Segfault when using %stdin

Steps to reproduce:

1. Run 'gs %stdin'

Expected result: Ghostscript waits to receive input on stdin

Actual result: Ghostscript segfaults

Here's the stacktrace from GDB:

Program received signal SIGSEGV, Segmentation fault.
stdin_open (iodev=0x62f100, access=0x7ffff7503d99 "r", ps=0x7fffffffbe88, mem=0x604778) at ./psi/ziodevsc.c:123
123 ./psi/ziodevsc.c: No such file or directory.
(gdb) bt
#0 stdin_open (iodev=0x62f100, access=0x7ffff7503d99 "r", ps=0x7fffffffbe88, mem=0x604778) at ./psi/ziodevsc.c:123
#1 0x00007ffff71e913b in zlibfile (i_ctx_p=i_ctx_p@entry=0x655d48) at ./psi/zfile.c:603
#2 0x00007ffff71f1a48 in specific_vm_op (i_ctx_p=0x655d48, opproc=0x7ffff71e90a0 <zlibfile>, space=<optimized out>) at ./psi/zsysvm.c:40
#3 0x00007ffff71d7ee2 in interp (pi_ctx_p=pi_ctx_p@entry=0x6033f0, pref=<optimized out>, perror_object=perror_object@entry=0x7fffffffd010) at ./psi/interp.c:1561
#4 0x00007ffff71d8999 in gs_call_interp (perror_object=0x7fffffffd010, pexit_code=0x7fffffffd00c, user_errors=1, pref=0x7fffffffceb0, pi_ctx_p=0x603490) at ./psi/interp.c:510
#5 gs_interpret (pi_ctx_p=pi_ctx_p@entry=0x6033f0, pref=pref@entry=0x7fffffffcfe0, user_errors=1, user_errors@entry=-12272, pexit_code=0x7fffffffd00c, perror_object=0x7fffffffd010) at ./psi/interp.c:468
#6 0x00007ffff71cdd49 in gs_main_interpret (perror_object=<optimized out>, pexit_code=<optimized out>, user_errors=-12272, pref=0x7fffffffcfe0, minst=<optimized out>) at ./psi/imain.c:241
#7 gs_main_run_string_end (minst=minst@entry=0x603350, user_errors=<optimized out>, pexit_code=pexit_code@entry=0x7fffffffd00c, perror_object=perror_object@entry=0x7fffffffd010) at ./psi/imain.c:621
#8 0x00007ffff71cddce in gs_main_run_string_with_length (minst=minst@entry=0x603350, str=str@entry=0x976c50 "<25737464696e>.runfile", length=<optimized out>, user_errors=<optimized out>, pexit_code=pexit_code@entry=0x7fffffffd00c, perror_object=perror_object@entry=0x7fffffffd010) at ./psi/imain.c:579
#9 0x00007ffff71cde0d in gs_main_run_string (minst=minst@entry=0x603350, str=str@entry=0x976c50 "<25737464696e>.runfile", user_errors=<optimized out>, pexit_code=pexit_code@entry=0x7fffffffd00c, perror_object=perror_object@entry=0x7fffffffd010) at ./psi/imain.c:561
#10 0x00007ffff71cf219 in run_string (minst=minst@entry=0x603350, str=str@entry=0x976c50 "<25737464696e>.runfile", options=options@entry=3) at ./psi/imainarg.c:897
#11 0x00007ffff71cf398 in runarg (minst=0x603350, pre=0x7ffff7506f7d "", arg=<optimized out>, post=0x7ffff747e40d ".runfile", options=3) at ./psi/imainarg.c:887
#12 0x00007ffff71d086b in gs_main_init_with_args (minst=0x603350, argc=2, argv=0x7fffffffdbf8) at ./psi/imainarg.c:233
#13 0x00000000004009bb in main (argc=2, argv=0x7fffffffdbf8) at ./psi/dxmainc.c:86

Don't know if this is upstream bug, I haven't build a more recent version to check.