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
/
tuning
Age
Commit message (
Expand
)
Author
2018-03-06
Fixed compilation issue in Xger tuner
Cedric Nugteren
2018-03-06
First version of the tuning API, added interface for copy-kernel, added sample
Cedric Nugteren
2018-03-03
Separate kernel tuners in .cpp with main and .hpp with settings
Cedric Nugteren
2018-02-20
Fixed several issues in the new invert tuner
Cedric Nugteren
2018-01-25
Moved some constants from global scope to a function; removed unnecessary inc...
Cedric Nugteren
2018-01-25
Changed the default number of runs for the GEMV tuner to fix issues for FP16
Cedric Nugteren
2018-01-18
Made GEMM routine tuning a bit more generic in preparation of possible separa...
Cedric Nugteren
2018-01-15
Factored out the generic parts of the GEMM routine tuner
Cedric Nugteren
2018-01-06
Fixed a vendor naming bug in the tuners and in the database
Cedric Nugteren
2017-12-23
Fixed unused variable warnings showing up with Clang
Cedric Nugteren
2017-12-23
Now calling main TRSV routine again to fix compilation in MSVC
Cedric Nugteren
2017-12-23
Split the invert kernel in two parts to prevent error C1091 in MSVC 2013
Cedric Nugteren
2017-12-23
Updated the database to use the new TRSV and Invert tuners
Cedric Nugteren
2017-12-23
Added TRSV block-size tuner
Cedric Nugteren
2017-12-19
Added skeleton for a tuner for the invert kernel
Cedric Nugteren
2017-12-18
Reformatted tuning code to make compilation faster
Cedric Nugteren
2017-12-17
Fixed an issue with the tuner: it was using platform vendor rather than devic...
Cedric Nugteren
2017-12-17
Fixed an unnecessary overflow issue on 32-bit systems
Cedric Nugteren
2017-12-10
Split GEMM kernel in 4 files instead of 3 due to MSVC 2013 string length limit
Cedric Nugteren
2017-12-10
Fixed an issue in the tuners to prevent error -14 from persisting (CL_EXEC_ST...
Cedric Nugteren
2017-12-09
Made the pre-processor run by default for ARM and Qualcomm GPUs
Cedric Nugteren
2017-11-30
Integrated pre-processor in compilation flow, default is still disabled
Cedric Nugteren
2017-11-20
Fixes some displaying issues in the GEMM routine tuner
Cedric Nugteren
2017-11-19
Fixed a variety of warnings and an error for MSVC2013 compilation
Cedric Nugteren
2017-11-19
Added compilation timing and better compilation error reporting
Cedric Nugteren
2017-11-19
Some fixed for the new auto-tuner to be compatible with the Python scripts
Cedric Nugteren
2017-11-19
Revived the GEMM routine tuner; minor formatting changes
Cedric Nugteren
2017-11-19
Modified the kernel tuners to use the newly integrated auto-tuner
Cedric Nugteren
2017-11-17
Moved some tuning functions from .hpp to .cpp
Cedric Nugteren
2017-11-17
Moved compilation function to separate file; removed dependency of tuners of ...
Cedric Nugteren
2017-11-16
Added printing of the best parameters for the new tuner
Cedric Nugteren
2017-11-15
Added first version of integrated and re-written auto-tuner
Cedric Nugteren
2017-11-06
Changed GEMM routine tuner's scoring to use L2 measure instead for better ave...
Cedric Nugteren
2017-11-02
Integrated the GEMM routine tuner for kernel selection; added first tuning re...
Cedric Nugteren
2017-10-30
Added collecting and printing of scores for the kernel-selection tuner
Cedric Nugteren
2017-10-28
Added initial version of a GEMM kernel selection tuner
Cedric Nugteren
2017-10-03
Gemm in-direct implementation now uses only 1 larger instead of max 3 optiona...
Cedric Nugteren
2017-09-30
Refactored the tuning architecture: less duplicate now; more defaults
Cedric Nugteren
2017-09-10
Added the new vendor-architecture-name hierarchy to the tuners as well
Cedric Nugteren
2017-08-31
Fixed some things in the tuner: bugs, style, and defaults to random search
Cedric Nugteren
2017-08-21
Minor updates after merging in the PSO addition to the tuners
Cedric Nugteren
2017-08-21
Remove multistrategy and related functions
mcian
2017-08-09
Revert the xgemm strategy to default. If user wants to use multistrategy can ...
mcian
2017-08-09
Use cltune::SearchMethod enum instead of int values
mcian
2017-07-23
Code refactoring
mcian
2017-07-17
Add PSO parameters support and search strategy selection from command line
mcian
2017-05-11
Re-added random tuning for GEMM after accidental removal
Cedric Nugteren
2017-04-22
Increased the default number of runs for the tuner from 2 up to 10 for fast k...
Cedric Nugteren
2017-04-21
Increased the default number of runs for GEMV tuning; updated GEMV tuning res...
Cedric Nugteren
2017-04-17
Fixed a namespace clash with CUDA FP16 for the half-datatype
Cedric Nugteren
[next]