Regression in exec.c (print_section_info) ?

[Joel Brobecker]

> This works for me, e.g. with "info target" but perhaps there is a bigger
> picture and this line was removed for another reason.

I reviewed the history of the change, and I think it was an oversight.
The change that removed the newline was meant to get rid of some uses
of current_gdbarch, no more.

Could you test the attached? I think it's generally better for
internationalization to have one string rather than put together several
string blocks. I doubt it would make much difference in this case, but
might as well.

I would also like to see a new test if we don't already have one to
prevent this type of regression in the future. Please confirm that
you did run the patch against the testcase on at least one architecture.

One last thing: You also forgot to provide a ChangeLog entry. But *thank
you* for sending a unified diff - I just can't read context diffs!

-- 
Joel
-------------- next part --------------
Index: exec.c
===================================================================
RCS file: /cvs/src/src/gdb/exec.c,v
retrieving revision 1.73
diff -u -p -r1.73 exec.c
--- exec.c	19 Apr 2008 02:07:19 -0000	1.73
+++ exec.c	29 Apr 2008 03:00:23 -0000
@@ -543,10 +543,8 @@ print_section_info (struct target_ops *t
   wrap_here ("        ");
   printf_filtered (_("file type %s.\n"), bfd_get_target (abfd));
   if (abfd == exec_bfd)
-    {
-      printf_filtered (_("\tEntry point: "));
-      fputs_filtered (paddress (bfd_get_start_address (abfd)), gdb_stdout);
-    }
+    printf_filtered (_("\tEntry point: %s\n"),
+                     paddress (bfd_get_start_address (abfd)))
   for (p = t->to_sections; p < t->to_sections_end; p++)
     {
       printf_filtered ("\t%s", hex_string_custom (p->addr, wid));