[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20180910140340.GR28860@dell>
Date: Mon, 10 Sep 2018 15:03:40 +0100
From: Lee Jones <lee.jones@...aro.org>
To: Marcel Ziswiler <marcel@...wiler.com>
Cc: linux-kernel@...r.kernel.org,
Marcel Ziswiler <marcel.ziswiler@...adex.com>
Subject: Re: [RESEND PATCH] mfd: stmpe: honor gpio interrupt probe deferral
On Tue, 14 Aug 2018, Marcel Ziswiler wrote:
> From: Marcel Ziswiler <marcel.ziswiler@...adex.com>
>
> Actually honor probe deferral in trying to get the GPIO interrupt as
> of_get_named_gpio_flags() in stmpe_of_probe() may as well just do so.
>
> Signed-off-by: Marcel Ziswiler <marcel.ziswiler@...adex.com>
>
> ---
>
> drivers/mfd/stmpe.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/mfd/stmpe.c b/drivers/mfd/stmpe.c
> index 566caca4efd8..4d6ffd3ef398 100644
> --- a/drivers/mfd/stmpe.c
> +++ b/drivers/mfd/stmpe.c
> @@ -1331,6 +1331,8 @@ int stmpe_probe(struct stmpe_client_info *ci, enum stmpe_partnum partnum)
> return -ENOMEM;
>
> stmpe_of_probe(pdata, np);
> + if (-EPROBE_DEFER == pdata->irq_gpio)
> + return -EPROBE_DEFER;
EPROBE_DEFER usually means "stop what you're doing and return", rather
than "do a few more things, then return". With that in mind, I would
rather you convert stmpe_of_probe() to return and int containing the
appropriate error as soon as it is detected.
--
Lee Jones [李琼斯]
Linaro Services Technical Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog
Powered by blists - more mailing lists