Hi Oleg On Wed, 2009-09-09 at 10:56 -0600, Yakovenko Oleg wrote: > This looks like a merge issue, however I prepared this patch to get my > feet wet ;) Please point me if I broke the process anywhere. Ouch .. that's weird -- yes, it is a merge issue. Why I wonder is where did I mess up that the compiler didn't catch it. Haven't been able to build the NAP branch for a while -- that could be it. The patch itself is ok, but yes, you broke the process :) I have posted a long list of tips and stuff I have collected over the years; I hope it'll be useful: http://linuxwimax.org/Developer/Tips Please submit patches inline vs attached -- use 'git send-email' to send the patches. This way they can be commented upon and tools import them directly from email gathering information such as date they were sent, etc and adding that to the repo. If you inline manually, watchout for tab-to-space conversion and line wrap around. I've inlined for commenting > From dc82a3bd18d197774083c640342a825d4e910cfd Mon Sep 17 00:00:00 2001 > From: Oleg Yakovenko <oyakovenko at yotateam.ru> > Date: Wed, 9 Sep 2009 20:15:37 +0400 > Subject: [PATCH] Remove a double definition of i2400m_power_save_disabled Missing subsystem header. Add a "wimax/i2400m:" header, so in the kernel commit logs, the single line description can tell what the patch applies to (wimax subsystem, i2400m driver). Obviosuly for other driver, it'd be something like 'scsi/aha912x:'. Missing longer description: even if this is a one liner, explain why the commit: "Module paraemter i2400m_power_save_disabled was defined twice; seems it wasn't caught by the empty headed maintainer...yadah yadah...". Missing "Signed-off-by": this tracks authorship of patches and the fact that you own the code: Signed-off-by: Your Name <your.name at email> Use the pointers in the Linux kernel Documentation/ Submit* documents, they are quite useful. Also the check-patch script in scripts/ helps a lot. > I also have a couple of questions (mostly process ones): > > > > It looks like there are duplicate repositories, e.g > linux/networking/wimax/wimax.git and linux/kernel/git/inaky/wimax.git. > Could you please comment? This was a confusion by the git.kernel.org admin; I'll ask it is removed. The primary one is linux/kernel/git/inaky/wimax.git, although that might change in the future to linux/kernel/git/wimax/wimax.git. > Is my understanding correct that Inaky maintains the whole wimax > source (both drivers and the stack) and sometimes it gets merged to > the mainline kernel? Yes. whenever the merge window open, I submit feature enhancements and minor bug fixes. Major bug fixes I send as they pop up. > What happens to the compat-* repositories? Is it a common practice to > replicate the whole kernel repository and do periodical updates from > the mainline? I do so every now and then, specially when big features or bugs come up, or when people ask...it's almost automatic. -- -- Inaky
