index
:
debian-clblast
debian/sid
upstream/latest
Debian package for CLBlast.
gspr@nonempty.org
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
src
/
kernels
/
level3
Age
Commit message (
Expand
)
Author
2017-12-09
Fixed defines parsing and substituting in pre-processor; fixed some variable ...
Cedric Nugteren
2017-12-07
Added register promotion to the main GEMM kernel
Cedric Nugteren
2017-12-03
Added GEMM (direct and in-direct) to the pre-processor testing; modified the ...
Cedric Nugteren
2017-12-03
Reformated transpose kernels for the pre-processor; extended the amount of tests
Cedric Nugteren
2017-11-29
Reformatted unrollable kernel loops and added the new promote_to_registers pr...
Cedric Nugteren
2017-10-14
Fixed a kernel/attribute order bug in the direct GEMM kernels
Cedric Nugteren
2017-10-14
Make local memory pointers a define in OpenCL; some fixes to the recently cha...
Cedric Nugteren
2017-10-14
Made transpose kernel struct init proper according to the C standard
Cedric Nugteren
2017-10-03
Gemm in-direct implementation now uses only 1 larger instead of max 3 optiona...
Cedric Nugteren
2017-07-08
Made the inline keyword in kernels optional currently only enabled for NVIDIA...
Cedric Nugteren
2017-06-30
Fixed an if-statement in the direct GEMM kernel causing a bug with specific s...
Cedric Nugteren
2017-05-14
Fixed a missing synchronization barrier in the invert kernel; fixes TRSM tests
Cedric Nugteren
2017-03-19
Added an (optional) non-direct implementation of the batched GEMM routine
Cedric Nugteren
2017-03-19
Added batched versions of the pad/copy/transpose kernels
Cedric Nugteren
2017-03-11
Added initial naive version of the batched GEMM routine based on the direct G...
Cedric Nugteren
2017-03-04
Added a proper data-preparation function for the TRSM tests
Cedric Nugteren
2017-02-26
Fixed an out-of-bounds memory access when filling a matrix with a constant
Cedric Nugteren
2017-02-26
Fixes division in the kernel for inversion of complex numbers
Cedric Nugteren
2017-02-25
Added PrepareData function for TRSM to create proper test input
Cedric Nugteren
2017-01-18
Added first version of the TRSM routine based on the diagonal invert kernel
Cedric Nugteren
2017-01-15
Added a first version of the diagonal block invert routine in preparation of ...
Cedric Nugteren
2016-12-18
Fixed a bug when using offsets in the direct GEMM kernels
Cedric Nugteren
2016-10-22
Fixed a bug in the SYRK/SYR2K/HERK/HER2K routines that would occur with speci...
Cedric Nugteren
2016-10-22
Fixed a bug in the SYRK/SYR2K/HERK/HER2K routines that would occur with speci...
Cedric Nugteren
2016-10-03
Fixed a const-correctness issue with complex conjugation in the GEMM direct k...
Cedric Nugteren
2016-10-03
Added functions to load from off-chip to local memory without vector loads fo...
Cedric Nugteren
2016-10-03
Re-organised GEMM direct kernel and added faster fall-back version for incomp...
Cedric Nugteren
2016-10-02
Specialised the GEMM direct kernel in four ways for transposing/non-transposi...
Cedric Nugteren
2016-10-02
Split the GEMM direct kernel into two files; set the default tuning target to...
Cedric Nugteren
2016-10-01
Added padding to the local memory of the GEMM direct kernel
Cedric Nugteren
2016-09-25
Added a first version of a tuner for the GEMM direct kernel; collapsed MWGD, ...
Cedric Nugteren
2016-09-25
Separated the tuning parameters of the new direct GEMM kernel from the indire...
Cedric Nugteren
2016-09-25
Added a first version of the direct version of GEMM with local memory
Cedric Nugteren
2016-09-21
Merge branch 'development' into gemm_direct
Cedric Nugteren
2016-09-12
Split the XGEMM kernel further up: now in 3 parts. This is done because MSVC ...
Cedric Nugteren
2016-09-04
The GEMM kernel no longer adds beta*C in case beta is zero; this would cause ...
Cedric Nugteren
2016-08-20
Merge branch 'master' of https://github.com/dvasschemacq/CLBlast into dvassch...
Cedric Nugteren
2016-08-18
Adapt opencl files for 1.1 OpenCL
D. Van Assche
2016-07-17
Improved the GEMM direct kernel by adding register blocking. Still not fast t...
Cedric Nugteren
2016-07-16
Created infrastructure to support a direct GEMM kernel; added correct but slo...
Cedric Nugteren
2016-07-10
Now passing alpha/beta to the kernel as arguments as before fp16 support; in ...
Cedric Nugteren
2016-06-16
Added XOMATCOPY routines to perform out-of-place matrix scaling, copying, and...
Cedric Nugteren
2016-06-14
Re-organised the level-3 supporting kernels (copy, pad, transpose, convert) a...
Cedric Nugteren
2016-06-08
Added global memory synchronisation for better cache performance on ARM Mali ...
Cedric Nugteren
2016-05-18
Merged in latest changes from 0.7.1 release
Cedric Nugteren
2016-05-16
Prepared GEMM and supporting kernels and tuners for half-precision support
Cedric Nugteren
2016-02-08
Separated the GEMM kernel in two parts to reduce string length for MSVC
Cedric Nugteren
2015-10-13
Added guards for routine-specific level-3 pad kernels
CNugteren
2015-10-12
Moved level3 kernel files to a subfolder
CNugteren