Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Wed, 13 Jul 2022 19:11:29 +0300
From:   Andy Shevchenko <andriy.shevchenko@linux.intel.com>
To:     Vamshi Gajjela <vamshigajjela@google.com>
Cc:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Jiri Slaby <jirislaby@kernel.org>,
        Miquel Raynal <miquel.raynal@bootlin.com>,
        Phil Edworthy <phil.edworthy@renesas.com>,
        Emil Renner Berthing <kernel@esmil.dk>,
        Heikki Krogerus <heikki.krogerus@linux.intel.com>,
        Johan Hovold <johan@kernel.org>, linux-serial@vger.kernel.org,
        linux-kernel@vger.kernel.org, manugautam@google.com
Subject: Re: [PATCH v3] serial: 8250_dw: Avoid pslverr on reading empty
 receiver fifo
Message-ID: <Ys7usW3W3kemtVc+@smile.fi.intel.com>
References: <20220713131722.2316829-1-vamshigajjela@google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220713131722.2316829-1-vamshigajjela@google.com>
Organization: Intel Finland Oy - BIC 0357606-4 - Westendinkatu 7, 02160 Espoo
Precedence: bulk

On Wed, Jul 13, 2022 at 06:47:22PM +0530, Vamshi Gajjela wrote:
> From: VAMSHI GAJJELA <vamshigajjela@google.com>
> 
> With PSLVERR_RESP_EN parameter set to 1, the device generates an error
> response when an attempt to read an empty RBR with FIFO enabled.
> 
> This happens when LCR writes are ignored when UART is busy.
> dw8250_check_lcr() in retries to update LCR, invokes dw8250_force_idle()
> to clear and reset FIFO and eventually reads UART_RX causing the error.
> 
> Avoid this by not reading RBR/UART_RX when no data is available.

From code perspective looks good,
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

but maybe Ilpo or others have something to add based on the practical use.

> Signed-off-by: VAMSHI GAJJELA <vamshigajjela@google.com>
> ---
> v3:
> - check lsr based on FIFO enablement
> v2:
> - update as per review comments (re-format comments, xmas tree ordering)
>  drivers/tty/serial/8250/8250_dw.c | 13 +++++++++++++
>  1 file changed, 13 insertions(+)
> 
> diff --git a/drivers/tty/serial/8250/8250_dw.c b/drivers/tty/serial/8250/8250_dw.c
> index f57bbd32ef11..7573904579f6 100644
> --- a/drivers/tty/serial/8250/8250_dw.c
> +++ b/drivers/tty/serial/8250/8250_dw.c
> @@ -82,8 +82,21 @@ static inline int dw8250_modify_msr(struct uart_port *p, int offset, int value)
>  static void dw8250_force_idle(struct uart_port *p)
>  {
>  	struct uart_8250_port *up = up_to_u8250p(p);
> +	unsigned int lsr;
>  
>  	serial8250_clear_and_reinit_fifos(up);
> +
> +	/*
> +	 * With PSLVERR_RESP_EN parameter set to 1, the device generates an
> +	 * error response when an attempt to read an empty RBR with FIFO
> +	 * enabled.
> +	 */
> +	if (up->fcr & UART_FCR_ENABLE_FIFO) {
> +		lsr = p->serial_in(p, UART_LSR);
> +		if (!(lsr & UART_LSR_DR))
> +			return;
> +	}
> +
>  	(void)p->serial_in(p, UART_RX);
>  }
>  
> -- 
> 2.37.0.144.g8ac04bfd2-goog
> 

-- 
With Best Regards,
Andy Shevchenko