Olivier Aumage
|
7acc0949a7
- merge trunk
|
11 anos atrás |
Olivier Aumage
|
0bb92b846b
- more ongoing work on the omp runtime support
|
11 anos atrás |
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 anos atrás |
Samuel Thibault
|
092f322b1c
Add CUDA concurrent kernel execution support through the STARPU_NWORKER_PER_CUDA environment variable.
|
11 anos atrás |
Samuel Thibault
|
2963604386
port r12554 from 1.1: Disable timing on events for transfers, we don't use it
|
11 anos atrás |
Samuel Thibault
|
28308c10be
Fix array size
|
11 anos atrás |
Samuel Thibault
|
f400e9b510
Update variables to unset
|
11 anos atrás |
Samuel Thibault
|
f249eeb264
Optimize getting the current task
|
11 anos atrás |
Samuel Thibault
|
d996ed4da4
drop unused variable
|
11 anos atrás |
Samuel Thibault
|
38fbca49e9
Fix _starpu_datawizard_progress prototype: it does not actually return anything
|
11 anos atrás |
Samuel Thibault
|
8bfbe1ee39
Drop mentioning Gordon
|
11 anos atrás |
Samuel Thibault
|
8bcae573b8
Drop duplicate wait
|
11 anos atrás |
Samuel Thibault
|
e8a62c8fce
move spurious variable initialization
|
11 anos atrás |
Olivier Aumage
|
18cb0e9e8c
- merge trunk
|
11 anos atrás |
Olivier Aumage
|
934fa3edfc
- initial thread handling
|
11 anos atrás |
Samuel Thibault
|
c60d91cb3f
Enable asynchronous OpenCL wherever possible
|
11 anos atrás |
Olivier Aumage
|
1712f84e3b
- merge trunk
|
11 anos atrás |
Olivier Aumage
|
09ea840391
- ongoing work on openmp runtime support
|
11 anos atrás |
Samuel Thibault
|
840af37490
push pending requests at the back of the list of pending requests, not the front
|
11 anos atrás |
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 anos atrás |
Samuel Thibault
|
9e7c9936ce
Use STARPU_MAIN_RAM where appropriate
|
11 anos atrás |
Samuel Thibault
|
9aa1c5ba6c
Use STARPU_MAIN_RAM where appropriate
|
11 anos atrás |
Samuel Thibault
|
2bafbc890a
fix build
|
11 anos atrás |
Samuel Thibault
|
92d443d920
Make documentation clearer
|
11 anos atrás |
Samuel Thibault
|
ea5d8f596b
Make a copy of the interface to the memchunk only when the latter gets detached from the data, and thus the interface code will not work on it. Drop the copy when the memchunk gets reattached. This allows interfaces to modify pointers in the interface in unpack, notably
|
11 anos atrás |
Samuel Thibault
|
fb03301fdf
port r12523 from 1.1: Fix detaching a memchunk from a handle: the memchunk is not supposed to still have a pointer to the replicate
|
11 anos atrás |
Samuel Thibault
|
1aad553cc2
Fix crash with perfmodels having type common
|
11 anos atrás |
Andra Hugo
|
aed609a645
rollback r12518: don't allow unregister in callback or task codelets
|
11 anos atrás |
Andra Hugo
|
84e77b5ad7
add prolog callbacks to mpi
|
11 anos atrás |
Olivier Aumage
|
84fe3d8fb6
- merge trunk
|
11 anos atrás |