Re: [Ada][PATCH][PING^2] GNAT/RTEMS patch for SVN Trunk ping

[Joel Sherrill <joel dot sherrill at oarcorp dot com>]

Arnaud Charlet wrote:
> > Any reason not to approve this patch?
> >     
>
> Last time I looked at this patch, it was missing changes from trunk,
>   
That was December.
> so I'll need to review the whole patch again to catch other possible
> inconsistencies, which takes time and is no fun. Also, as I said, there
> are other pending changes in this file that were delayed during stage3 that
> I'm about to commit, so this patch will need to take these changes
> into account.
>   

Not to be rude but how is it fair that you guys keep committing
hundreds of changes which were never sent to gcc-patches and
then throw back any outstanding patches your changes break?
I am not the only one with outstanding Ada patches in the
FSF GCC world.

This delay ignores the fact that you want me to edit the VxWorks port
(and I did) as part of my patch but so far not one AdaCore patch
has updated an RTEMS target specific file as part of your changes.

For example, there was bulk change to s-osinte-* to change the
standard comment and the RTEMS version wasn't touched. Cheap
change not done. 

Similarly there was a new constant to the sockets binding which
wasn't propagated to all targets.  I know this binding file is
automatically generated but the files were not edited to mark
them as needing updating and since I wasn't pinged, I assume
no target specific maintainers were asked to look at them.

The first checkin was >30K line diff touching over 300 files.
There have been checkins since then that were also large.

Is there a double standard here?

--joel
> Arno