From mboxrd@z Thu Jan  1 00:00:00 1970
From: Andrew Cagney <ac131313@cygnus.com>
To: DJBARROW@de.ibm.com
Cc: gdb-patches@sourceware.cygnus.com, s390-patches@gnu.org, schwidefsky@de.ibm.com, ARENZ@de.ibm.com
Subject: Re: New gdb 31 & 64 bit patches for S/390
Date: Wed, 04 Jul 2001 21:02:00 -0000
Message-id: <3B43DB38.5040702@cygnus.com>
References: <C1256A00.006F3FCE.00@d12mta09.de.ibm.com>
X-SW-Source: 2001-07/msg00071.html

Hello,

I'm trying to understand the code:

+void
+_initialize_s390_tdep ()
+{
+  const bfd_arch_info_type *s390_arch_ptr =
+    bfd_lookup_arch (bfd_arch_s390, 0);
+
+#if GDB_MULTI_ARCH
+  /* Hook us into the gdbarch mechanism.  */
+  register_gdbarch_init (bfd_arch_s390, s390_gdbarch_init);
+#endif
+  if (!tm_print_insn)          /* Someone may have already set it */
+    tm_print_insn = gdb_print_insn_s390;
+  if (s390_arch_ptr)
+    tm_print_insn_info.mach = s390_arch_ptr->mach;
+  else
+    internal_error (__FILE__, __LINE__,
+                   "_initialize_s390_tdep: bfd_lookup_arch failed for ....
+}

but don't understand the reason for the call to bfd_lookup_arch().

printcmd.c:print_insn() will set the ->mach to the target architectures 
machine regardless of the above.

	Andrew