https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67219
--- Comment #4 from Thomas Koenig <tkoenig at gcc dot gnu.org> ---
The backtrace shows where this comes from:
#0 gfc_warning_now (opt=201, gmsgid=0x1437d10 "Change of value in conversion
from %qs to %qs at %L") at ../../trunk/gcc/fortran/error.c:1187
#1 0x000000000061a05d in gfc_int2real (src=0x1ea1de0, kind=4) at
../../trunk/gcc/fortran/arith.c:2079
#2 0x00000000006971b0 in match_sym_complex_part (result=0x7fffffffce60) at
../../trunk/gcc/fortran/primary.c:1286
#3 match_complex_part (result=0x7fffffffce60) at
../../trunk/gcc/fortran/primary.c:1311
#4 0x0000000000697262 in match_complex_constant
(result=result@entry=0x7fffffffd0f0) at ../../trunk/gcc/fortran/primary.c:1344
#5 0x0000000000697448 in gfc_match_literal_constant
(result=result@entry=0x7fffffffd0f0, signflag=signflag@entry=0)
at ../../trunk/gcc/fortran/primary.c:1448
#6 0x0000000000675c33 in match_primary (result=0x7fffffffd0f0) at
../../trunk/gcc/fortran/matchexp.c:149
#7 match_level_1 (result=<synthetic pointer>) at
../../trunk/gcc/fortran/matchexp.c:211
Seems like the gfc_warning_now isn't appropriate here. I will see what happens
when this is converted to gfc_warning.
