On 10/15/2014 09:39 AM, Albert ARIBAUD wrote:
Hi Stephen,
On Wed, 15 Oct 2014 09:34:49 -0600, Stephen Warren swarren@wwwdotorg.org wrote:
On 10/15/2014 05:43 AM, Albert ARIBAUD wrote:
Hi Tom,
On Tue, 14 Oct 2014 23:14:24 +0000, Tom Warren TWarren@nvidia.com wrote:
Albert,
Please pull u-boot-tegra/master into ARM/master. Thanks!
The following changes since commit a7f99bf139b3aaa0d5494693fd0395084355e41a:
arm: Fix _start for CONFIG_SYS_DV_NOR_BOOT_CFG (2014-09-11 18:04:39 +0200)are available in the git repository at:
git://git.denx.de/u-boot-tegra.git masterfor you to fetch changes up to 90a565acf3a18c61170ec0e0b2046c98cb7ecc85:
arm: tegra: initial support for apalis t30 (2014-09-17 11:15:04 -0700)
Marcel Ziswiler (1): arm: tegra: initial support for apalis t30
Stephen Warren (1): ARM: tegra: add PCIe-related pins to the Jetson TK1 pinmux tables
This new board fails when I merge this branch to u-boot-arm/master, with the following diagnostics:
include/configs/apalis_t30.h:17:0: warning: "CONFIG_OF_SEPARATE" redefined [enabled by default] In file included from /home/albert.u.boot/src/u-boot-arm/include/linux/kconfig.h:4:0, from <command-line>:0: include/generated/autoconf.h:8:0: note: this is the location of the previous definition
(actually, there is a lot of these, for various config options)
Device Tree Source is not correctly specified. Please define 'CONFIG_DEFAULT_DEVICE_TREE' or build with 'DEVICE_TREE=<device_tree>' argument
(This one is fatal even though it does not explicitly state "error:")
This failure does not happen on the tegra branch; it is obviously due to the Kbuild stuff.
This issue should be fixed by:
https://patchwork.ozlabs.org/patch/398793/ [U-Boot] tegra: apalis_t30: master revamp
However, I would assert that at this point during the release cycle, we probably shouldn't be adding new boards/features, but just fixing bugs. Perhaps we should drop the Apalis patch completely and re-apply it once the release is out?
Ok. Should I still apply the TK1 pinmux patch, or should I consider the whole PR to be for -next?
I consider the Jetson TK1 pinmux patch a bug-fix; it adds entries that always should have been there.