 Nathalie Furmento
|
31f6ba2030
use STARPU_MAIN_RAM when appropriate
|
12 years ago |
|
Nathalie Furmento
|
3989f504bd
src: give name to internal tasks
|
12 years ago |
|
Nathalie Furmento
|
9bb5a09263
src: minor fixes
|
12 years ago |
 Andra Hugo
|
9bb7bdb07f
Increment/decrement submitted/ready tasks per context (instead of doing it per context & globally)
|
12 years ago |
|
Andra Hugo
|
5a22d83b5c
hypervisor: feft_lp: compute max by using the execution of the last sample instead of the idle time
|
12 years ago |
 Samuel Thibault
|
1ad22db615
port r11138 from 1.1: Replace global mutex by global rwlock
|
12 years ago |
|
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 years ago |
|
Andra Hugo
|
c4873d4813
hypervisor lp: max is computed using the time of executing tasks instead of idle time
|
12 years ago |
|
Samuel Thibault
|
4cae2beddb
port r11283 from 1.1: Prefetch data for tasks with execute_on_a_specific_worker too
|
12 years ago |
|
Andra Hugo
|
52459c8508
*eager: wake up workers when we add them to a context
|
12 years ago |
|
Nathalie Furmento
|
4d363ebb6b
src/core/sched_policy.c: do not display message scheduling policy not found when calling with STARPU_SCHED=help
|
12 years ago |
|
Andra Hugo
|
646c61a828
improve idle computations in order to better compute max_workers for the resizing lp
|
12 years ago |
 Pierre-André Wacrenier
|
2f7b61d57f
notify idleness at pop level in order to assign more pricesily idle time among contexts sharing a worker
|
12 years ago |
|
Andra Hugo
|
04e86519ef
uniform name for prologue
|
12 years ago |
|
Samuel Thibault
|
5f9af65fb6
Rather use the proper english writing, prologue. Prolog is really essentially a language :)
|
12 years ago |
|
Andra Hugo
|
f108a67609
add a prolog callback to a task (+ example)
|
12 years ago |
|
Andra Hugo
|
39cfbbcb7f
consider the flops of ready tasks when resizing and compute the speed for a sample of time too
|
12 years ago |
|
Samuel Thibault
|
5ac25e5658
fix build
|
12 years ago |
|
Andra Hugo
|
127e19b535
add iterating over all contexts at pop time when blocked_workers are alowed
|
12 years ago |
|
Andra Hugo
|
1c1a6aaddf
* add function to check if the workerid returned by the func starpu_worker_get_id belongs to a context,
|
12 years ago |
|
Andra Hugo
|
e4d5bcda38
remove the functions that force a worker to pop in a certain context, for the moment no need for that we do round robin
|
12 years ago |
 Thibaud Lambert
|
b2615f92ee
mic: merge trunk
|
12 years ago |
|
Thibaud Lambert
|
3b7abcc2cf
mic: Fix bugs
|
12 years ago |
|
Andra Hugo
|
c6ae0527ab
workers keep a chained list of contexts (instead of a sparse list) and remove as many loops as possible at pop time
|
12 years ago |
|
Samuel Thibault
|
cf2fcb6937
update to temanejo 1.0rc2
|
12 years ago |
|
Thibaud Lambert
|
e8e81e14ca
mic: merge trunk
|
12 years ago |
|
Thibaud Lambert
|
b17db603a8
mic: Correction nbmaxcombinedworkers + update _starpu_mic_sink_deinit
|
12 years ago |
|
Andra Hugo
|
5fa6ed1bb7
allow passing the pointer of a scheduler to a context
|
12 years ago |
|
Andra Hugo
|
70ee761ed3
correct speed computations for the hyp
|
12 years ago |
|
Andra Hugo
|
6c954c729c
workers must release the scheduling mutex before taking the role of hypervisor, update triggering issues for the other resizing strategies too
|
12 years ago |
