On Tuesday 25 March 2008, Samuel Thibault wrote: > Frans Pop, le Tue 25 Mar 2008 19:29:56 +0100, a écrit : > > (Please don't CC me on list mail.) > > Then tell your mailer to use followup-to :) That's an unofficial (or at least fairly recent) header that unfortunately my MUA does not support. However, the mailing list policy is quite clear on how list mail should be replied to, completely independent on how m-f-t is set to. If m-f-t (or reply-to) is set, you should honor it. But if it is not set, you should follow the mailing list rules. > > On Tuesday 25 March 2008, Samuel Thibault wrote: > > > The idea is to compile the speakup module out-of-tree but still > > > include it in d-i. > > > > Why would you want to compile a module that is in-tree as an > > out-of-tree module? > > Mmm, maybe I don't understand what you mean by "in-tree". Out-of-tree: not included in official upstream kernel source In-tree: included in upstream kernel source > The idea is not to compile speakup built into the kernel, but just as a > module, and then it doesn't need to be integrated to the kernel build > system, Compiling it into the kernel is not what I'm talking about. It definitely should be built as a module, but as it is part of the upstream kernel, it should be built just like any other regular module and included in linux-image-2.6.X-Y-<flavor>. If it is included in the kernel source, there is absolutely no reason to build it as part of a separate source package. In fact, that would be plain dumb. > but just be compiled separately, just like exmap, ndiswrapper, spca5xx > etc. The reason they are in linux-modules extra is exactly because they are _not_ included in the official upstream kernel source, and thus need to be in a different source package. If you wanted speakup to be in linux-modules-extra, you should not have sent your patches to lkml for inclusion...
Attachment:
signature.asc
Description: This is a digitally signed message part.