 Samuel Thibault
|
a93bf83bb4
Do not change the new replicate fields, this is handled by the caller. Fix the cache allocation to actually reuse old buffers.
|
12 년 전 |
|
Samuel Thibault
|
e1bdf3cd44
merge changelog change
|
12 년 전 |
|
Samuel Thibault
|
1f1954c43b
Add OpenCL kernel submission pipelining, to overlap costs
|
12 년 전 |
|
Samuel Thibault
|
f99ef13c88
Add CUDA kernel submission pipelining, to overlap costs and allow concurrent
|
12 년 전 |
|
Samuel Thibault
|
aa5c3e5c50
Let interfaces declare which transfers they allow with the can_copy methode.
|
12 년 전 |
|
Samuel Thibault
|
89ece6b14f
changelog
|
12 년 전 |
|
Samuel Thibault
|
2a1bb7bf49
Update 1.1.2 svn revision
|
12 년 전 |
|
Samuel Thibault
|
14b8f531a6
Lock performance model files while writing and reading them to avoid issues on parallel launches, MPI runs notably.
|
12 년 전 |
|
Samuel Thibault
|
f43bf86292
Fix complexity of implicit task/data dependency, from quadratic to linear
|
12 년 전 |
|
Samuel Thibault
|
3db86a2ede
* Add STARPU_VARIABLE_NBUFFERS to be set in cl.nbuffers, and nbuffers and
|
12 년 전 |
 Nathalie Furmento
|
53a11e7402
workers: new function starpu_worker_get_type_as_string()
|
12 년 전 |
|
Samuel Thibault
|
3b7a4ec8f9
Mention MIC too
|
12 년 전 |
|
Samuel Thibault
|
3f9b8cd37a
Add the starpu_paje_summary statistics tool.
|
12 년 전 |
|
Samuel Thibault
|
214c709b08
port r 13033 from 1.1: Add changelog
|
12 년 전 |
|
Samuel Thibault
|
8d8e0903d0
port r12948 from 1.1: Add changelog entry
|
12 년 전 |
|
Samuel Thibault
|
7dd6b42215
Document that blocking spinlocks are only available on Linux
|
12 년 전 |
|
Samuel Thibault
|
039524c7cf
Spinlocks now block after a hundred tries. This avoids typical 10ms pauses
|
12 년 전 |
|
Samuel Thibault
|
0c9c7c0ba3
advertise lws
|
12 년 전 |
|
Samuel Thibault
|
4705435be8
Add STARPU_TAG_ONLY, to specify a tag for traces without making StarPU manage the tag.
|
12 년 전 |
|
Samuel Thibault
|
778b4c444a
* Add codelet size, footprint and tag id in the paje trace.
|
12 년 전 |
|
Nathalie Furmento
|
b43914770b
ChangeLog: move updates to the correct section, some updates were wrongly listed as part of release 1.1.x instead of 1.2.x
|
12 년 전 |
|
Samuel Thibault
|
1d8e85f121
port r12828 from 1.1: Add changelog for STARPU_CALIBRATE_MINIMUM
|
12 년 전 |
|
Samuel Thibault
|
d592415088
port r12811 from 1.1: Add changelog for the "scheduling" state in traces
|
12 년 전 |
|
Nathalie Furmento
|
06778e0bb6
sched_ctx: new function starpu_sched_ctx_display_workers() to display the worker names of a context
|
12 년 전 |
|
Nathalie Furmento
|
90879cd44c
configure: allow to set --enable-verbose=extra to increase debug messages verbosity
|
12 년 전 |
|
Samuel Thibault
|
f255d2c95a
port r12703 from 1.1: Allow data to be uninitialized but the application provide an init_cl to initialize it
|
12 년 전 |
|
Nathalie Furmento
|
cdb8ad6182
ChangeLog: update
|
12 년 전 |
|
Nathalie Furmento
|
68d0971b70
starpufft: By default, StarPU FFT examples are not compiled and checked, the configure option --enable-starpufft-examples needs to be specified to change this behaviour.
|
12 년 전 |
|
Samuel Thibault
|
f2b1d0f595
port r12584 from 1.1: Re-enable asynchronous transfers from CPUs, now that the issue has been found to be in application code
|
12 년 전 |
|
Samuel Thibault
|
092f322b1c
Add CUDA concurrent kernel execution support through the STARPU_NWORKER_PER_CUDA environment variable.
|
12 년 전 |
