From mboxrd@z Thu Jan  1 00:00:00 1970
From: mturquette@linaro.org (Michael Turquette)
Date: Fri, 08 May 2015 12:00:10 -0700
Subject: [PATCH v2 0/3] clk: si5351: Fixes for v4.1-rc1
In-Reply-To: <1430773456-15203-1-git-send-email-sebastian.hesselbarth@gmail.com>
References: <1430773456-15203-1-git-send-email-sebastian.hesselbarth@gmail.com>
Message-ID: <20150508190010.16410.26923@quantum>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

Quoting Sebastian Hesselbarth (2015-05-04 14:04:13)
> Mike, Stephen,
> 
> this is v2 of dealing with recently reported clk-si5351 issues. Compared
> to v1 [1], I decided to first split-off the pure fixes and postpone the
> improvements to a later patch set. Patches 1 and 3 should go though clk
> fixes while patch 2 should be taken by mvebu maintainers.
> 
> This fixes are intended for v4.1-rc1 and deal with issues where DT claimed
> parent clocks are not properly detected starting with v4.0. Looking at the
> code of clk-si5351, I just realized that the way the driver deals with
> parent clocks is utter nonsense. Russell King also mentioned that passing
> struct clk though platform_data shouldn't be done at all.
> 
> Therefore, this 3 patches rework parent clock handling of clk-si5351 to
> make use of (a) named parent clocks, (b) devm_clk_get() for both DT- and
> platform_data passed parent clocks, and (c) properly clk_prepare_enable()
> valid parent clocks.
> 
> However, I do have a stripped down backport of the fix for stable v4.0
> that just fixes the real issue, i.e. DT based probing and missing
> clk_prepare_enable(). If you agree with this fixes for v4.1-rc1 we should
> consider the backport for v4.0 stable.

I've pushed patches #1 and #3 to clk-fixes, now merged into clk-next.
I'll give it a couple of cycles in linux-next and then submit it for
4.1-rc4.

Regards,
Mike

> 
> Overall changes compared to v1:
> - Postpone improvement patch ("clk: si5351: Reset PLL after rate change")
>   for later patch set.
> - Reword parent clock check error message. (Suggested by Fabio Estevam)
> 
> Sebastian
> 
> [1] https://lkml.org/lkml/2015/4/30/688
> 
> Sebastian Hesselbarth (3):
>   clk: si5351: Mention clock-names in the binding documentation
>   ARM: dove: Add clock-names to CuBox Si5351 clk generator
>   clk: si5351: Do not pass struct clk in platform_data
> 
>  .../devicetree/bindings/clock/silabs,si5351.txt    |  4 +-
>  arch/arm/boot/dts/dove-cubox.dts                   |  1 +
>  drivers/clk/clk-si5351.c                           | 63 +++++++++++++++-------
>  include/linux/platform_data/si5351.h               |  4 --
>  4 files changed, 49 insertions(+), 23 deletions(-)
> 
> ---
> Cc: Mike Turquette <mturquette@linaro.org>
> Cc: Stephen Boyd <sboyd@codeaurora.org>
> Cc: Jean-Francois Moine <moinejf@free.fr>
> Cc: Michael Welling <mwelling@ieee.org>
> Cc: Russell King <rmk+linux@arm.linux.org.uk>
> Cc: Jason Cooper <jason@lakedaemon.net>
> Cc: Andrew Lunn <andrew@lunn.ch>
> Cc: Gregory Clement <gregory.clement@free-electrons.com>
> Cc: linux-clk at vger.kernel.org
> Cc: linux-arm-kernel at lists.infradead.org
> Cc: linux-kernel at vger.kernel.org
> -- 
> 2.1.0
>