svn commit: trunk/busybox/libbb

Rob Landley rob at landley.net
Wed Jun 21 02:00:13 UTC 2006


On Tuesday 20 June 2006 7:28 pm, Mike Frysinger wrote:
> On Tuesday 20 June 2006 12:17, landley at busybox.net wrote:
> > Author: landley
> > Date: 2006-06-20 09:17:24 -0700 (Tue, 20 Jun 2006)
> > New Revision: 15439
> >
> > Log:
> > Revert the last two patches to go back to a state before this file was
> > controversial.
>
> i can understand revertting your patch since it was broken, but there was
> nothing wrong with
> http://busybox.net/cgi-bin/viewcvs.cgi?view=rev&rev=15412

It made the code bigger without a config option, added lots of ugly #ifdefs, 
was checked in after the feature freeze without any discussion on the list, 
and the "why does busybox need to add this new feature at all, even 115k is 
barely standard for a serial port and we already did 4 times that" which has 
_since_ been discussed but hadn't been up until then?

During a feature freeze it's polite to at least _ask_ "should this go in now".  
I don't even know who danf is.  (At this point I consider it a 1.3 issue, at 
which point it'll probably go back in but I'd like to at least think about 
finding a less ugly way to do it.  Possibly some "raw" syntax to specify the 
number to feed to the ioctl for the really fast custom speeds rather than us 
having a platform-specific lookup table composed entirely of #ifdefs.  I 
dunno, I'd have to think about it...)

> -mike

Rob
-- 
Never bet against the cheap plastic solution.



More information about the busybox mailing list