 Samuel Thibault
|
0977720ba0
port r12586 from 1.1: Make sure the user has synchronized with the local stream
|
11 years ago |
|
Samuel Thibault
|
f2b1d0f595
port r12584 from 1.1: Re-enable asynchronous transfers from CPUs, now that the issue has been found to be in application code
|
11 years ago |
 Andra Hugo
|
a0c9760232
free topology tree + fix small bug of exec_paralle_code
|
11 years ago |
|
Samuel Thibault
|
d0eab8ef8c
Do not rely on the .h file for functions actually available at link time, do check them explicitly
|
11 years ago |
|
Samuel Thibault
|
deb5a22184
Document STARPU_NWORKER_PER_CUDA
|
11 years ago |
 Nathalie Furmento
|
a5977c4431
src/datawizard/memalloc.c: interface may be a keyword depending on the used compiler
|
11 years ago |
|
Nathalie Furmento
|
292f8101d8
tests/perfmodels/user_base.c: fix function return type
|
11 years ago |
|
Nathalie Furmento
|
4ce96baf81
src/datawizard/data_request.c: add parenthesis to enhance clarity
|
11 years ago |
|
Nathalie Furmento
|
4f0c5ac5c6
include/starpu_perfmodel.h: define a invalid value for perfmodel type
|
11 years ago |
|
Nathalie Furmento
|
95b0599a65
doxygen: fix ref commands and small typos
|
11 years ago |
|
Nathalie Furmento
|
baa674826b
doxygen: silent warnings
|
11 years ago |
|
Samuel Thibault
|
83a38dfa85
clEnqueueMarker is deprecated by clEnqueueMarkerWithWaitList. Be able to use both
|
11 years ago |
 Olivier Aumage
|
1ff35b63df
- fix another wrong ifdef
|
11 years ago |
|
Olivier Aumage
|
f357759b05
- fix wrong ifdef test
|
11 years ago |
|
Andra Hugo
|
5a219c8a8f
* patch Terry: fix book workers (book workers that have already been booked and wake up eventually workers that we don't want anymore in the group)
|
11 years ago |
|
Samuel Thibault
|
092f322b1c
Add CUDA concurrent kernel execution support through the STARPU_NWORKER_PER_CUDA environment variable.
|
11 years ago |
|
Samuel Thibault
|
2963604386
port r12554 from 1.1: Disable timing on events for transfers, we don't use it
|
11 years ago |
|
Samuel Thibault
|
28308c10be
Fix array size
|
11 years ago |
|
Samuel Thibault
|
f400e9b510
Update variables to unset
|
11 years ago |
|
Samuel Thibault
|
f249eeb264
Optimize getting the current task
|
11 years ago |
|
Samuel Thibault
|
d996ed4da4
drop unused variable
|
11 years ago |
|
Samuel Thibault
|
38fbca49e9
Fix _starpu_datawizard_progress prototype: it does not actually return anything
|
11 years ago |
|
Samuel Thibault
|
8bfbe1ee39
Drop mentioning Gordon
|
11 years ago |
|
Samuel Thibault
|
8bcae573b8
Drop duplicate wait
|
11 years ago |
|
Samuel Thibault
|
e8a62c8fce
move spurious variable initialization
|
11 years ago |
|
Samuel Thibault
|
c60d91cb3f
Enable asynchronous OpenCL wherever possible
|
11 years ago |
|
Samuel Thibault
|
840af37490
push pending requests at the back of the list of pending requests, not the front
|
11 years ago |
|
Samuel Thibault
|
058e113825
port r12533 from 1.1: Do not let non-CUDA workers use non-0 streams, CUDA seems not very threadsafe with that. Make the coherency engine avoid selecting non-CUDA workers to issue transfers, to avoid letting them use the 0 stream.
|
11 years ago |
|
Samuel Thibault
|
9e7c9936ce
Use STARPU_MAIN_RAM where appropriate
|
11 years ago |
|
Samuel Thibault
|
9aa1c5ba6c
Use STARPU_MAIN_RAM where appropriate
|
11 years ago |
