gdb/testsuite: fix failure in gdb.mi/mi-pending.exp with extended-remote
authorAndrew Burgess <aburgess@redhat.com>
Sat, 18 Feb 2023 22:50:52 +0000 (22:50 +0000)
committerAndrew Burgess <aburgess@redhat.com>
Tue, 28 Feb 2023 10:56:28 +0000 (10:56 +0000)
commit05ac6365e558a207fcf7fa39c3fc8c7b1d0601aa
tree5b149ccc9b2ce68177ce0c3d1b9c725e327dd354
parent47171eeb9468fd38e3785b0181d63a2abea3bf29
gdb/testsuite: fix failure in gdb.mi/mi-pending.exp with extended-remote

I currently see this failure when running the gdb.mi/mi-pending.exp
test using the native-extended-remote board:

  -break-insert -f -c x==4 mi-pendshr.c:pendfunc2
  &"No source file named mi-pendshr.c.\n"
  ^done,bkpt={number="2",type="breakpoint",disp="keep",enabled="y",addr="<PENDING>",pending="mi-pendshr.c:pendfunc2",cond="x==4",evaluated-by="host",times="0",original-location="mi-pendshr.c:pendfunc2"}
  (gdb)
  FAIL: gdb.mi/mi-pending.exp: MI pending breakpoint on mi-pendshr.c:pendfunc2 if x==4 (unexpected output)

The failure is caused by the 'evaluated-by="host"' string, which only
appears in the output when the test is run using the
native-extended-remote board.

I could fix this by just updating the pattern in
gdb.mi/mi-pending.exp, but I have instead updated mi-pending.exp to
make more use of the support procs in mi-support.exp.  This did
require making a couple of adjustments to mi-support.exp, but I think
the result is that mi-pending.exp is now easier to read, and I see no
failures with native-extended-remote anymore.

One of the test names has changed after this work, I think the old
test name was wrong - it described a breakpoint as pending when the
breakpoint was not pending, I suspect a copy & paste error.

But there's no changes to what is actually being tested after this
patch.

Approved-By: Pedro Alves <pedro@palves.net>
gdb/testsuite/gdb.mi/mi-pending.exp
gdb/testsuite/lib/mi-support.exp