Philippe Tillet
|
f257884eb7
|
some cleaning
|
2019-06-24 09:31:34 -07:00 |
|
Philippe Tillet
|
67989e7d18
|
fixup
|
2019-06-13 20:03:28 -07:00 |
|
Philippe Tillet
|
f7dcea1187
|
Now doing double-buffering
|
2019-06-13 19:48:02 -07:00 |
|
Philippe Tillet
|
36e3667a9a
|
removed shared conflicts for 8x32x4 and 32x8x4 configurations
|
2019-06-13 17:51:54 -07:00 |
|
Philippe Tillet
|
21a9b92c87
|
disabling interleaving
|
2019-06-13 17:16:00 -07:00 |
|
Philippe Tillet
|
d487cf31ce
|
trying 128 bits loads
|
2019-06-12 21:07:01 -07:00 |
|
Philippe Tillet
|
1c6372711b
|
added interleaving
|
2019-06-12 20:30:28 -07:00 |
|
Philippe Tillet
|
a6b580ec05
|
interleaving fails with B
|
2019-06-12 19:46:43 -07:00 |
|
Philippe Tillet
|
1b5a742a88
|
[triton/codegen] added shared memory padding for HMMA arguments and vectorized loads
|
2019-06-11 19:51:08 -07:00 |
|
Philippe Tillet
|
cbd916994d
|
[example/tensorflow] no longer hardcoding library dir
|
2019-06-11 11:06:02 -07:00 |
|
Philippe Tillet
|
7d50b87681
|
[selection/codegen] bugfix in distributed tile indices initialization
|
2019-06-11 10:45:19 -07:00 |
|
Philippe Tillet
|
06b5992509
|
[feature] added basic tensor core support
|
2019-06-11 10:24:49 -07:00 |
|
Philippe Tillet
|
d074a166e2
|
[feature] basic tensor core utilization works
|
2019-06-08 14:39:45 -07:00 |
|
Philippe Tillet
|
5f3d48c1d0
|
[tensor cores] added basic codegen template for using wmma
|
2019-06-07 21:19:47 -07:00 |
|
Philippe Tillet
|
ec4c6aaaaa
|
Added inline PTX for mma.sync
|
2019-06-07 19:39:33 -07:00 |
|
Philippe Tillet
|
6fce9f28ae
|
added fragmented axis
|
2019-06-07 10:32:56 -07:00 |
|
Philippe Tillet
|
781b6d377d
|
seleciton now segfault (expected
|
2019-06-06 20:34:56 -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
|
f58c9a4d2b
|
[general] hmma baseline setup
|
2019-06-05 14:43:38 -07:00 |
|
Philippe Tillet
|
49fcfd6fc7
|
[examples/tensorflow] fixed #include issue
|
2019-06-05 11:09:41 -07:00 |
|
Philippe Tillet
|
383b5b2a2a
|
[triton/ast] renamed ast -> lang in namespace and file structure
|
2019-05-28 17:28:02 -04:00 |
|
Philippe Tillet
|
d2a46afe00
|
[triton/ast]: cleaned the ast module
|
2019-05-28 17:07:54 -04:00 |
|
Philippe Tillet
|
8102efc064
|
[triton/examples/cpp] removed common.hpp helper
|
2019-05-28 14:14:33 -04:00 |
|
Philippe Tillet
|
a9d078c06f
|
[triton/dnn/conv] merged optimizations branch
- Added forward/backward support for strided convolution
- Added support for bias
- Added support for reduction splitting
|
2019-05-28 14:04:53 -04:00 |
|
Philippe Tillet
|
e526ffc62b
|
[examples/pytorch] added a bunch of models for more thorough testing
|
2019-05-28 14:04:31 -04:00 |
|
Philippe Tillet
|
3f3eb1c2a4
|
[dnn/conv] Added the option to have look-up table for filters for all
operations
|
2019-05-22 19:03:33 -04:00 |
|
Philippe Tillet
|
f8291af7ef
|
[dnn/conv] removed divergent paths in LUT computations
|
2019-05-22 17:49:40 -04:00 |
|
Philippe Tillet
|
2672812ad0
|
[dnn/conv] No more divergent path in conv::set_arg
|
2019-05-22 15:25:43 -04:00 |
|
Philippe Tillet
|
e8f23bcade
|
[dnn/conv] Added bias and forward stride
|
2019-05-22 13:27:08 -04:00 |
|
Philippe Tillet
|
f33a1f3fe3
|
[examples/pytorch] Fixed issues in backward pass of conv
|
2019-05-19 01:31:08 -04:00 |
|
Philippe Tillet
|
b2b55c52c9
|
[triton/python/conv]: Added cache for compiled kernels
|
2019-05-18 11:51:49 -04:00 |
|
Philippe Tillet
|
600aef72d5
|
[conv/dnn] now created a separate .h and .cpp file
|
2019-05-17 12:29:11 -04:00 |
|
Philippe Tillet
|
34f8617709
|
[dnn/conv] fixed formatting of generated Triton-C code
|
2019-05-16 15:48:02 -04:00 |
|
Philippe Tillet
|
ece7beea3c
|
[dnn/conv]: now using look-up table for wgrad computation as well
|
2019-05-16 15:26:16 -04:00 |
|
Philippe Tillet
|
15a967c81e
|
[dnn/conv] minor cleaning
|
2019-05-15 11:32:47 -04:00 |
|
Philippe Tillet
|
be2ba03382
|
[dnn/conv] optimizations of backpropagation with look-up tables
|
2019-05-14 19:10:59 -04:00 |
|
Philippe Tillet
|
cbfbe72e46
|
[general] added LICENSE file
|
2019-05-13 22:29:53 -04:00 |
|
Philippe Tillet
|
5941501f70
|
[dnn] added Triton-C derivative computations in conv
|
2019-05-13 18:04:11 -04:00 |
|
Philippe Tillet
|
f6fe9492e4
|
[dnn/conv] added triton-c code for wgrad
|
2019-05-11 18:09:23 -04:00 |
|
Philippe Tillet
|
fc4daf11dd
|
[examples/conv] now deferring shape computations to conv configuration
|
2019-05-08 13:58:25 -04:00 |
|
Philippe Tillet
|
54f888a270
|
[dnn/conv] some minor fixes
|
2019-05-08 10:09:30 -04:00 |
|
Philippe Tillet
|
615569287e
|
more cleaning of conv
|
2019-05-06 19:30:22 -04:00 |
|
Philippe Tillet
|
fd91368f98
|
[general] creation of dnn module for gemm/conv triton routines
|
2019-05-06 17:47:06 -04:00 |
|
Philippe Tillet
|
f80441017c
|
[codegen] added leading dimension padding for transposition in shared
memory
|
2019-05-06 11:53:35 -04:00 |
|
Philippe Tillet
|
4813bb007c
|
[codegen] bugfix in builder insert point for predicated instructions
|
2019-05-04 12:09:27 -04:00 |
|
Philippe Tillet
|
30833c18f1
|
[codegen/tune] bugfix in heuristics for nano-tile sizes
|
2019-05-04 01:32:34 -04:00 |
|
Philippe Tillet
|
0d694445e6
|
[examples] added skeleton for pytorch wrapper
|
2019-05-03 14:30:06 -04:00 |
|