ENH: use bundled MPI_Test when checking processor interface ready()
- fewer calls, potentially more consistent ENH: update sendRequest state after recvRequest wait - previously had this type of code: // Treat send as finished when recv is done UPstream::waitRequest(recvRequest_); recvRequest_ = -1; sendRequest_ = -1; Now refined as follows: // Require receive data. Update the send request state. UPstream::waitRequest(recvRequest_); recvRequest_ = -1; if (UPstream::finishedRequest(sendRequest_)) sendRequest_ = -1; Can potentially investigate with requiring both, but this may be over-contrained. Example, // Require receive data, but also wait for sends too UPstream::waitRequestPair(recvRequest_, sendRequest_);
Showing
- src/OpenFOAM/matrices/lduMatrix/lduAddressing/lduInterface/lduCalculatedProcessorField/lduCalculatedProcessorField.C 7 additions, 11 deletions...lduCalculatedProcessorField/lduCalculatedProcessorField.C
- src/OpenFOAM/matrices/lduMatrix/solvers/GAMG/interfaceFields/processorGAMGInterfaceField/processorGAMGInterfaceField.C 5 additions, 4 deletions...processorGAMGInterfaceField/processorGAMGInterfaceField.C
- src/finiteArea/fields/faPatchFields/constraint/processor/processorFaPatchField.C 18 additions, 21 deletions...aPatchFields/constraint/processor/processorFaPatchField.C
- src/finiteVolume/fields/fvPatchFields/constraint/calculatedProcessor/calculatedProcessorFvPatchField.C 15 additions, 18 deletions...int/calculatedProcessor/calculatedProcessorFvPatchField.C
- src/finiteVolume/fields/fvPatchFields/constraint/processor/processorFvPatchField.C 19 additions, 21 deletions...vPatchFields/constraint/processor/processorFvPatchField.C
- src/overset/lduPrimitiveProcessorInterface/GAMG/calculatedProcessorGAMGInterfaceField.C 5 additions, 4 deletions...sorInterface/GAMG/calculatedProcessorGAMGInterfaceField.C
Please register or sign in to comment