This is the mail archive of the
mailing list for the GDB project.
Re: [RFA]: Clean up debug printing of pc in gdbserver
- From: Pedro Alves <pedro at codesourcery dot com>
- To: gdb-patches at sourceware dot org
- Cc: Doug Evans <dje at google dot com>
- Date: Sun, 24 May 2009 17:47:58 +0100
- Subject: Re: [RFA]: Clean up debug printing of pc in gdbserver
- References: <20090507031109.0CBFB84890@localhost>
> The debug printing of pc in linux-i386-low.c/linux-x86-64-low.c can
> use some clean up.
> - why restrict the printing to just x86?
> - the text that is printed for linux_resume_one_lwp is confusing
> Ok to check in?
> 2009-05-06 ?Doug Evans ?<firstname.lastname@example.org>
> ????????* linux-x86-64-low.c (debug_threads): Remove declaration.
> ????????(x86_64_get_pc,x86_64_set_pc): Remove debug printing of pc.
> ????????* linux-i386-low.c (debug_threads): Remove declaration.
> ????????(i386_get_pc,i386_set_pc): Remove debug printing of pc.
> ????????* linux-low.c (get_stop_pc): Print pc if debug_threads.
> ????????(check_removed_breakpoint, linux_wait_for_lwp): Ditto.
> ????????(linux_resume_one_lwp): Ditto.
Looks OK to me. Small nit below.
On Thursday 07 May 2009 04:11:08, Doug Evans wrote:
> if (debug_threads
> - && WIFSTOPPED (*wstatp))
> + && WIFSTOPPED (*wstatp)
> + && the_low_target.get_pc != NULL)
> struct thread_info *saved_inferior = current_inferior;
> + CORE_ADDR stop_pc = (*the_low_target.get_pc) ();
> current_inferior = (struct thread_info *)
> find_inferior_id (&all_threads, child->head.id);
> - /* For testing only; i386_stop_pc prints out a diagnostic. */
> - if (the_low_target.get_pc != NULL)
> - get_stop_pc ();
> + fprintf (stderr, "linux_wait_for_lwp: pc is %08lx\n", (long) stop_pc);
> current_inferior = saved_inferior;
Can we rename that `stop_pc' variable to, say, `pc', so we're
consistent throughout? We can think of the `stop_pc' as having
always the decr_pc_after_break adjustment applied.