[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <317f1f81-74da-68ae-200f-42a969774ae8@oracle.com>
Date: Thu, 5 Jan 2017 09:10:35 -0800
From: Santosh Shilimkar <santosh.shilimkar@...cle.com>
To: Sekhar Nori <nsekhar@...com>, David Lechner <david@...hnology.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Santosh Shilimkar <ssantosh@...nel.org>
Cc: Franklin S Cooper Jr <fcooper@...com>,
Rob Herring <robh+dt@...nel.org>,
Mark Rutland <mark.rutland@....com>,
Kevin Hilman <khilman@...nel.org>,
Axel Haslam <ahaslam@...libre.com>,
Alexandre Bailon <abailon@...libre.com>,
Bartosz Golaszewski <bgolaszewski@...libre.com>,
Jiri Slaby <jslaby@...e.com>, linux-serial@...r.kernel.org,
devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH v2 4/4] ARM: dts: keystone: Add "ti,da830-uart" compatible
string
On 1/5/2017 1:04 AM, Sekhar Nori wrote:
> Hi Santosh,
>
> On Thursday 05 January 2017 03:30 AM, Santosh Shilimkar wrote:
>> On 1/4/2017 12:30 PM, David Lechner wrote:
>>> The TI Keystone SoCs have extra UART registers beyond the standard 8250
>>> registers, so we need a new compatible string to indicate this. Also, at
>>> least one of these registers uses the full 32 bits, so we need to specify
>>> reg-io-width in addition to reg-shift.
>>>
>>> "ns16550a" is left in the compatible specification since it does work as
>>> long as the bootloader configures the SoC UART power management
>>> registers.
>>>
>> NAK!!
>> We can't break the booting boards with existing boot loaders.
>
> Sorry, but it not clear to me how this breaks booting with older
> bootloaders? If older DTB is ROM'ed, it will continue to work because of
> match with ns16550a.
>
> I just verified boot on K2E with these patches applied and using 2016.05
> based U-Boot from a TI release.
>
> http://pastebin.ubuntu.com/23744719/
>
Thanks for test. As long as it doesn't break the boot, am fine
with it.
>> I suggest you to first get the driver updated to take care of
>> the UART PM register and then enable the support for it.
>
> Isn't that what patch 2/4 is doing?
>
I see that now. Thanks for clarifying.
Serial patch needs to go via Greg's tree. I will pick up the
DTS bits.
Regards,
Santosh
Powered by blists - more mailing lists