On Sat, Nov 8, 2008 at 11:37 AM, Mauro Carvalho Chehab <mchehab@xxxxxxxxxxxxx> wrote: > On Sat, 8 Nov 2008, Markus Rechberger wrote: > >> As written earlier already I don't think that I didn't follow any >> rules here since I provided single >> patches at the very first beginning >> >> http://mcentral.de/v4l-dvb/ >> (this is all kernel code and only kernel code). >> >> That work didn't get attention and due a different decision of >> framework changes (which that codebase relied >> on) it all had to be rebased, I doubt that anyone >> would have reworked all that patch for patch. Instead it went into one >> repository and finally got modified to work again >> with the available framework rather than relying onto any such >> modifications. > > One thing is to rebase a tree, another is to merge all patches into a big > one, not preserving the original authorships. > > Development trees sometimes need rebase. This is done by popping all newer > patches from the tree, applying the upstream patches, and then pushing again > every individual patches, fixing the ones that don't apply well, but > preserving their authorships. > > The modified patches should receive a special tag before the maintainer's > SOB, like: > > [me@mymail: I did this to apply this patch] > > as stated at the kernel docs. > > This method will reduce a lot the risk of breaking improvements and other > fixes that happened upstream, and will properly preserve authorship of > individual patches. > > If you were doing a rebase, your patches would likely be accepted. > Should I start picking patches from the linuxtv.org tree where patches from my tree are taken and where the Sign off is not provided? Just recently one patch went into not even stating that it comes from my codebase, although not worth the time making an elephant out of it. Markus _______________________________________________ linux-dvb mailing list linux-dvb@xxxxxxxxxxx http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
