From mboxrd@z Thu Jan  1 00:00:00 1970
From: Andy Shevchenko <andriy.shevchenko-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
Subject: Re: [PATCH v2 2/3] spi: pxa2xx: Prepare for edge-triggered
 interrupts
Date: Mon, 16 Jan 2017 21:07:17 +0200
Message-ID: <1484593637.2133.152.camel@linux.intel.com>
References: <cover.1484592296.git.jan.kiszka@siemens.com>
         <7b15a0910a3ad861fd32161c72559bafa7b71e29.1484592296.git.jan.kiszka@siemens.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8bit
Cc: linux-spi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org,
        Daniel Mack <daniel-cYrQPVfZoowdnm+yROfE0A@public.gmane.org>,
        Haojian Zhuang <haojian.zhuang-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>,
        Robert Jarzmik <robert.jarzmik-GANU6spQydw@public.gmane.org>,
        linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
        Mika Westerberg <mika.westerberg-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>,
        Jarkko Nikula <jarkko.nikula-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
To: Jan Kiszka <jan.kiszka-kv7WeFo6aLtBDgjK7y7TUQ@public.gmane.org>,
        Mark Brown <broonie-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
Return-path: <linux-spi-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>
In-Reply-To: <7b15a0910a3ad861fd32161c72559bafa7b71e29.1484592296.git.jan.kiszka-kv7WeFo6aLtBDgjK7y7TUQ@public.gmane.org>
Sender: linux-spi-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
List-ID: <linux-spi.vger.kernel.org>

On Mon, 2017-01-16 at 19:44 +0100, Jan Kiszka wrote:
> When using the a device with edge-triggered interrupts, such as MSIs,
> the interrupt handler has to ensure that there is a point in time
> during
> its execution where all interrupts sources are silent so that a new
> event can trigger a new interrupt again.
> 
> This is achieved here by looping over SSSR evaluation. We need to take
> into account that SSCR1 may be changed by the transfer handler, thus
> we
> need to redo the mask calculation, at least regarding the volatile
> interrupt enable bit (TIE).
> 

So, more comments/questions below.

>  
>  	sccr1_reg = pxa2xx_spi_read(drv_data, SSCR1);
>  
> -	/* Ignore possible writes if we don't need to write */
> -	if (!(sccr1_reg & SSCR1_TIE))
> -		mask &= ~SSSR_TFS;
> -
>  	/* Ignore RX timeout interrupt if it is disabled */
>  	if (!(sccr1_reg & SSCR1_TINTE))
>  		mask &= ~SSSR_TINT;
>  
> -	if (!(status & mask))
> -		return IRQ_NONE;
> +	while (1) {

Can we switch to do-while and move previous block here? Btw, can TINTE
bit be set again during a loop?

> +		/* Ignore possible writes if we don't need to write
> */
> +		if (!(sccr1_reg & SSCR1_TIE))
> +			mask &= ~SSSR_TFS;
>  
> -	if (!drv_data->master->cur_msg) {
> -		handle_bad_msg(drv_data);
> -		/* Never fail */
> -		return IRQ_HANDLED;
> -	}
> +		if (!(status & mask))
> +			return ret;
> +
> +		if (!drv_data->master->cur_msg) {
> +			handle_bad_msg(drv_data);
> +			/* Never fail */
> +			return IRQ_HANDLED;
> +		}
> +

> +		ret |= drv_data->transfer_handler(drv_data);

So, we might call handler several times. This needs to be commented in
the code why you do so.

>  
> -	return drv_data->transfer_handler(drv_data);
> +		status = pxa2xx_spi_read(drv_data, SSSR);

Would it be possible to get all 1:s from the register
(something/autosuspend just powered off it by timeout?) ?

> +		sccr1_reg = pxa2xx_spi_read(drv_data, SSCR1);
> +	}
>  }
>  
>  /*

-- 
Andy Shevchenko <andriy.shevchenko-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
Intel Finland Oy
--
To unsubscribe from this list: send the line "unsubscribe linux-spi" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html