Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.32 as permitted sender) client-ip=23.128.96.32;
Date:   Tue, 21 Nov 2023 11:15:04 -0500
From:   Hugo Villeneuve <hugo@hugovil.com>
To:     Jiri Slaby <jirislaby@kernel.org>
Cc:     gregkh@linuxfoundation.org, linux-kernel@vger.kernel.org,
        linux-serial@vger.kernel.org, lech.perczak@camlingroup.com,
        Hugo Villeneuve <hvilleneuve@dimonoff.com>
Message-Id: <20231121111504.6fca4b22bd6d3cc05c6fb668@hugovil.com>
In-Reply-To: <90ba18b8-1a31-498d-a103-5c91b80cea5b@kernel.org>
References: <20231120163931.2872431-1-hugo@hugovil.com>
        <20231120163931.2872431-2-hugo@hugovil.com>
        <90ba18b8-1a31-498d-a103-5c91b80cea5b@kernel.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Subject: Re: [PATCH v2 1/1] serial: sc16is7xx: improve regmap debugfs by
 using one regmap per port
Precedence: bulk

On Tue, 21 Nov 2023 06:57:56 +0100
Jiri Slaby <jirislaby@kernel.org> wrote:

> On 20. 11. 23, 17:39, Hugo Villeneuve wrote:
> > From: Hugo Villeneuve <hvilleneuve@dimonoff.com>
> > 
> > With this current driver regmap implementation, it is hard to make sense
> > of the register addresses displayed using the regmap debugfs interface,
> > because they do not correspond to the actual register addresses documented
> > in the datasheet. For example, register 1 is displayed as registers 04 thru
> > 07:
> ...
> > --- a/drivers/tty/serial/sc16is7xx.c
> > +++ b/drivers/tty/serial/sc16is7xx.c
> ...
> >   #ifdef CONFIG_SERIAL_SC16IS7XX_SPI
> >   static int sc16is7xx_spi_probe(struct spi_device *spi)
> >   {
> >   	const struct sc16is7xx_devtype *devtype;
> > -	struct regmap *regmap;
> > +	struct regmap *regmaps[2];
> 
> IMO, in all places, it would make sense to declare a VLA with 
> devtype->nr_uart.
 
Hi Jiri,
are VLA now allowed in kernel code?

 
> > +	unsigned int i;
> >   	int ret;
> >   
> >   	/* Setup SPI bus */
> > @@ -1732,11 +1736,20 @@ static int sc16is7xx_spi_probe(struct spi_device *spi)
> >   		devtype = (struct sc16is7xx_devtype *)id_entry->driver_data;
> >   	}
> >   
> > -	regcfg.max_register = (0xf << SC16IS7XX_REG_SHIFT) |
> > -			      (devtype->nr_uart - 1);
> > -	regmap = devm_regmap_init_spi(spi, &regcfg);
> > +	for (i = 0; i < devtype->nr_uart; i++) {
> > +		regcfg.name = sc16is7xx_regmap_name(i);
> > +		/*
> > +		 * If read_flag_mask is 0, the regmap code sets it to a default
> > +		 * of 0x80. Since we specify our own mask, we must add the READ
> > +		 * bit ourselves:
> > +		 */
> > +		regcfg.read_flag_mask = sc16is7xx_regmap_port_mask(i) |
> > +			SC16IS7XX_SPI_READ_BIT;
> > +		regcfg.write_flag_mask = sc16is7xx_regmap_port_mask(i);
> > +		regmaps[i] = devm_regmap_init_spi(spi, &regcfg);
> 
> As you trip over the array until devtype->nr_uart which is of course up 
> to 2. For now.
> 
> Or at least add an asserion: devtype->nr_uart <= 2 somewhere.

devtype->nr_uart is coming from const struct sc16is7xx_devtype, so it
is always a constant, do we need an assert in this case?

Hugo.


> thanks,
> -- 
> js
> suse labs
> 
>