dump-guest-memory: more descriptive lookup_type failure

We've seen a few reports of

 (gdb) source /usr/share/qemu-kvm/dump-guest-memory.py
 Traceback (most recent call last):
   File "/usr/share/qemu-kvm/dump-guest-memory.py", line 19, in <module>
     UINTPTR_T = gdb.lookup_type("uintptr_t")
 gdb.error: No type named uintptr_t.

This occurs when symbols haven't been loaded first, i.e. neither a
QEMU binary was loaded nor a QEMU process was attached first. Let's
better inform the user of how to fix the issue themselves in order
to avoid more reports.
Acked-by: NJanosch Frank <frankja@linux.vnet.ibm.com>
Signed-off-by: NAndrew Jones <drjones@redhat.com>
Message-Id: <20180314153820.18426-1-drjones@redhat.com>
Reviewed-by: NFam Zheng <famz@redhat.com>
Tested-by: NFam Zheng <famz@redhat.com>
Reviewed-by: NLaszlo Ersek <lersek@redhat.com>
Reviewed-by: NMarc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: NMarc-André Lureau <marcandre.lureau@redhat.com>

scripts/dump-guest-memory.py

@@ -16,7 +16,12 @@ the COPYING file in the top-level directory.
 import ctypes
 import struct
 
-UINTPTR_T = gdb.lookup_type("uintptr_t")
+try:
+    UINTPTR_T = gdb.lookup_type("uintptr_t")
+except Exception as inst:
+    raise gdb.GdbError("Symbols must be loaded prior to sourcing dump-guest-memory.\n"
+                       "Symbols may be loaded by 'attach'ing a QEMU process id or by "
+                       "'load'ing a QEMU binary.")
 
 TARGET_PAGE_SIZE = 0x1000
 TARGET_PAGE_MASK = 0xFFFFFFFFFFFFF000