[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <72515578-AF72-4F37-AB69-EC1930D27FF4@konsulko.com>
Date: Mon, 17 Oct 2016 22:15:27 +0300
From: Pantelis Antoniou <pantelis.antoniou@...sulko.com>
To: Ralf Ramsauer <ralf@...ses-pyramidenbau.de>
Cc: Mark Brown <broonie@...nel.org>,
Geert Uytterhoeven <geert@...ux-m68k.org>,
Wolfram Sang <wsa@...-dreams.de>,
Linux SPI <linux-spi@...r.kernel.org>,
Linux I2C <linux-i2c@...r.kernel.org>,
"linux-kernel @ vger . kernel . org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2 0/2] spi, i2c: mark device nodes only in case of successful instantiation
Hi Ralf,
> On Oct 17, 2016, at 16:59 , Ralf Ramsauer <ralf@...ses-pyramidenbau.de> wrote:
>
> Hi,
>
> this one fixes initialisation of I2C/SPI nodes. Upon failure during
> intialisation, nodes were erroneously populated and never unmarked.
>
> This lead to the problem that re-loaded drivers will never probe those devices
> again and can easily be fixed by clearing the OF_POPULATE flag when the node
> doesn't successfully initialise.
>
> For the discussion of v1, see
> https://lkml.org/lkml/2016/10/14/483
>
> Ralf
>
> changes since v1:
> - also fix I2C core driver
> - keep the atomic test-and-set, as Geert suggested
>
> Ralf Ramsauer (2):
> spi: mark device nodes only in case of successful instantiation
> i2c: mark device nodes only in case of successful instantiation
>
> drivers/i2c/i2c-core.c | 11 ++++++++++-
> drivers/spi/spi.c | 5 ++++-
> 2 files changed, 14 insertions(+), 2 deletions(-)
>
> --
> 2.10.1
>
Thanks for catching this.
Acked-by: Pantelis Antoniou <pantelis.antoniou@...sulko.com>
Powered by blists - more mailing lists