Re: [RESEND][PATCH 5/5] Eliminate mi_run_to_main, introduce mi_clean_restart

[Tom de Vries <tdevries@suse.de>]

On 10/13/20 11:36 PM, Pedro Alves wrote:
> On 10/13/20 7:19 PM, Simon Marchi wrote:
>> On 2020-10-12 2:56 p.m., Pedro Alves wrote:
>>> diff --git a/gdb/testsuite/lib/mi-support.exp b/gdb/testsuite/lib/mi-support.exp
>>> index 693c7d2c467..0de3aa8703a 100644
>>> --- a/gdb/testsuite/lib/mi-support.exp
>>> +++ b/gdb/testsuite/lib/mi-support.exp
>>> @@ -1023,28 +1023,41 @@ proc mi_run_with_cli {args} {
>>>      return [eval mi_run_cmd_full 0 $args]
>>>  }
>>>
>>> -#
>>> -# Just like run-to-main but works with the MI interface
>>> -#
>>> -
>>> -proc mi_run_to_main { } {
>>> -    global suppress_flag
>>> -    if { $suppress_flag } {
>>> -	return -1
>>> -    }
>>> +# Starts fresh GDB binary and loads an optional executable into GDB.
>>> +# Usage: mi_clean_restart [executable]
>>> +# EXECUTABLE is the basename of the binary.
>>> +# Return -1 if starting gdb or loading the executable failed.
>>>
>>> +proc mi_clean_restart { args } {
>>
>> Again, for a new optional argument I'd suggest using parse_args,
>> meaning calling mi_clean_restart with an explicit binfile would be:
>>
>>   mi_clean_restart -binfile ${binfile}
>>
>> But, this is not very important, the next person to add an argument to
>> mi_clean_restart can do it (if it ever happens).  The patch LGTM as it
>> is.
> 
> Yeah, I left this one as is to have it mirror the CLI's clean_restart.
> 
> Thanks for the quick reviews.  I've merged the series now.

Caused this and similar regressions:
...
FAIL: gdb.ada/mi_catch_ex.exp: continue to exception catchpoint hit
(unknown output after running)
...

Filed as https://sourceware.org/bugzilla/show_bug.cgi?id=26732

Thanks,
- Tom