This is the mail archive of the
gdb-patches@sourceware.cygnus.com
mailing list for the GDB project.
Re: [RFA] Don't talk about ptrace when there's none
> Date: Sun, 9 Apr 2000 09:58:34 -0400 (EDT)
> From: Eli Zaretskii <eliz@delorie.com>
>
> So what do people prefer, eventually? I could:
>
> (a) install the changes posted by Elena, and have the error message
> mention ptrace only if HAVE_PTRACE is defined;
>
> or
>
> (b) change the wording of the message so that it doesn't mention
> ptrace at all, as suggested by Michael.
>
> I like the second alternative better, especially since, as I now
> understand, ptrace is unused on many platforms.
>
> Comments?
>
> I'm in favor for the second aternative too, especially since the Hurd
> does have ptrace that is only sporadically used in GDB. It isn't used
> for inserting breakpoints, so mentioning ptrace in the error message
> would be misleading for the Hurd.
Okay, then how about the following patch? Good enough to commit?
2000-04-11 Eli Zaretskii <eliz@is.elta.co.il>
* infrun.c (proceed, normal_stop): Change the error message about
failure to insert breakpoints/watchpoints so that it makes sense
even if ptrace is not used or no other processes can be active.
--- gdb/infrun.c~1 Sat Apr 1 19:41:12 2000
+++ gdb/infrun.c Tue Apr 11 19:37:02 2000
@@ -1044,9 +1044,11 @@ proceed (CORE_ADDR addr, enum target_sig
int temp = insert_breakpoints ();
if (temp)
{
- print_sys_errmsg ("ptrace", temp);
+ print_sys_errmsg ("insert_breakpoints", temp);
error ("Cannot insert breakpoints.\n\
-The same program may be running in another process.");
+The same program may be running in another process,\n\
+or you may have requested too many hardware\n\
+breakpoints and/or watchpoints.\n");
}
breakpoints_inserted = 1;
@@ -3384,9 +3386,11 @@ normal_stop (void)
if (breakpoints_failed)
{
target_terminal_ours_for_output ();
- print_sys_errmsg ("ptrace", breakpoints_failed);
+ print_sys_errmsg ("While inserting breakpoints", breakpoints_failed);
printf_filtered ("Stopped; cannot insert breakpoints.\n\
-The same program may be running in another process.\n");
+The same program may be running in another process,\n\
+or you may have requested too many hardware breakpoints\n\
+and/or watchpoints.\n");
}
if (target_has_execution && breakpoints_inserted)