This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH] Postpone connecting from gdb_start to gdb_load in native-extended-gdbserver
- From: Doug Evans <dje at google dot com>
- To: Yao Qi <yao at codesourcery dot com>
- Cc: gdb-patches at sourceware dot org
- Date: Fri, 14 Sep 2012 15:57:14 -0700
- Subject: Re: [PATCH] Postpone connecting from gdb_start to gdb_load in native-extended-gdbserver
- References: <1346731269-2549-1-git-send-email-yao@codesourcery.com>
On Mon, Sep 3, 2012 at 9:01 PM, Yao Qi <yao@codesourcery.com> wrote:
> Hi,
> When running test cases with board file native-extended-gdbserver.exp
> (to test my 'general async notification' in all-stop mode), I find
> gdb connects to remote in proc gdb_start instead of gdb_load, so that some
> settings in test doesn't happen in remote, because they have to be set
> before GDB connects to remote.
> [...]
> --- a/gdb/testsuite/boards/native-extended-gdbserver.exp
> +++ b/gdb/testsuite/boards/native-extended-gdbserver.exp
> @@ -55,43 +55,14 @@ set_board_info gdb_server_prog "../gdbserver/gdbserver"
>
> send_user "configuring for gdbserver local testing (extended-remote)\n";
>
> +set_board_info gdb,do_reload_on_run 1
fwiw, I agree this needs to be done differently.
For example, I thought one of the features of extended-remote is that
one didn't have to do a reload on run (so to speak).