On 2016-06-26 13:31, Ben Greear wrote: > On 06/26/2016 03:52 AM, Felix Fietkau wrote: >> On 2016-06-22 00:00, gree...@candelatech.com wrote: >>> From: Ben Greear <gree...@candelatech.com> >>> >>> This lets one use the ath9k and ath10k driver instead of the built-in >>> ath10k/ath9k driver from the upstream kernel (or backports). >>> >>> For ath10k, this should be a drop-in replacement, as well as enabling >>> better CT firmware support. >>> >>> For ath9k, this enables some 5Mhz channel and 4.9Ghz support, >>> but may loose some existing patches that LEDE carries for ath9k. >>> >>> Signed-off-by: Ben Greear <gree...@candelatech.com> >> I would really like to have this use kmod-ath from the main mac80211 >> patches in order to support using ath9k from the main package along with >> ath10k-ct. >> The packagea at least it needs a split between the ath9k and ath10k parts. > > That sounds good to me. I'm mostly interested in ath10k at this > point...probably > any ath9k changes could just be applied as patches instead of a fully external > driver. > > I guess I need to change any 'include "../foo.h" to <foo.h> and then somehow > fix > the make command to point to the backports tree? We could also export any relevant header files from the mac80211 package and make your package copy them into the tree (and then not build the ath module).
> Any suggestions on how to better sync with kernel compile options? Seems > ath10k > has a lot of different CONFIG* options. I hard-coded some in the make-file, > but > I am guessing that is fragile at best. We can export backport-include/backport/autoconf.h from the mac80211 package. From C code context you can use all the non-ath10k-specific options from there, e.g. CONFIG_ATH_DEBUG. Any other options should probably be defined like in the mac80211 package (though with different config symbol names to prevent conflicts). > If you have time to help with a bit of whatever is needed to get it properly > linking/compiling > against the proper code, I can work on making the build actually work. Sure, no problem. - Felix _______________________________________________ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
