All of lore.kernel.org
 help / color / mirror / Atom feed
From: dgilbert@interlog.com (Douglas Gilbert)
To: linux-arm-kernel@lists.infradead.org
Subject: at91sam9x5: uart (not usart) broken
Date: Wed, 07 Aug 2013 11:45:41 -0400	[thread overview]
Message-ID: <52026BA5.40707@interlog.com> (raw)
In-Reply-To: <5201EA4D.8030106@overkiz.com>

On 13-08-07 02:33 AM, boris brezillon wrote:
> Hello Doug,
>
> On 07/08/2013 02:32, Douglas Gilbert wrote:
>> Between lk 3.10.0 and lk 3.11.0-rc4 the uarts (not usarts)
>> have been broken in the at91sam9x5 family. The DT files have
>> been re-factored for uarts but look correct.
>>
>> So suspicion moves to the atmel_serial driver which enjoyed
>> a lot of changes in June.
>>
>
> My patch (the last one in log history) is adding error checks on clk_get and
> clk_enable (clk_prepare_enable) calls.
> I think the issue you're having comes from these new checks, and the fact
> that uart clks DT entries are missing.
>
> The previous version was just silently ignoring these errors.
> It may work if these clks (uart clks) are already enabled by bootstrap or
> bootloader.
>
> Anyway, I think the cleaner way for resolving this issue is to apply the patch
> I joined in attachment.
>
> Could you test it (I don't have any sam9x5 boards), and if this works I'll submit
> it.

Hi Boris,
Yes, your patch works with lk 3.11.0-rc4 on a at91sam9g25
(aria G25). So please apply to the current rc series.

Also since in the lk 3.10 series this is silently broken and/or
works by accident (e.g. when a boot loader sets the uart clocks)
then perhaps this fix should be sent to Greg K-H for inclusion
in the lk 3.10.x stable series.

Doug Gilbert

>> Tested on a at91sam9g25 (Aria G25) which is a member of the
>> at91sam9x5 family with two uarts (four usarts and a debug
>> serial port). The failure looks like this in dmesg:
>>
>>  console [ttyS0] enabled
>>  f801c000.serial: ttyS1 at MMIO 0xf801c000 (irq = 23) is a ATMEL_SERIAL
>>  f8020000.serial: ttyS2 at MMIO 0xf8020000 (irq = 24) is a ATMEL_SERIAL
>>  f8024000.serial: ttyS3 at MMIO 0xf8024000 (irq = 25) is a ATMEL_SERIAL
>>  atmel_usart: probe of f8040000.serial failed with error -2
>>  f8028000.serial: ttyS4 at MMIO 0xf8028000 (irq = 32) is a ATMEL_SERIAL
>>
>> The device at 0xf8040000 is uart0 . My DT file was trying
>> to bring up uart0 but not uart1.
>>
>> Doug Gilbert
>

      reply	other threads:[~2013-08-07 15:45 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-08-07  0:32 at91sam9x5: uart (not usart) broken Douglas Gilbert
2013-08-07  6:33 ` boris brezillon
2013-08-07 15:45   ` Douglas Gilbert [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=52026BA5.40707@interlog.com \
    --to=dgilbert@interlog.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.