Re: [PATCH] Increase timeout on gdb.base/exitsignal.exp

[Pedro Alves <palves@redhat.com>]

On 08/25/2015 07:35 PM, Sergio Durigan Junior wrote:
> On Tuesday, August 25 2015, Pedro Alves wrote:
> 
>> On 08/25/2015 07:24 PM, Sergio Durigan Junior wrote:
>>
>>>> This (many instances) doesn't make sense to me.  And I think wouldn't
>>>> fix anything.  Seems to me the bumped timeout, if any, should be around
>>>> the continue that caused the first time out:
>>>>
>>>> # Continue until the end.
>>>> gdb_test "continue" "Program terminated with signal SIGSEGV.*" \
>>>>     "program terminated with SIGSEGV"
>>>
>>> Sorry, I am confused...  The timeout does not occur on this command: it
>>> occurs on the print commands.  So I think we must extend the timeout for
>>> the print commands; don't we?
>>
>> Not AFAICS from what you pasted.  The print time outs seem to me to be
>> cascading fails.  Highlighting what you pasted:
> 
> Oh, you're right, I missed that (don't know how) when I was creating the
> patch.  Thanks for the catch; I'll update the patch.

I still don't think we should just bump the timeout for this without
understanding it better.  It very much looks like to me there's
a bug somewhere.  I see that it sometimes fails on the x86 buildbots too:

 http://gdb-build.sergiodj.net/builders/Fedora-x86_64-native-extended-gdbserver-m32/builds/1805

Thanks,
Pedro Alves