This is on ia64, where kind=10 is padded to 16 bytes.
I suspect it will be the same on i686 with the same padding.
$ cat compl.f
program main
complex (kind=10) a, b
real(kind=10) rr,ri
a = (3.2_10, -2.1_10)
open (10,form="unformatted",status="replace")
write(10) a
rewind(10)
read(10) b
print *,b
rewind(10)
read(10) rr,ri
print *,rr,ri
end
$ gfortran compl.f
$ ./a.out
( 3.20000000000000 , 2.132901171296497E+2959)
3.20000000000000 2.132901171296497E+2959
$ gfortran -v
Using built-in specs.
Target: ia64-unknown-linux-gnu
Configured with: ../gcc-4.1-20050806/configure --prefix=/home/zfkts --enable-
languages=c,f95
Thread model: posix
gcc version 4.1.0 20050806 (experimental)
$ ls -l fort.10
-rw-r--r-- 1 zfkts hvt 36 Aug 16 13:49 fort.10
--
Summary: unformatted complex I/O with kind=10
Product: gcc
Version: 4.1.0
Status: UNCONFIRMED
Keywords: wrong-code
Severity: normal
Priority: P2
Component: libfortran
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: tkoenig at gcc dot gnu dot org
CC: gcc-bugs at gcc dot gnu dot org
GCC target triplet: ia64-unknown-linux-gnu
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23419
