gaspersic/triton
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
461 Commits 14 Branches 9 Tags
acd460402d8f13f6018cd8ee6dc546262978b4d3
Commit Graph

51 Commits

Author SHA1 Message Date
Philippe Tillet
acd460402d Kernels/REDUCE: added temporary workspace information 2015-12-18 18:14:29 -05:00
Philippe Tillet
2d487f894d Kernels: Fixed CUDA compatibility issues 2015-12-17 04:12:57 -05:00
Philippe Tillet
b9700b029a Fixup for cuda 2015-12-17 03:48:44 -05:00
Philippe Tillet
722dae528c Kernels/GEMM: AMD optimizations are no longer beneficial with latest driver => revert 2015-12-17 00:51:04 -05:00
Philippe Tillet
0b6a952d2e Kernels: OpenCL fix for 1D/2D reductions 2015-12-16 16:48:01 -05:00
Philippe Tillet
83feed534c Code quality: more renaming 2015-12-16 16:34:36 -05:00
Philippe Tillet
042aa070bb Code Quality: More sensible names 2015-12-12 21:19:59 -05:00
Philippe Tillet
46dad59e10 Tests: Fixed typos and polished test names 2015-12-12 13:31:14 -05:00
Philippe Tillet
f975ea7621 GEMM: reverted AMD optimizations 2015-12-12 01:29:08 -05:00
Philippe Tillet
004eebc038 Kernels: fixed kernels fusion for DOT, GEMV 2015-12-05 19:14:09 -05:00
Philippe Tillet
386963a6cc Core: added queue-wise temporary workspace. WARNING: breaks the fused computation of multiple DOT/GEMV operations 2015-11-27 18:43:46 -05:00
Philippe Tillet
6fc94c0c0b Kernels: Fixed various corner cases for the kernel templates and BLAS 2015-11-26 19:49:44 -05:00
Philippe Tillet
da1b0a9571 GEMM: performance regression fix 2015-11-19 20:49:38 -05:00
Philippe Tillet
e2cdb88338 Core: included bugfixes from the SVD branch 2015-11-19 12:37:18 -05:00
Philippe Tillet
714e0f5634 API: Fixed single-element indexing 2015-10-07 01:13:55 -04:00
Philippe Tillet
07b8ba20de API: some fixes with 1D slices 2015-10-06 16:34:47 -04:00
Philippe Tillet
8daf13da2e Code quality: some renaming here and there 2015-10-05 14:35:46 -04:00
Philippe Tillet
07e7bd862c API: added diag(matrix) 2015-10-04 17:05:06 -04:00
Philippe Tillet
740f5def49 API: polished slice construction 2015-10-03 19:30:50 -04:00
Philippe Tillet
feeb1e9862 Feature: Merged kernel-fusion branch 
* Fuses multiple AXPY kernel
* Possibility to add thread-wise for loops in AXPY-like kernels
 2015-09-30 15:31:41 -04:00
Philippe Tillet
836a955663 GEMV: bugfix with CUDA 2015-08-30 02:35:55 -04:00
Philippe Tillet
1e77703f7f Android: various fixes 2015-08-28 13:48:54 -04:00
Philippe Tillet
53dcbfa1e0 Kernels [GEMM]: restored vector types on CUDA 2015-08-27 22:55:38 -04:00
Philippe Tillet
f06a3bdf53 Bugfix: fixed bug in dynamic kernel selection 2015-08-26 19:11:09 -04:00
Philippe Tillet
67a35a62bd Driver: now loading the backend dynamically on Linux 2015-08-25 17:06:51 -04:00
Philippe Tillet
868df9dad2 Kernels: uncommented some GEMM code 2015-08-24 23:09:23 -04:00
Philippe Tillet
95f2564c1a Tuning: Android UI improvement 2015-08-24 23:03:37 -04:00
Philippe Tillet
10524ebdee CUDA: various improvements 2015-08-24 17:03:31 -04:00
Philippe Tillet
feda5a8c9c C++: added access to CL's platform id via driver::Platform 2015-08-18 14:21:12 -07:00
Philippe Tillet
da20db711b Tune: Further file hierarchy improvements 2015-08-18 02:31:44 -07:00
Philippe Tillet
a23e976ca7 Kernels: Bugfix with large temporary sizes 2015-08-17 16:32:10 -07:00
Philippe Tillet
9bed56fff4 Profiles: Now throwing operation_not_supported exception at the right place 2015-08-17 16:32:10 -07:00
U-AMR\ptillet
b34c611802 Code quality: Added consistency between int_t and size_t. Fixed warnings for Win64 2015-08-13 16:00:49 -07:00
Philippe Tillet
f7cb4ac960 Code quality: fixed implicit conversions from size_t to int_t 2015-08-13 14:30:11 -07:00
Philippe Tillet
0bb73602f9 Kernel: Merged gemv-simd code 2015-08-13 10:15:32 -07:00
Philippe Tillet
ff4cf94df7 Code quality: significant cleaning of namespaces, etc.. 2015-08-12 00:47:58 -07:00
Philippe Tillet
5a8cfede45 Code quality: renamed model/ to database/ 2015-08-11 20:18:39 -07:00
Philippe Tillet
b5cc1f7ddc Kernels: Now SizeType is always "int". Right now I don't expect data-structure to have more than 2**31 entries. Improves performance on a number of routines. 2015-08-11 11:50:49 -07:00
Philippe Tillet
963867574f Kernels: merged start1, start2 and stride1, stride2 into start and stride for matrices 2015-08-10 22:45:48 -07:00
U-AMR\ptillet
0a506fa433 Code quality: fixed build issues with MSVC 2015-08-10 17:04:57 -07:00
Philippe Tillet
f60b82af25 Kernels: more generic temporary workspace checks 2015-08-10 10:19:50 -07:00
Philippe Tillet
9572b94ac1 Models: added sanity check for the predicted template 2015-08-07 23:05:44 -07:00
Philippe Tillet
d05296603a Kernels: Fixup GEMM broken by last commit 2015-08-07 22:11:37 -07:00
Philippe Tillet
1399404f04 Code quality: Fixed issue with to_string for proper compilation on Cygwin GCC 2015-08-06 20:20:08 -07:00
Philippe Tillet
33fea11547 Code quality: more cleaning of files architecture 2015-08-06 19:34:26 -07:00
Philippe Tillet
e4ff883688 Code quality: Cleaned a bit file hierarchy in kernel templates 2015-08-06 16:14:33 -07:00
Philippe Tillet
db090d7942 Code quality: Large clean-up of the codebase and especially of the include/ folder 2015-08-06 12:05:12 -07:00
Philippe Tillet
df9f6142ef Models: added basic database structure 2015-08-05 17:01:42 -07:00
Philippe Tillet
bb4d2d62e3 Code quality: disabled the use of strcat / sprintf for safety issues on windows... 2015-08-05 11:42:08 -07:00
Philippe Tillet
f4c597b294 Code quality: fixed compilation errors/warnings with Clang 2015-08-05 09:26:50 -07:00