 Samuel Thibault
|
84e795a1db
fix memleaks
|
8 년 전 |
 Nathalie Furmento
|
514a24fefe
sc_hypervisor: fix cppcheck warnings
|
8 년 전 |
|
Nathalie Furmento
|
3cdc863cb4
fix bugs detected by coverity
|
9 년 전 |
 Samuel Pitoiset
|
92fff34d63
hypervisor: fix wrong use of printf() in debug mode
|
9 년 전 |
 Andra Hugo
|
ba3e4d7403
small fixes done while writing my thesis + task window at context level (before pushing tasks to the scheduler) expressed w.r.t. to time and not number of tasks
|
10 년 전 |
|
Samuel Thibault
|
717fbf7ab1
Always call init_iterator, it does not make sense otherwise
|
10 년 전 |
|
Andra Hugo
|
5f388d2a95
hierarchical contexts - bug fixing and DEBUG macro
|
11 년 전 |
|
Andra Hugo
|
491a6c6124
fix hierarchical resizing of contexts
|
11 년 전 |
|
Andra Hugo
|
60d44eb737
resize hierarchic ctxs hierarchically :)
|
11 년 전 |
|
Nathalie Furmento
|
26110479f9
rename starpu_get_nready_flops_of_sched_ctx to starpu_sched_ctx_get_nready_flops
|
11 년 전 |
|
Nathalie Furmento
|
56986e683e
rename starpu_get_nready_tasks_of_sched_ctx to starpu_sched_ctx_get_nready_tasks
|
11 년 전 |
|
Andra Hugo
|
12dec4c5cc
hypervisor optimisation (avoid unnecessary notifications/locks/trylocks) + add the hyp the traces
|
11 년 전 |
|
Andra Hugo
|
5a22d83b5c
hypervisor: feft_lp: compute max by using the execution of the last sample instead of the idle time
|
11 년 전 |
|
Andra Hugo
|
994728b26b
hypervisor lp: max computation for the contexts sharing resources ( do not consider shared workers as full workers)
|
11 년 전 |
|
Andra Hugo
|
1289532233
whenever there are a few resources the lp cannot assign, all contexts share these resources
|
11 년 전 |
|
Andra Hugo
|
082dc8501c
hypervisor: adapt lp solution when max has an important impact on the solution (exec lp twice for now, next co share resources)
|
11 년 전 |
|
Andra Hugo
|
52459c8508
*eager: wake up workers when we add them to a context
|
11 년 전 |
|
Andra Hugo
|
9502f53f25
introduce min max as a constraint for the linear program
|
11 년 전 |
|
Andra Hugo
|
9f20f53ce9
fix size_ctxs (do not consider ready tasks for size, only for resize), speed is computed on fixed value of the time
|
11 년 전 |
|
Andra Hugo
|
f108a67609
add a prolog callback to a task (+ example)
|
11 년 전 |
|
Andra Hugo
|
39cfbbcb7f
consider the flops of ready tasks when resizing and compute the speed for a sample of time too
|
11 년 전 |
|
Andra Hugo
|
1c1a6aaddf
* add function to check if the workerid returned by the func starpu_worker_get_id belongs to a context,
|
11 년 전 |
|
Andra Hugo
|
339104ed55
for feft_lp distribute resources at each resize instead of moving idle workers from one ctx to another
|
12 년 전 |
|
Andra Hugo
|
c365c9f927
bug fixing: when a ctx has a single worker left it does not give it away, it keeps it and it shares it with a another ctx
|
12 년 전 |
|
Andra Hugo
|
109eac0300
remove contraint for tmax in lp
|
12 년 전 |
|
Andra Hugo
|
abebb0c602
cleaning ispeed_lp code + free hwloc_worker_set when deleting the context
|
12 년 전 |
|
Andra Hugo
|
f40ece4bb1
share at least a resource when the hypervisor allows no resources to a context
|
12 년 전 |
|
Andra Hugo
|
e66dd53eab
cleaning + better managing the types of workers
|
12 년 전 |
|
Andra Hugo
|
b0b9e3f308
small fix to resize ctxs when only cpus are available
|
12 년 전 |
|
Andra Hugo
|
054afc9556
allow calling the resizing linear programs from the appl
|
12 년 전 |
