[Ipopt] Ipopt version 3.12.9

[Stefan Vigerske]

Hi,

you won't have to use coinbrew to use a current Ipopt.

But the problem you report also comes up with current Ipopt (mainly 
since it's not an Ipopt problem). It's the MUMPS source that isn't 
compatible with the Fortran language version that Gfortran 10 assumes.

Getting -fallow-argument-mismatch or -std=legacy into the Fortran 
compiler flags should work around this.

Stefan

On 6/4/20 5:55 PM, Brad Bell wrote:
> I am still using the ThridParty directory, instead of coinbrew, because 
> it works.
> (Similar to how the coin optimization suite still uses autotools to 
> install CppAD even though that feature has been deprecated for 10 years).
> I am getting a new error with a new version of gfortran. Would it be 
> fixed if I switched over to using coinbrew ?
> 
> Ipopt Version: 3.12.9
> gfortan Vection: 10.1.1-1
> 
> Error Message:
>   gfortran -I. -I../../../ThirdParty/Mumps 
> -I../../../ThirdParty/Mumps/MUMPS/src 
> -I../../../ThirdParty/Mumps/MUMPS/libseq 
> -I../../../ThirdParty/Mumps/MUMPS/include -g -O0 -pipe -Dmetis -c 
> ../../../ThirdParty/Mumps/MUMPS/src/dmumps_comm_buffer.F -o 
> dmumps_comm_buffer.o
> ../../../ThirdParty/Mumps/MUMPS/src/dmumps_comm_buffer.F:2667:23:
> 
>   2658 |         CALL MPI_PACK( WHAT, 1, MPI_INTEGER,
>        |                       2
> ......
>   2667 |         CALL MPI_PACK( LIST_SLAVES, NSLAVES, MPI_INTEGER,
>        |                       1
> Error: Rank mismatch between actual argument at (1) and actual argument 
> at (2) (scalar and rank-1)
> 
> 
> _______________________________________________
> Ipopt mailing list
> Ipopt at list.coin-or.org
> https://list.coin-or.org/mailman/listinfo/ipopt