Re: [PATCH 0/4] Handle test execution timeout consistently across protocols

Maciej W. Rozycki wrote:
> On Tue, 12 Dec 2023, Jacob Bachmeyer wrote:
>   
>>>  A change for the timeout factor will follow.  As it consists of a DejaGNU
>>> part and a GCC part I chose to send it separately so as not to cross-post
>>> this whole series between the DejaGNU and GCC mailing lists.
>>>
>>>  Any questions, comments, or concerns?  Otherwise please apply.
>>>       
>> Patches 1, 2, and 3 applied as offered on branch mwr-patch-20231212.  Patch 4
>> has been revised and applied on the branch because the command to be executed
>> is free-form text while the timeout is a simple number; the revised format
>> ("Executing on MACHINE with timeout TIMEOUT: COMMAND") is easier to reliably
>> parse.
>>     
>
>  Well:
>
> "Executing on $hostname: $program $pargs $inp $outp (timeout = $timeout)"
>
> is the format `remote_exec' has been using for decades (the message comes 
> from before the beginning of our repo),[...]

Indeed those do.  (I still have yet to fully dig into lib/remote.exp.)  
I had thought that patch 4 was adding a new message, thus the 
willingness to change it.

> I made the new messages consistent with the existing one and I'd prefer 
> that they (all) used the long-established format.  There's just been too 
> much mental imprint to it at this stage.  Therefore please reconsider your 
> decision.

Done.  Branch mwr-patch-20231212 is withdrawn and replaced with branch 
mwr-patch-20231212-1, which contains patch 4 as offered as well.  If 
these are good, I will merge that branch to master soon.

Lastly, I also await your reply on the questions I had about the 
"test_timeout_factor" patch.


-- Jacob