Cannot build with IPASIR - cryptominisat

[jmony]

CBMC version: master and Release 5.12
Operating system: Linux (Debian and Ubuntu)
Exact command line resulting in the issue: make -C src IPASIR=/home/ubuntu/cryptominisat/src LIBS="/home/ubuntu/cryptominisat/build/lib/libipasircryptominisat5.a"
(cryptominisat built OK)
What behaviour did you expect: Successful build
What happened instead:

Entering solvers

make -C solvers
make[1]: Entering directory '/home/ubuntu/cbmc-cbmc-5.12/src/solvers'
g++ -c -DHAVE_IPASIR -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -MMD -MP -std=c++11 -O2 -DHAVE_IPASIR -I .. -I /home/ubuntu/cryptominisat/src -o sat/satcheck_ipasir.o sat/satcheck_ipasir.cpp
sat/satcheck_ipasir.cpp: In constructor ‘satcheck_ipasirt::satcheck_ipasirt()’:
sat/satcheck_ipasir.cpp:161:36: error: no matching function for call to ‘cnf_solvert::cnf_solvert()’
satcheck_ipasirt::satcheck_ipasirt()
^
In file included from sat/satcheck_ipasir.h:15:0,
from sat/satcheck_ipasir.cpp:20:
sat/cnf.h:72:12: note: candidate: cnf_solvert::cnf_solvert(message_handlert&)
explicit cnf_solvert(message_handlert &message_handler)
^~~~~~~~~~~
sat/cnf.h:72:12: note: candidate expects 1 argument, 0 provided
sat/cnf.h:69:7: note: candidate: cnf_solvert::cnf_solvert(const cnf_solvert&)
class cnf_solvert:public cnft
^~~~~~~~~~~
sat/cnf.h:69:7: note: candidate expects 1 argument, 0 provided
sat/cnf.h:69:7: note: candidate: cnf_solvert::cnf_solvert(cnf_solvert&&)
sat/cnf.h:69:7: note: candidate expects 1 argument, 0 provided
../common:222: recipe for target 'sat/satcheck_ipasir.o' failed
make[1]: *** [sat/satcheck_ipasir.o] Error 1
make[1]: Leaving directory '/home/ubuntu/cbmc-cbmc-5.12/src/solvers'
Makefile:97: recipe for target 'solvers.dir' failed
make: *** [solvers.dir] Error 2
make: Leaving directory '/home/ubuntu/cbmc-cbmc-5.12/src'

[jmony]

I forgot to mention that I edited config.inc to uncomment IPASIR (and only IPASIR).

[peterschrammel]

This might be fixed by #5305.

[jmony]

I tried with #5305 but it had no effect.

[tautschnig]

I actually messed this up in 89641a2. We'll need #5305 (or some variant thereof) plus an additional fix.