Re: [PATCH 2/2] gdb/testsuite: tighten up some end-of-line patterns

[Andrew Burgess <aburgess@redhat.com>]

Tom de Vries <tdevries@suse.de> writes:

> On 12/8/23 23:25, Tom de Vries wrote:
>> On 11/30/23 20:23, Tom Tromey wrote:
>>>>>>>> "Andrew" == Andrew Burgess <aburgess@redhat.com> writes:
>>>
>>> Andrew> Following on from the previous commit, I searched the 
>>> testsuite for
>>> Andrew> places where we did:
>>>
>>> Andrew>   set eol "<some pattern>"
>>>
>>> Andrew> in most cases the <some pattern> could be replaced with "\r\n" 
>>> though
>>> Andrew> in the stabs test I've switched to using the multi_line proc 
>>> as that
>>> Andrew> seemed like a better choice.
>>>
>>> Andrew> In gdb.ada/info_types.exp I did need to add an extra use of 
>>> $eol as
>>> Andrew> the previous pattern would match multiple newlines, and in 
>>> this one
>>> Andrew> place we were actually expecting to match multiple newlines.  The
>>> Andrew> tighter pattern only matches a single newline, so we now need 
>>> to be
>>> Andrew> explicit when multiple newlines are expected -- I think this 
>>> is a good
>>> Andrew> thing.
>>>
>>> Andrew> All the tests are still passing for me after these changes.
>>>
>>> Looks good to me, thank you.
>>> Approved-By: Tom Tromey <tom@tromey.com>
>> 
>> I  see some regressions:
>> ...
>> FAIL: gdb.ada/catch_assert_if.exp: continuing to expected failed assertion
>> FAIL: gdb.ada/catch_ex.exp: continuing to first exception
>> FAIL: gdb.ada/catch_ex.exp: continuing to second exception
>> FAIL: gdb.ada/catch_ex.exp: continuing to Program_Error exception
>> FAIL: gdb.ada/catch_ex.exp: continuing to failed assertion
>> FAIL: gdb.ada/catch_ex.exp: continuing to unhandled exception
>> FAIL: gdb.ada/catch_ex.exp: continuing to temporary catchpoint
>> FAIL: gdb.ada/excep_handle.exp: continuing to first Constraint_Error 
>> exception handlers
>> FAIL: gdb.ada/excep_handle.exp: continuing and stopping in Storage_Error 
>> exception handlers
>> FAIL: gdb.ada/excep_handle.exp: continuing without stopping to 
>> Program_Error exception handlers
>> FAIL: gdb.ada/excep_handle.exp: continuing without stopping to 
>> Storage_Error exception handlers
>> FAIL: gdb.ada/excep_handle.exp: continuing to second Constraint_Error 
>> exception handlers
>> FAIL: gdb.ada/excep_handle.exp: continuing to Program_Error exception 
>> handlers
>> ...
>> 
>> For instance:
>> ...
>> (gdb) PASS: gdb.ada/catch_assert_if.exp: Check catch assertions with 
>> condition
>> continue^M
>> Continuing.^M
>> ^M
>> Catchpoint 2, failed assertion at 0x000000000040201a in bla () at 
>> /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/catch_assert_if/bla.adb:32^M
>> 32            pragma Assert (Global_Var = 1, "Error #2"); -- STOP^M
>> (gdb) FAIL: gdb.ada/catch_assert_if.exp: continuing to expected failed 
>> assertion
>> ...
>
> I see this failure both with openSUSE Leap 15.4 and openSUSE Tumbleweed.
>
> Fixed on both by:
> ...
> diff --git a/gdb/testsuite/gdb.ada/catch_assert_if.exp 
> b/gdb/testsuite/gdb.ada/catch_assert_if.exp
> index 9b094d88dc8..dd9db10ebe1 100644
> --- a/gdb/testsuite/gdb.ada/catch_assert_if.exp
> +++ b/gdb/testsuite/gdb.ada/catch_assert_if.exp
> @@ -52,7 +52,7 @@ set bp_location [gdb_get_line_number "STOP" 
> ${testdir}/bla.adb]
>   set catchpoint_msg \
>     "Catchpoint $decimal, failed assertion at $hex in bla \\\(\\\).*at 
> .*bla.adb:$bp_location"
>   gdb_test "continue" \
> -         "Continuing\.$eol$catchpoint_msg$eol.*STOP" \
> +         "Continuing\.$eol$eol$catchpoint_msg$eol.*STOP" \
>            "continuing to expected failed assertion"
>
>   gdb_test "continue" \
> ...
>
> Does this pass for you without this patch?

All the tests you listed are showing as UNSUPPORTED for me due to the
runtime debuginfo package being missing.  So far I've been unable to
find/install the correct package.

The change above looks reasonable to me, I suspect all these tests will
be fixed by similar updates.

FWIW, have a pre-approval for fixes for these regressions.  And thanks
for looking into these.

Approved-By: Andrew Burgess <aburgess@redhat.com>

Thanks,
Andrew