About this list Date view Thread view Subject view Author view Attachment view

From: Herbert Pötzl (herbert_at_13thfloor.at)
Date: Thu 10 Jul 2003 - 16:35:18 BST


On Thu, Jul 10, 2003 at 05:19:45PM +0200, Dariush Pietrzak wrote:
> > those bastards, doing a 1200+ patches and not
> > keeping them up to date ...
> The idea is that with patchset it's almost trivial to keep it up to date -
> you've got your ctx16 patch from patchset that integrates nicely with the
> rest of your kernel, ctx17 comes out, you make ctx16->ctx17 patch, and
> you've got less work to do integrating.

hmm, depends, but for merely orthogonal patches
this is definitely true, nonetheless some developer
do not keep splitted patch sets, not only because
the diff between versions of incorporated patches
can also be applied to the whole patchdisaster ...

do not get me wrong, I've stood up against monster
mega patches several times, and personally I would
prefer 10-100 line patches which can be easily
reviewed/understood ...

> > I guess this is the fate of almost every
> > patch-combo starting as patchsets ...
> Nope. The patchset is there, only that it's secret.
> Me and few other people spent some time creating patchset with
> newer stuff in ( and against 2.4.21 ), and when we wanted to
> give it back, we heard 'yeah, but it doesen't integrate with
> my secret sooper-dooper patchset so it's useless to me'.

so why not start your own patch SET and give back
your 'imporvements/updates' to the community?

If people with a preference for split up patches
(like you and me) are given the choice, they will
choose the 'better' solution ...

> Friggin' great.
> ... i'll calm down now;)

best,
Herbert


About this list Date view Thread view Subject view Author view Attachment view
[Next/Previous Months] [Main vserver Project Homepage] [Howto Subscribe/Unsubscribe] [Paul Sladen's vserver stuff]
Generated on Thu 10 Jul 2003 - 16:54:27 BST by hypermail 2.1.3