CNugteren
|
ede78fe499
|
Fixed the license for the correctness testers
|
2015-06-29 20:39:51 +02:00 |
|
CNugteren
|
2914a285d4
|
Re-organized the performance-client infrastructure to avoid code duplication
|
2015-06-29 20:38:34 +02:00 |
|
CNugteren
|
e5c0edbfd7
|
Re-organized the test infrastructure to avoid code duplication
|
2015-06-28 15:52:57 +02:00 |
|
CNugteren
|
cf1892d22c
|
Added buffer structure and sizes to arguments
|
2015-06-28 15:37:38 +02:00 |
|
Cedric Nugteren
|
77e2157485
|
Merge pull request #9 from CNugteren/level3_routines
Added SYRK and SYR2K level-3 routines
|
2015-06-26 20:56:21 +02:00 |
|
CNugteren
|
e27e339ebf
|
Replaced crosses with tickmarks
|
2015-06-26 17:43:17 +02:00 |
|
CNugteren
|
7c8d16147a
|
Added the SYR2K routine, tester, and client
|
2015-06-26 08:12:56 +02:00 |
|
CNugteren
|
75f263ce3a
|
Added symmetric matrix support for the ABC performance tester
|
2015-06-26 08:10:23 +02:00 |
|
CNugteren
|
ff9f9fac57
|
Added option to test only symmetric matrices (m=n)
|
2015-06-25 20:39:34 +02:00 |
|
CNugteren
|
57c705dbf2
|
Clarified comment
|
2015-06-25 20:38:34 +02:00 |
|
CNugteren
|
96e4012349
|
Added SSYRK performance graphs
|
2015-06-25 19:19:31 +02:00 |
|
CNugteren
|
3de4471afe
|
Added the SYRK routine
|
2015-06-24 07:52:19 +02:00 |
|
CNugteren
|
60a88aac86
|
Added the SYRK routine, tester, and client
|
2015-06-24 07:50:18 +02:00 |
|
CNugteren
|
a17297937d
|
Added performance-client for AC routines
|
2015-06-23 22:31:27 +02:00 |
|
CNugteren
|
9fc38cdf5e
|
Added a lower/upper triangular version of the GEMM kernel
|
2015-06-23 17:58:51 +02:00 |
|
CNugteren
|
0a3831e6d1
|
Updated bandwidth computation for GEMM and SYMM
|
2015-06-23 08:09:46 +02:00 |
|
CNugteren
|
20eb3506d6
|
Added a condition to update only lower/upper triangular parts in the un-pad kernels
|
2015-06-23 08:09:07 +02:00 |
|
CNugteren
|
4c2a166bc5
|
Added test infrastructure for AB and AC routines
|
2015-06-21 12:57:38 +02:00 |
|
CNugteren
|
e3829c1067
|
Added prototypes of SYRK and SYR2K
|
2015-06-21 12:44:03 +02:00 |
|
CNugteren
|
ea7da6a497
|
Fixed support for complex data-types for GEMM and SYMM clients
|
2015-06-21 11:21:03 +02:00 |
|
Cedric Nugteren
|
18251df848
|
Merge pull request #7 from CNugteren/development
Update to version 0.2.0
|
2015-06-21 09:15:41 +02:00 |
|
CNugteren
|
985eeac503
|
Updated to version 0.2.0
|
2015-06-21 09:13:08 +02:00 |
|
CNugteren
|
6aac23be86
|
Updated performance graphs for Intel Iris GPUs
|
2015-06-21 09:12:42 +02:00 |
|
Cedric Nugteren
|
e19893ff15
|
Merge pull request #5 from CNugteren/improved_testing
Improved testing infrastructure
|
2015-06-20 17:02:23 +02:00 |
|
CNugteren
|
84dd6ba1d7
|
Updated changelog with testing improvements
|
2015-06-20 16:47:50 +02:00 |
|
CNugteren
|
c365614eb5
|
More detailed test passed/skipped/failure reporting
|
2015-06-20 16:43:50 +02:00 |
|
CNugteren
|
0f486d9b74
|
Automatically skips tests with unsupported precision
|
2015-06-20 14:13:54 +02:00 |
|
CNugteren
|
3ea3ba2bee
|
Distinguish between a short smoke test and a full test
|
2015-06-20 13:33:50 +02:00 |
|
CNugteren
|
dfbc336531
|
Moved the argument parsing to the common Tester base class
|
2015-06-20 11:26:01 +02:00 |
|
CNugteren
|
e26742c629
|
Added additional absolute error checking when testing
|
2015-06-20 10:58:21 +02:00 |
|
Cedric Nugteren
|
d7097a063a
|
Merge pull request #4 from CNugteren/host_code_perf
Host-code performance
|
2015-06-19 07:35:48 +02:00 |
|
CNugteren
|
41ce480c51
|
Updated changelog with host-code performance optimisation
|
2015-06-19 07:34:00 +02:00 |
|
CNugteren
|
ab55df703d
|
Added const-ref accessors to all CL++11 classes
|
2015-06-19 07:28:35 +02:00 |
|
CNugteren
|
682c01a80c
|
Now returns program from database by reference
|
2015-06-18 18:44:14 +02:00 |
|
Cedric Nugteren
|
af40b8e667
|
Merge pull request #3 from CNugteren/conjugate_transpose
Added complex conjugate transpose
|
2015-06-17 07:15:30 +02:00 |
|
CNugteren
|
8b2dbdba98
|
Updated with conjugate transpose and CGEMM/ZGEMM CSYMM/ZSYMM
|
2015-06-17 07:12:45 +02:00 |
|
CNugteren
|
7e176ccac9
|
Added support for conjugate transpose in GEMV
|
2015-06-16 08:42:52 +02:00 |
|
CNugteren
|
d7a0d970e0
|
Moved layout and transpose types to generic tester and AXY routines
|
2015-06-16 08:07:26 +02:00 |
|
CNugteren
|
af78a04eca
|
Updated the tuners to set the conjugate argument
|
2015-06-16 07:50:45 +02:00 |
|
CNugteren
|
e03582a112
|
Added support for CGEMM/ZGEMM and CSYMM/ZSYMM
|
2015-06-16 07:45:09 +02:00 |
|
CNugteren
|
6ebca73d72
|
Created data-type dependent transpose values to test
|
2015-06-16 07:44:32 +02:00 |
|
CNugteren
|
8f01c644b5
|
Added support for complex conjugate transpose
|
2015-06-16 07:43:19 +02:00 |
|
Cedric Nugteren
|
9e2fba9ab9
|
Merge pull request #2 from CNugteren/xgemv
Added GEMV routine
|
2015-06-15 18:35:27 +02:00 |
|
CNugteren
|
f925d47dad
|
Added GEMV to changelog and readme
|
2015-06-15 08:41:37 +02:00 |
|
CNugteren
|
ce703a2f5a
|
Added tuning for DGEMV on Iris and SGEMV on K40m
|
2015-06-15 08:41:13 +02:00 |
|
CNugteren
|
e63d20e0b6
|
Enable tests of DGEMV/CGEMV/ZGEMV
|
2015-06-15 08:40:41 +02:00 |
|
CNugteren
|
01726197ab
|
Fixed a bug in AXPBY defines for complex data-types
|
2015-06-15 08:38:24 +02:00 |
|
CNugteren
|
7776323cb4
|
Added GEMV performance-graph script
|
2015-06-14 12:05:30 +02:00 |
|
CNugteren
|
294a3e3d41
|
Split the three variations of the GEMV kernel for maximal tuning freedom
|
2015-06-14 11:15:53 +02:00 |
|
CNugteren
|
ab0064dab7
|
Fixed number of threads launched for GEMV
|
2015-06-14 10:08:56 +02:00 |
|