Author | Changes | Lines of Code | Lines per Change |
---|---|---|---|
Totals | 16 (100.0%) | 60 (100.0%) | 3.7 |
qymeng | 7 (43.8%) | 19 (31.7%) | 2.7 |
jthornoc | 2 (12.5%) | 18 (30.0%) | 9.0 |
ahumphrey | 5 (31.3%) | 18 (30.0%) | 3.6 |
jsutherland | 2 (12.5%) | 5 (8.3%) | 2.5 |
This one has been a long time coming.
Rip out the standard namespace opening that has been polluting header files throughout the framework. This crept into all sorts of places.
DO NOT open namespaces in header files. Ever.
5 lines of code changed in 2 files:
use d_allmpi to check if running under MPI
0 lines of code changed in 2 files:
Do not call MPI_Comm_dup if not running under MPI/threads.
1 lines of code changed in 1 file:
Allocate MPI communicators based on taskgraph.
This fixed a hung when there are large number of reduction tasks in the taskgraph.
17 lines of code changed in 3 files:
Add some extra offerings for sensible multi-threaded output.
2 lines of code changed in 2 files:
Fix for subtle logic error that was causing sus not to exit correctly
when an error occurred during UPS validation. The re-'throw'-ing of the
exception needs to occur on all processors, not just proc 0 where the
output is displayed.
Also, fixed catching of the case when a UPS_SPEC file is not found.
18 lines of code changed in 2 files:
#define proc0thread0cout.
This is to further reduce excess spew on large parallel runs when using the threaded scheduler.
7 lines of code changed in 1 file:
Beginnings of generalized "device" tasks and work queues.
This will ultimately provide a unified approach to both GPU and MIC tasks.
9 lines of code changed in 2 files:
rename coarsenDelt task name to reduceSysVar
call MPI reduce directly from reduceSysVar task
1 lines of code changed in 1 file: