ICODE-ADC/lammps
4
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

avoid overflow when computing memory address offsets on 64-bit machines

Browse Source
This commit is contained in:
Axel Kohlmeyer
2021-02-05 19:42:07 -05:00
parent e223ea3784
commit 12f49c4c89
4 changed files with 12 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
src/irregular.cpp
View File

@ -930,11 +930,11 @@ void Irregular::exchange_data(char *sendbuf, int nbytes, char *recvbuf)
// post all receives, starting after self copies
bigint offset = num_self*(bigint)nbytes;
bigint offset = (bigint)num_self*(bigint)nbytes;
for (int irecv = 0; irecv < nrecv_proc; irecv++) {
MPI_Irecv(&recvbuf[offset],num_recv[irecv]*nbytes,MPI_CHAR,
proc_recv[irecv],0,world,&request[irecv]);
offset += num_recv[irecv]*nbytes;
offset += (bigint)num_recv[irecv]*nbytes;
}
// reallocate buf for largest send if necessary