StarPU

StarPU Nightly Tests

Latest nightly tarball

The latest nightly tarball successfully passing 'make distcheck' is available at starpu-nightly-latest.tar.gz (produced on 2018-11-14).

The coverage report is available as HTML or as TXT.

The StarPU documentation is available in PDF and in HTML.

Tests

Date Revision Result Comments Log Coverage HTML Coverage Text
2018-11-14 e7e99e55acf0ccb107ef6a21a8c933ff08849901 success -SKIP:openmp/init_exit_01-openmp/init_exit_02-openmp/environment-openmp/api_01-openmp/parallel_01-openmp/parallel_02-openmp/parallel_03-openmp/parallel_barrier_01-openmp/parallel_master_01-openmp/parallel_master_inline_01-openmp/parallel_single_wait_01-openmp/parallel_single_nowait_01-openmp/parallel_single_inline_01-openmp/parallel_single_copyprivate_01-openmp/parallel_single_copyprivate_inline_01-openmp/parallel_critical_01-openmp/parallel_critical_inline_01-openmp/parallel_critical_named_01-openmp/parallel_critical_named_inline_01-openmp/parallel_simple_lock_01-openmp/parallel_nested_lock_01-openmp/parallel_for_01-openmp/parallel_for_02-openmp/parallel_for_ordered_01-openmp/parallel_sections_01-openmp/parallel_sections_combined_01-openmp/task_01-openmp/task_02-openmp/task_03-openmp/taskwait_01-openmp/taskgroup_01-openmp/taskgroup_02-openmp/array_slice_01-openmp/cuda_task_01-datawizard/readonly-datawizard/locality.sh-stencil/stencil5_lb-cpp/add_vectors_interface-sched_ctx/gpu_partition-matmul/matmul- log link link
2018-11-12 51600bb636f0e7069e7d3796ddf295d038dd8095 success -SKIP:openmp/init_exit_01-openmp/init_exit_02-openmp/environment-openmp/api_01-openmp/parallel_01-openmp/parallel_02-openmp/parallel_03-openmp/parallel_barrier_01-openmp/parallel_master_01-openmp/parallel_master_inline_01-openmp/parallel_single_wait_01-openmp/parallel_single_nowait_01-openmp/parallel_single_inline_01-openmp/parallel_single_copyprivate_01-openmp/parallel_single_copyprivate_inline_01-openmp/parallel_critical_01-openmp/parallel_critical_inline_01-openmp/parallel_critical_named_01-openmp/parallel_critical_named_inline_01-openmp/parallel_simple_lock_01-openmp/parallel_nested_lock_01-openmp/parallel_for_01-openmp/parallel_for_02-openmp/parallel_for_ordered_01-openmp/parallel_sections_01-openmp/parallel_sections_combined_01-openmp/task_01-openmp/task_02-openmp/task_03-openmp/taskwait_01-openmp/taskgroup_01-openmp/taskgroup_02-openmp/array_slice_01-openmp/cuda_task_01-datawizard/readonly-datawizard/locality.sh-stencil/stencil5_lb-cpp/add_vectors_interface-sched_ctx/gpu_partition-matmul/matmul-openmp/init_exit_01-openmp/init_exit_02-openmp/environment-openmp/api_01-openmp/parallel_01-openmp/parallel_02-openmp/parallel_03-openmp/parallel_barrier_01-openmp/parallel_master_01-openmp/parallel_master_inline_01-openmp/parallel_single_wait_01-openmp/parallel_single_nowait_01-openmp/parallel_single_inline_01-openmp/parallel_single_copyprivate_01-openmp/parallel_single_copyprivate_inline_01-openmp/parallel_critical_01-openmp/parallel_critical_inline_01-openmp/parallel_critical_named_01-openmp/parallel_critical_named_inline_01-openmp/parallel_simple_lock_01-openmp/parallel_nested_lock_01-openmp/parallel_for_01-openmp/parallel_for_02-openmp/parallel_for_ordered_01-openmp/parallel_sections_01-openmp/parallel_sections_combined_01-openmp/task_01-openmp/task_02-openmp/task_03-openmp/taskwait_01-openmp/taskgroup_01-openmp/taskgroup_02-openmp/array_slice_01-openmp/cuda_task_01-datawizard/readonly-datawizard/locality.sh-stencil/stencil5_lb-cpp/add_vectors_interface-sched_ctx/gpu_partition-matmul/matmul- log link link
2018-11-10 51600bb636f0e7069e7d3796ddf295d038dd8095 success -SKIP:openmp/init_exit_01-openmp/init_exit_02-openmp/environment-openmp/api_01-openmp/parallel_01-openmp/parallel_02-openmp/parallel_03-openmp/parallel_barrier_01-openmp/parallel_master_01-openmp/parallel_master_inline_01-openmp/parallel_single_wait_01-openmp/parallel_single_nowait_01-openmp/parallel_single_inline_01-openmp/parallel_single_copyprivate_01-openmp/parallel_single_copyprivate_inline_01-openmp/parallel_critical_01-openmp/parallel_critical_inline_01-openmp/parallel_critical_named_01-openmp/parallel_critical_named_inline_01-openmp/parallel_simple_lock_01-openmp/parallel_nested_lock_01-openmp/parallel_for_01-openmp/parallel_for_02-openmp/parallel_for_ordered_01-openmp/parallel_sections_01-openmp/parallel_sections_combined_01-openmp/task_01-openmp/task_02-openmp/task_03-openmp/taskwait_01-openmp/taskgroup_01-openmp/taskgroup_02-openmp/array_slice_01-openmp/cuda_task_01-datawizard/readonly-datawizard/locality.sh-stencil/stencil5_lb-cpp/add_vectors_interface-sched_ctx/gpu_partition-matmul/matmul- log link link
2018-11-08 d2a33ce51d9e54c6cb67ea4bb6cfc0107c3bc06d success -SKIP:openmp/init_exit_01-openmp/init_exit_02-openmp/environment-openmp/api_01-openmp/parallel_01-openmp/parallel_02-openmp/parallel_03-openmp/parallel_barrier_01-openmp/parallel_master_01-openmp/parallel_master_inline_01-openmp/parallel_single_wait_01-openmp/parallel_single_nowait_01-openmp/parallel_single_inline_01-openmp/parallel_single_copyprivate_01-openmp/parallel_single_copyprivate_inline_01-openmp/parallel_critical_01-openmp/parallel_critical_inline_01-openmp/parallel_critical_named_01-openmp/parallel_critical_named_inline_01-openmp/parallel_simple_lock_01-openmp/parallel_nested_lock_01-openmp/parallel_for_01-openmp/parallel_for_02-openmp/parallel_for_ordered_01-openmp/parallel_sections_01-openmp/parallel_sections_combined_01-openmp/task_01-openmp/task_02-openmp/task_03-openmp/taskwait_01-openmp/taskgroup_01-openmp/taskgroup_02-openmp/array_slice_01-openmp/cuda_task_01-datawizard/readonly-datawizard/locality.sh-stencil/stencil5_lb-cpp/add_vectors_interface-sched_ctx/gpu_partition-matmul/matmul- log link link
2018-11-06 34cb952bf0ecc0147e8feb81a219950f8dce0212 success -SKIP:openmp/init_exit_01-openmp/init_exit_02-openmp/environment-openmp/api_01-openmp/parallel_01-openmp/parallel_02-openmp/parallel_03-openmp/parallel_barrier_01-openmp/parallel_master_01-openmp/parallel_master_inline_01-openmp/parallel_single_wait_01-openmp/parallel_single_nowait_01-openmp/parallel_single_inline_01-openmp/parallel_single_copyprivate_01-openmp/parallel_single_copyprivate_inline_01-openmp/parallel_critical_01-openmp/parallel_critical_inline_01-openmp/parallel_critical_named_01-openmp/parallel_critical_named_inline_01-openmp/parallel_simple_lock_01-openmp/parallel_nested_lock_01-openmp/parallel_for_01-openmp/parallel_for_02-openmp/parallel_for_ordered_01-openmp/parallel_sections_01-openmp/parallel_sections_combined_01-openmp/task_01-openmp/task_02-openmp/task_03-openmp/taskwait_01-openmp/taskgroup_01-openmp/taskgroup_02-openmp/array_slice_01-openmp/cuda_task_01-datawizard/readonly-datawizard/locality.sh-stencil/stencil5_lb-cpp/add_vectors_interface-sched_ctx/gpu_partition-matmul/matmul- log link link
2018-11-05 cd50843aa7a7b59ba751f81e30b80f5642c991eb failure - log - -
2018-11-03 1b5ba4ea8da90ee8a7beafd47d6e6353de3873b1 failure - log - -
2018-11-01 cd6764e1eb7c6fbd502e64346beccc6f0c817f07 failure - log - -
2018-10-31 c987eaf4490c06d8c2c9f043801f789d80d080f3 failure - log - -
2018-10-30 f3828f41d5c7de3747fd99bd3bd8bd5121cea9f1 failure - log - -
See also the tests archive.

Micro-benchmarks

Raw data

The black line at revision 11490 marks when we changed the system where the microbenchs are run.

The purle line at revision 12298.5 marks when the default scheduler was switched to use tree-based worker iterators, and the green line at revision 17026.5 marks when this was reverted.

The black line at revision 19182.5 marks when we changed again the system where the microbenchs are run.

The black lines at revisions 19320.5 and 19451.5 mark the period when the kernel was switched to 4.8.11 instead of the usual Debian 4.5.0

Tasks Overhead

This is the time to submit a task, from the main thread:

This is the time to execute an empty tasks:

This is the total time to submit & execute an empty tasks:

Synchronous Tasks Overhead

This is the total time to submit & execute a synchronous task:

Asynchronous Tasks Overhead

This is the total time to submit & execute an asynchronous task without dependencies:

Tasks size Overhead

This shows the speedup of running small tasks sizes on 60 cores of a 64-core machine. The highest curve (in blue) is for 4096µs tasks, the next curve (in green) is for 2048µs tasks, the next curve (in purple) is for 1024µs tasks, etc.


eager

modular-eager-prefetching

modular-eager

prio

modular-prio-prefetching

modular-prio

ws

modular-ws

lws

graph_test

dm

dmda

dmdar

dmdas

modular-heft

modular-heft-prio

modular-heft2

dmdasd

heteroprio

random

modular-random

modular-random-prefetching

modular-random-prio

modular-random-prio-prefetching

peager

pheft
eager
ws
heft
random
misc

Registering a Matrix as a Vector

Last updated on 2018/11/14 at 04:58.