 Samuel Thibault
|
840af37490
push pending requests at the back of the list of pending requests, not the front
|
11 lat temu |
|
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 lat temu |
|
Samuel Thibault
|
9e7c9936ce
Use STARPU_MAIN_RAM where appropriate
|
11 lat temu |
|
Samuel Thibault
|
9aa1c5ba6c
Use STARPU_MAIN_RAM where appropriate
|
11 lat temu |
|
Samuel Thibault
|
2bafbc890a
fix build
|
11 lat temu |
|
Samuel Thibault
|
92d443d920
Make documentation clearer
|
11 lat temu |
|
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 lat temu |
|
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 lat temu |
|
Samuel Thibault
|
1aad553cc2
Fix crash with perfmodels having type common
|
11 lat temu |
 Andra Hugo
|
aed609a645
rollback r12518: don't allow unregister in callback or task codelets
|
11 lat temu |
|
Andra Hugo
|
84e77b5ad7
add prolog callbacks to mpi
|
11 lat temu |
|
Andra Hugo
|
a78e7bdc6a
allow unregisters in callbacks and codelets when we have only CPUs
|
11 lat temu |
|
Andra Hugo
|
2938dd4805
patch Terry: add a new prologue callback (one executed at pop time)
|
11 lat temu |
|
Andra Hugo
|
e991168dec
rollback book_workers
|
11 lat temu |
|
Samuel Thibault
|
5c4a168e72
Do not destroy containers before using them
|
11 lat temu |
|
Samuel Thibault
|
5a9ffce947
backport r12506: fix disabling out-of-order for data transfers
|
11 lat temu |
|
Samuel Thibault
|
bf8aa6338d
Let the OpenCL driver progress while the GPU is computing
|
11 lat temu |
|
Samuel Thibault
|
7d2bf0c630
Enable asynchronous CUDA kernels wherever possible
|
11 lat temu |
|
Samuel Thibault
|
7ce4ced85d
Let the CUDA driver progress while the GPU is computing
|
11 lat temu |
|
Samuel Thibault
|
42925b7f8e
Do not define _XOPEN_SOURCE, as it disables GNU extensions, which users don't expect
|
11 lat temu |
 Olivier Aumage
|
40432cdc2c
- first word after \section is the section name
|
11 lat temu |
|
Samuel Thibault
|
163ef6b095
Make statistics only on the part of the trace between start_profiling and stop_profiling
|
11 lat temu |
|
Andra Hugo
|
aa59ce38c7
rollback
|
11 lat temu |
|
Andra Hugo
|
1c2f50cec5
fix exec_parallel code when called by the app thread
|
11 lat temu |
|
Andra Hugo
|
0c40ac6f8a
make the hypervisor point to the good includes
|
11 lat temu |
|
Andra Hugo
|
044a3e8e31
enable lots of cpus on mic native
|
11 lat temu |
 Nathalie Furmento
|
48b95f97ee
doc/doxygen/Makefile.am: add chapters/paje_draw_histogram.pdf to EXTRA_DIST
|
11 lat temu |
|
Samuel Thibault
|
e1e5feddca
Pass stability option to sort tool, not grep :)
|
11 lat temu |
|
Samuel Thibault
|
4e93f7f8cb
Keep order of events when they have the same time
|
11 lat temu |
|
Samuel Thibault
|
fee359d0fd
fix copyright
|
11 lat temu |
