 Samuel Thibault
|
7c3516dd73
add missing initialization
|
12 年 前 |
|
Samuel Thibault
|
d9bfb2bacc
port r11334 from 1.1: Prefetching is now done for all schedulers when it can be done whatever the scheduling decision.
|
12 年 前 |
|
Samuel Thibault
|
e2dd2f3d8f
port r11332 from 1.1: Fix test against batched allocation
|
12 年 前 |
|
Samuel Thibault
|
0eff9a928f
print number of pending requests
|
12 年 前 |
|
Samuel Thibault
|
98cc8a4d3e
Add starpu-print-datas-summary contributed by Cyril Bordage
|
12 年 前 |
|
Samuel Thibault
|
7f56eab83f
port r11326 from 1.1: fix typo
|
12 年 前 |
 Nathalie Furmento
|
b7a3574823
mic-configure: add make target 'showcheck'
|
12 年 前 |
|
Samuel Thibault
|
84f45afab5
port r11322 from 1.1: fix prio scheduler termination
|
12 年 前 |
|
Samuel Thibault
|
336612abd1
port r11320 from 1.1: Do not error out if the user didn't release memory allocated with starpu_malloc_on_node
|
12 年 前 |
|
Samuel Thibault
|
4bb3fc4b7b
port r11318 from 1.1: Simply use the root object total memory, like in the commented code
|
12 年 前 |
|
Samuel Thibault
|
c58de18431
port r11316 from 1.1: Do not limit ourself to NUMA node 0, we still allocate on the whole machine ATM.
|
12 年 前 |
|
Samuel Thibault
|
9529be1e71
port r11314 from 1.1: Add missing include
|
12 年 前 |
|
Samuel Thibault
|
0e385974a9
port r11312 from 1.1: Small CUDA allocations (<= 4MiB) are now batched to avoid the huge cudaMalloc overhead.
|
12 年 前 |
|
Samuel Thibault
|
970c99c558
port r11310 from 1.1: Fix looking through a list while removing elements
|
12 年 前 |
|
Samuel Thibault
|
e1501b5592
port r11308 from 1.1: record allocation size in trace
|
12 年 前 |
|
Samuel Thibault
|
f850d2f26a
port r11305 from 1.1: the handle interfaces are already freed at this point, use the footprint pre-computed in the mc
|
12 年 前 |
|
Nathalie Furmento
|
4bdac6b139
configure.ac: add option --coverage to CXXFLAGS (to avoid error error 'Hidden symbol atexit is referenced by DSO' when using libtool with gcov)
|
12 年 前 |
|
Samuel Thibault
|
c4dc808530
port r11299 from 1.1: Use a hash table to speed up allocation cache reuse by hashing through the data footprint
|
12 年 前 |
|
Samuel Thibault
|
01aa31bde8
port r11297 from 1.1: Avoid looking through all data for not_important data if the application did not set any data as not_important.
|
12 年 前 |
 Andra Hugo
|
c4873d4813
hypervisor lp: max is computed using the time of executing tasks instead of idle time
|
12 年 前 |
|
Samuel Thibault
|
0bb629fed8
Avoid reverting the ordering of cache items
|
12 年 前 |
|
Nathalie Furmento
|
689a54fffe
fix doc and function prototypes
|
12 年 前 |
|
Nathalie Furmento
|
8d57290fd9
tests/main/starpu_init.c: use setenv instead of putenv, this is more valgrind friendly
|
12 年 前 |
|
Nathalie Furmento
|
e8a8200fc9
doc/doxygen/chapters/code/disk_copy.c: use setenv instead of putenv, this is more valgrind friendly
|
12 年 前 |
|
Nathalie Furmento
|
a2dd3ee669
tests/disk/disk_copy.c: use setenv instead of putenv, this is more valgrind friendly
|
12 年 前 |
|
Nathalie Furmento
|
a7383bc086
mpi/tests/insert_task_cache.c: use setenv instead of putenv, this is more valgrind friendly
|
12 年 前 |
|
Samuel Thibault
|
e9fbf7551e
Add more CUDA path candidates
|
12 年 前 |
|
Samuel Thibault
|
4cae2beddb
port r11283 from 1.1: Prefetch data for tasks with execute_on_a_specific_worker too
|
12 年 前 |
|
Nathalie Furmento
|
6f8c409637
tests/Makefile.am: have make clean delete core file (which is produced by expected to fail errorcheck/invalid_blocking_calls)
|
12 年 前 |
|
Nathalie Furmento
|
da74baa0b0
doc/doxygen/refman.tex: remove commands which are no longer valid with recent versions of doxygen (thanks again to Cyril Bordage for the BR)
|
12 年 前 |
