On 15 September 2016 at 12:19, Stephen Warren <swar...@wwwdotorg.org> wrote:
> From: Stephen Warren <swar...@nvidia.com>
> Some boards have a different set of USB controllers enabled in DT than
> the set referenced by /alias entries. This patch fixes that. For
> example, this avoids the following message while booting on Ventana,
> which is caused by the fact that the USB0 controller had no alias, and
> defaulted to wanting a sequence number of 0, which was later explicitly
> requested by the alias for USB controller 2.
> USB2: Device 'usb@c5008000': seq 0 is in use by 'usb@c5000000'
> This didn't affect USB operation in any way though.
> Related, there's no need for the USB controller aliases to have an order
> that's different from the HW order, so re-order any aliases to match the
> HW ordering. This has the benefit that since USB controller 0 is the only
> one that supports device-mode in HW, and U-Boot only supports enabling
> device move on controller 0, there's now good synergy in the ordering! For
> Tegra20, that's not relevant at present since USB device mode doesn't work
> correctly on that SoC, but it will save some head-scratching later.
> This patch doesn't fix the colibri_t20 board, even though it has the same
> issue, since Marcel already sent a patch for that.
> Cc: Marcel Ziswiler <marcel.ziswi...@toradex.com>
> Signed-off-by: Stephen Warren <swar...@nvidia.com>
> arch/arm/dts/tegra20-harmony.dts | 3 ++-
> arch/arm/dts/tegra20-seaboard.dts | 5 +++--
> arch/arm/dts/tegra20-trimslice.dts | 3 +--
> arch/arm/dts/tegra20-ventana.dts | 4 +++-
> 4 files changed, 9 insertions(+), 6 deletions(-)
Reviewed-by: Simon Glass <s...@chromium.org>
U-Boot mailing list