15:21:26 -04:00
test_to_numpy.py
…
test_transcendental.py lower test_transcendental fuzz test threshold for sin float64 (#5956) 2024-08-07 02:04:37 -04:00
test_uop_graph.py UOp pattern DEFINE_VAR with min==max is also CONST (#6095) 2024-08-15 12:09:44 -04:00
test_uops.py spec for in order LOAD/STORE indexing (#6073) 2024-08-14 19:18:00 +03:00
test_uops_stats.py UOpGraph not in renderer or Program [run_process_replay] (#5867) 2024-08-01 16:20:30 -07:00
test_verify_lazyop.py pretty print lazy op per default (#5505) 2024-07-18 09:34:08 -07:00
test_winograd.py rename lazyops to parents [run_process_replay] (#6091) 2024-08-15 17:27:32 +03:00
test_zero_copy.py
…