summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2017-07-08Made the inline keyword in kernels optional currently only enabled for ↵Cedric Nugteren
NVIDIA and ARM GPUs
2017-06-30Fixed an if-statement in the direct GEMM kernel causing a bug with specific ↵Cedric Nugteren
sets of input parameters
2017-06-25Fixed some Clang and MSVC warningsCedric Nugteren
2017-06-21Fixes some compilation issues related to the database structure changeCedric Nugteren
2017-06-20Changed the structure of the database to reduce compilation time and save memoryCedric Nugteren
2017-06-18Added tuning results for the Core i7-920 CPUCedric Nugteren
2017-06-18Fixed an overflow bug on 32-bit systems when chosing a GEMM kernelCedric Nugteren
2017-06-01Added tuning results for GeForce GT 650M (thanks to bzcheeseman)Cedric Nugteren
2017-05-27Merge pull request #158 from CNugteren/msvc_compilation_fixesCedric Nugteren
MSVC compilation fixes
2017-05-27Fixed comment decribing the order of program cache fieldsKirill Mavreshko
2017-05-26Fixed a compilation issue under MSVC 2013Cedric Nugteren
2017-05-26Fixes inability to run GEMM on multiple identical GPUs (issue #155)Kirill Mavreshko
2017-05-15Fixed an TRSM issue caused by incorrect block size calculationCedric Nugteren
2017-05-14Fixed a missing synchronization barrier in the invert kernel; fixes TRSM testsCedric Nugteren
2017-05-12Added the IxAMIN routines: absolute minimum version of IxAMAXCedric Nugteren
2017-05-12Fixed a bug in the TRSM routine; tests now passCedric Nugteren
2017-05-11Added tuning results for the AMD Radeon Fiji GPUCedric Nugteren
2017-05-11Re-added random tuning for GEMM after accidental removalCedric Nugteren
2017-04-23Re-added Titan X (Pascal) tuning results based on more averaging when tuningCedric Nugteren
2017-04-22Increased the default number of runs for the tuner from 2 up to 10 for fast ↵Cedric Nugteren
kernels
2017-04-22Fixed the direct vs indirect setting for NVIDIA GPUsCedric Nugteren
2017-04-21Increased the default number of runs for GEMV tuning; updated GEMV tuning ↵Cedric Nugteren
results for Iris Pro
2017-04-20Tuned the direct versus indirect GEMM kernel trade-off point for NVIDIA GPUsCedric Nugteren
2017-04-17Fixed a namespace clash with CUDA FP16 for the half-datatypeCedric Nugteren
2017-04-16Merge branch 'development' into benchmarkingCedric Nugteren
2017-04-14Added a new Xaxpy kernel in between the regular and fast version inCedric Nugteren
2017-04-13Fixed CUDA malloc and cuBLAS handles: cuBLAS as a performance-reference now ↵Cedric Nugteren
works
2017-04-10Merge branch 'development' into cublas_referenceCedric Nugteren
Conflicts: scripts/generator/generator.py
2017-04-10Fixed a compilation issue under MSVC and GCCCedric Nugteren
2017-04-10Removed const-vector-of-const-objects from the database class to remain ↵Cedric Nugteren
according to the C++11 standard
2017-04-07Added a special override database for the Apple CPU implementation on OS X: ↵Cedric Nugteren
this makes the test work, it does not focus on good performance
2017-04-07Uses float2 and double2 for base complex data-types instead of a custom ↵Cedric Nugteren
struct; fixes bug on Apple OpenCL
2017-04-07Added some missing const-nessCedric Nugteren
2017-04-02Layed the groundwork for cuBLAS comparisons in the clientsCedric Nugteren
2017-04-01Separated host-device and device-host memory copies from execution of the ↵Cedric Nugteren
CBLAS reference code; for fair timing and code de-duplication
2017-03-19Added an (optional) non-direct implementation of the batched GEMM routineCedric Nugteren
2017-03-19Added batched versions of the pad/copy/transpose kernelsCedric Nugteren
2017-03-14Added the possibility to tune batched kernelsCedric Nugteren
2017-03-11Added initial naive version of the batched GEMM routine based on the direct ↵Cedric Nugteren
GEMM kernel
2017-03-10Added API and test infrastructure for the batched GEMM routineCedric Nugteren
2017-03-10Added proper testing of the alpha parameter; finalized the batched AXPY ↵Cedric Nugteren
implementation
2017-03-10Fixed a small compilation bug for MSVC related to a floating-point constantCedric Nugteren
2017-03-08Implemented a batched version of the AXPY kernelCedric Nugteren
2017-03-08Make batched routines based on offsets instead of a vector of cl_mem objects ↵Cedric Nugteren
- undoing many earlier changes
2017-03-05Minor fixes to the client w.r.t. the addition of the batch countCedric Nugteren
2017-03-05Added first naive version of the batched AXPY routineCedric Nugteren
2017-03-05Adjusted the test-infrastructure to support testing of batched-versions of ↵Cedric Nugteren
routines
2017-03-05Changed the way the test-data is generated: now using a single MT generator ↵Cedric Nugteren
and distribution for all data
2017-03-05Prepared generator for batched routines; added batched AXPY routine interfaceCedric Nugteren
2017-03-04Added tuning results for the Radeon HD6750M GPU (Apple OpenCL)Cedric Nugteren
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-20 12:02:11 +0000