Commit Graph

76 Commits

Author SHA1 Message Date
Philippe Tillet
26c9849462 [ir][instructions] added permutations option for trans 2019-08-05 21:19:13 -07:00
Philippe Tillet
d62e581ab3 basic split-k across warps working for GEMM 2019-08-05 19:33:28 -07:00
Philippe Tillet
d869d9a924 [codegen][selection] more flexible instruction selection for reduce_inst 2019-08-04 16:34:36 -07:00
Philippe Tillet
6be532c6a2 [codegen][selection] adding support for reduction along arbitrary axis 2019-08-02 21:29:36 -07:00
Philippe Tillet
d9945692a9 [dnn] better specification of recompilation key 2019-08-02 17:42:48 -07:00
Philippe Tillet
3b92ddf7e6 [codegen/reassociation] now recursively takes pointer arguments into account as well 2019-07-31 18:41:56 -07:00
Philippe Tillet
f7bd976fc7 [dnn/blocksparse] added heuristics for block-sparse dot 2019-07-31 17:12:36 -07:00
Philippe Tillet
bb32ac56c9 [codegen/optimize_dce.cpp] fixed bugs whereby barriers were removed by DCE 2019-07-31 15:11:10 -07:00
Philippe Tillet
080bf1af88 [dnn/blocksparse/dot]: BlocksparseDx also working 2019-07-30 11:42:31 -07:00
Philippe Tillet
17cb2db356 [dnn/blocksparse/dot] prototype version seems to pass basic test 2019-07-27 21:21:36 -07:00
Philippe Tillet
2a377bc8b1 [ir] deleted mask/merge instructions; will be replaced by masked_load/store and select 2019-07-25 15:06:15 -07:00
Philippe Tillet
397d76156b progress on re-association 2019-07-23 17:21:24 -07:00
Philippe Tillet
c448876178 better benchmarking 2019-07-22 19:26:12 -07:00
Philippe Tillet
b1d81a5802 more work on heuristics 2019-07-21 18:11:54 -07:00
Philippe Tillet
d159455f7b [codegen/alignment_info] better alignment information 2019-07-20 21:44:18 -07:00
Philippe Tillet
28c250216c [dnn/gemm] added some bounds checking 2019-07-19 21:32:55 -07:00
Philippe Tillet
5215fb0424 [codegen] some more optimizations 2019-07-19 20:29:03 -07:00
Philippe Tillet
f0d8306437 [codegen/alignment_info] better handling of constants 2019-07-18 16:12:06 -07:00
Philippe Tillet
86f70f8224 [codegen/selection] performance fix-up when A is transposed for hmma 2019-07-17 21:46:23 -07:00
Philippe Tillet
2f0817b2cd [codegen/selection] tensor cores now used for transposed layotus 2019-07-17 17:20:38 -07:00
Philippe Tillet
bfa39b8992 preparing the field for tensor cores transposes 2019-07-17 13:20:33 -07:00
Philippe Tillet
791c91ee63 [dnn/shift] bugfix in static shape division 2019-07-17 11:39:17 -07:00
Philippe Tillet
a55b098e88 [dnn/shift] now using constant divisions 2019-07-16 21:05:21 -07:00
Philippe Tillet
07c964919c [dnn/shift] now strictly only shifting the interior 2019-07-16 20:18:48 -07:00
Philippe Tillet
ec24e1e7df trying to remove interior logic 2019-07-16 18:47:50 -07:00
Philippe Tillet
5f6dd23fc2 [dnn/dot] reverted back to peak tensorcores performance 2019-07-16 16:14:58 -07:00
Philippe Tillet
28959fe165 [runtime/jit] made auto-tuning silent 2019-07-16 14:41:38 -07:00
Philippe Tillet
f9db0449b7 [dnn] Adding batchnorm 2019-07-08 18:44:37 -07:00
Philippe Tillet
8fc253946c [codegen] shift: added sketch for shift-convolution backpropagation 2019-07-02 16:39:07 -07:00
Philippe Tillet
6cfb575d29 [lang] fixup in cast type 2019-06-30 17:43:18 -07:00
Philippe Tillet
c172bd518b more stuff 2019-06-30 16:55:02 -07:00
Philippe Tillet
d8c3d58593 more optimization 2019-06-28 20:22:52 -07:00
Philippe Tillet
f4dedb522c fixup 2019-06-27 17:05:48 -07:00
Philippe Tillet
6300ec5080 [examples] added conv2d op in tensorflow 2019-06-26 18:50:53 -07:00
Philippe Tillet
25e9a10917 changed auto-tuner parameter ranges 2019-06-25 19:27:49 -07:00
Philippe Tillet
d945ce5e1b Now showing valid parameter for NN 2019-06-25 19:18:43 -07:00
Philippe Tillet
64513fb407 [codegen] added fallback when tensor cores cannot be used 2019-06-25 15:49:58 -07:00
Philippe Tillet
06b5992509 [feature] added basic tensor core support 2019-06-11 10:24:49 -07:00
Philippe Tillet
6045209d5b Now find correct tuning configuration 2019-06-06 20:13:26 -07:00
Philippe Tillet
0a0b48e9a2 adding hmma tuning parameters 2019-06-06 19:51:02 -07:00
Philippe Tillet
81eba3e1ec ugh 2019-06-06 19:36:41 -07:00
Philippe Tillet
cdf5a0d011 [codegen/tune]: added fragmentation types 2019-06-06 16:48:32 -07:00
Philippe Tillet
30833c18f1 [codegen/tune] bugfix in heuristics for nano-tile sizes 2019-05-04 01:32:34 -04:00
Philippe Tillet
3413aad582 [general] major overhaul of triton-c/triton-ir/triton-jit:
- Added alloc const
- Added atomics
- Pruning tuning space
- Added example for dot/conv/shift
- Bugfixes
2019-04-25 16:18:15 -04:00
Philippe Tillet
0c607c9392 [examples] normalize benchmark by max_clock / current_clock 2019-03-28 07:58:37 -04:00
Philippe Tillet
2c3ae0675e [JIT] re-added nvidia compatibility 2019-03-27 21:12:01 -04:00
Philippe Tillet
fdf8559806 [general] added missing files 2019-03-27 20:01:35 -04:00
Philippe Tillet
bc2a257d5c [code generation] more flexibility in backend selection 2019-03-27 11:29:42 -07:00
Philippe Tillet
e04253c0dd [code generation] basic CPU backend 2019-03-27 11:13:36 -07:00
Philippe Tillet
8d35c98920 [code generation] search space pruning 2019-03-25 14:10:24 -07:00