On Mon, 27 Apr 2015 10:42:32 -0300 Ezequiel Garcia <ezequiel@xxxxxxxxxxxxxxxxxxxx> wrote: > On 04/24/2015 08:41 AM, Alban Bedel wrote: > > This series add OF bindings and code support for the interrupt > > controllers, clocks and GPIOs. However it was only tested on a > > TL-WR1043ND with an AR9132, others SoCs are untested, and a few are > > not supported at all. > > > Hi Alban, Hi, > I've booted a Carambola2 using this (plus a custom devicetree and some > small changes): > > Tested-by: Ezequiel Garcia <ezequiel@xxxxxxxxxxxxxxxxxxxx> > > Just a small comment/question: Shouldn't we allow to build all the > devicetree files, instead of just the one that will be built-in? > > I.e., something like this: > > dtb-$(CONFIG_MATCH_ATH79_DT) += ar9132_tl_wr1043nd_v1.dtb > dtb-$(CONFIG_MACH_ATH79_DT) += ar9331_carambola2.dtb > > It should be useful to catch errors, but also in general, as the > devicetree is supposed to be independent of the kernel and should be > built separate from it. Yes, that would be better, I'll fix that. > PS: This series depends on a previous patchset. It's usually useful to > mention this in the cover letter and make a poor tester's life > easier :) > > Most code changes base on the previous bug fix series: > > [PATCH v2 0/5] MIPS: ath79: Various small fix to prepare OF support Wasn't that clear enough? > Thanks for the work, Thanks for the testing. Alban
