| .. |
|
common
|
9bb7bdb07f
Increment/decrement submitted/ready tasks per context (instead of doing it per context & globally)
|
%!s(int64=12) %!d(string=hai) anos |
|
core
|
c0cb3f849f
port r 11459 from 1.1: Do not force calibration systematically when a history-based model is used. Let the scheduler trigger it when it actually needs an expected performance
|
%!s(int64=12) %!d(string=hai) anos |
|
datawizard
|
bb88fe0e68
port r11466 from 1.1: Even without allocation cache, data release is done lazily, we can not count on precision of value returned by starpu_memory_get_available
|
%!s(int64=12) %!d(string=hai) anos |
|
debug
|
7f56eab83f
port r11326 from 1.1: fix typo
|
%!s(int64=12) %!d(string=hai) anos |
|
drivers
|
020bb9e52a
port r11429 from 1.1: Enable gpu-gpu direct transfer. This does not seem to harm performance any more
|
%!s(int64=12) %!d(string=hai) anos |
|
profiling
|
e2b943451f
mic (perfmodel): merge trunk + finalize perfmodel
|
%!s(int64=13) %!d(string=hai) anos |
|
sched_policies
|
4eb6989308
Do not let the prio scheduler decrement task stats several times when popping a task with multiple available implementations
|
%!s(int64=12) %!d(string=hai) anos |
|
top
|
dbc393bed8
src/top/starpu_top_connection.c: make sure WINVER is defined before including windows include files
|
%!s(int64=13) %!d(string=hai) anos |
|
util
|
8f69d75bdd
move implementation of starpu_task_list functions in public .h file
|
%!s(int64=12) %!d(string=hai) anos |
|
worker_collection
|
3a7711af02
src: remove un-needed include pthread.h
|
%!s(int64=13) %!d(string=hai) anos |
|
.gitignore
|
27eaf1ca36
Add `.gitignore'.
|
%!s(int64=15) %!d(string=hai) anos |
|
Makefile.am
|
8f69d75bdd
move implementation of starpu_task_list functions in public .h file
|
%!s(int64=12) %!d(string=hai) anos |
|
dolib.c
|
f5dee9ca79
Add effective version in dll path
|
%!s(int64=14) %!d(string=hai) anos |
|
starpu_parameters.h
|
e2b943451f
mic (perfmodel): merge trunk + finalize perfmodel
|
%!s(int64=13) %!d(string=hai) anos |
Samuel Thibault
bb88fe0e68
port r11466 from 1.1: Even without allocation cache, data release is done lazily, we can not count on precision of value returned by starpu_memory_get_available