- Jan 15, 2019
-
-
Xavier Teruel authored
The service omp_get_thread_num() now returns -1 if thread is not part of the runtime yet; this query allows to determine either we need to call and admit_current_thread service or not; closes #1247
-
- Nov 02, 2018
-
-
Victor Lopez authored
We were unconditionally linking with libsqlite3 even if OpenCL was not configured. No other component uses sqlite3 so let's remove an unneeded dependency when possible.
-
- Oct 30, 2018
-
-
Jaume Bosch authored
-
- Oct 26, 2018
-
-
Jaume Bosch authored
The development was moved to a fork in the OmpSs-at-FPGA project. For more info visit: https://pm.bsc.es/ompss-at-fpga
-
- Oct 25, 2018
-
-
Victor Lopez authored
This partially reverts commit e5343f02. The fatal condition is too strict and is affecting several other scenarios.
-
- Oct 23, 2018
-
-
Victor Lopez authored
-
- Oct 22, 2018
-
-
Victor Lopez authored
The runtime should be able to create as many threads as available CPUs, or as requested by the user if CPUs are oversubscribed.
-
- Oct 09, 2018
-
-
Sergi Mateo Bellido authored
See #1246
-
- Aug 21, 2018
-
-
Victor Lopez authored
-
- Aug 17, 2018
-
-
Victor Lopez authored
Now the execution summary and the nanox-bindings utility print correctly the bindings mask when one thread is bound to multiple CPUs. Also, the string is now more human readable, as binding is grouped by thread if such is the case.
-
- Jul 10, 2018
-
-
Victor Lopez authored
-
Victor Lopez authored
Fix #1245
-
- Jun 22, 2018
-
-
Victor Lopez authored
Close #1244
-
- Jun 20, 2018
-
-
-
Jaume Bosch authored
-
- Jun 19, 2018
-
-
Jaume Bosch authored
Do not fail when the GPU or OpenCL plugin try to reserve an SMP processor but there are not available. Instead, the runtime will show a warning and share the core with another thread
-
- Jun 13, 2018
-
-
Victor Lopez authored
- Capture DLB_Init error code and disable DLB if needed - Add verbosity to some DLB errors
-
- Jun 04, 2018
-
-
Victor Lopez authored
This option limits the number of threads per core created in SMT machines, while keeping the same affinity mask by binding to multiple hwthreads inside the core.
-
Victor Lopez authored
-
Victor Lopez authored
-
Victor Lopez authored
-
Victor Lopez authored
-
Victor Lopez authored
-
- May 15, 2018
-
-
Victor Lopez authored
gcc-8 requires an alias function to have the same exact return type
-
- Apr 03, 2018
-
-
Victor Lopez authored
The main WD should be implicit to honor the implicit parallel of size 1 that executes the main program. As a side effect, the main WD in OmpSs can no longer ignore the barrier construct and must be explicitly prevented.
-
- Mar 09, 2018
-
-
Victor Lopez authored
Recent lock in lendCpu (5cd5ccbe) may cause deadlocks
-
- Mar 07, 2018
-
-
Victor Lopez authored
-
- Feb 23, 2018
-
-
Victor Lopez authored
-
Victor Lopez authored
Dynamic and Guided Worksharings WD submission now only creates a single WD representative instead of distributing the work. Idle threads will participate if needed, and will copy the WD at dequeue time.
-
- Feb 22, 2018
-
-
Jaume Bosch authored
-
Jaume Bosch authored
-
Jaume Bosch authored
Previously, the WD was assuming that the active device in the PE was the first one
-
Jaume Bosch authored
-
Jaume Bosch authored
Method WD::canRunIn(PE) moved to PE::canRun(WD) because it iterates over all PEdevices, it should be a PE method
-
Jaume Bosch authored
-
Jaume Bosch authored
-
Jaume Bosch authored
-
- Feb 09, 2018
-
-
Victor Lopez authored
-
- Feb 07, 2018
-
-
Victor Lopez authored
-
Victor Lopez authored
Close #1239
-