]> git.ipfire.org Git - thirdparty/binutils-gdb.git/commit
Match any kind of error after "cannot resolve name" on lib/gdbserver-support.exp...
authorSergio Durigan Junior <sergiodj@redhat.com>
Mon, 30 Jul 2018 19:23:27 +0000 (15:23 -0400)
committerSergio Durigan Junior <sergiodj@redhat.com>
Mon, 30 Jul 2018 21:16:20 +0000 (17:16 -0400)
commitfb66cde8a42a383b111f0f1f48eb9f6daf9d736c
treeccf6ff307ea44ffe22c6974c15b37c3d7474c5af
parenteb41b24898e9858852c98f9275e7a4adee860d7b
Match any kind of error after "cannot resolve name" on lib/gdbserver-support.exp:gdbserver_start

On commit:

commit 7f1f7e23939adc7d71036a17fc6081e3af7ca585
Author: Sergio Durigan Junior <sergiodj@redhat.com>
Date:   Fri Jul 13 16:20:34 2018 -0400

    Expect for another variant of error message when gdbserver cannot resolve hostname

I extended the regular expression being used to identify whether
gdbserver could not resolve a (host)name.  This was needed because the
error message being printed had a different variation across some
systems.  However, as it turns out, I've just noticed that the message
has yet another variation:

  target remote tcp8:123:2353
  tcp8:123:2353: cannot resolve name: System error
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  tcp8:123:2353: No such file or directory.
  (gdb) FAIL: gdb.server/server-connect.exp: tcp8: connect to gdbserver using tcp8:123

which is causing FAILs on some systems (namely, Fedora-i686 on
BuildBot).

So instead of trying to predict everything that can be printed, I
decided to just match anything after the "cannot resolve name: " part.
This patch implements that.

Regression tested on the BuildBot.

gdb/testsuite/ChangeLog:
2018-07-30  Sergio Durigan Junior  <sergiodj@redhat.com>

* lib/gdbserver-support.exp (gdbserver_start): Match any kind of
error after "cannot resolve name" string.
gdb/testsuite/ChangeLog
gdb/testsuite/lib/gdbserver-support.exp