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;
Message-ID: <36f37a18-0022-0368-bf7c-ebdd724b1558@linaro.org>
Date:   Wed, 19 Apr 2023 10:14:01 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.10.0
Subject: Re: [PATCH v2 2/4] spi: s3c64xx: add cpu_relax in polling loop
Content-Language: en-US
To:     Jaewon Kim <jaewon02.kim@samsung.com>,
        Mark Brown <broonie@kernel.org>, Andi Shyti <andi@etezian.org>,
        Alim Akhtar <alim.akhtar@samsung.com>
Cc:     linux-spi@vger.kernel.org, linux-samsung-soc@vger.kernel.org,
        linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org,
        Chanho Park <chanho61.park@samsung.com>
References: <20230419060639.38853-1-jaewon02.kim@samsung.com>
 <CGME20230419062755epcas2p43a646bbae5f01e3120331407ad873318@epcas2p4.samsung.com>
 <20230419060639.38853-3-jaewon02.kim@samsung.com>
From:   Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
In-Reply-To: <20230419060639.38853-3-jaewon02.kim@samsung.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 19/04/2023 08:06, Jaewon Kim wrote:
> Adds cpu_relax() to prevent long busy-wait.

How cpu_relax prevents long waiting?

> There is busy-wait loop to check data transfer completion in polling mode.
> 
> Signed-off-by: Jaewon Kim <jaewon02.kim@samsung.com>
> ---
>  drivers/spi/spi-s3c64xx.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/drivers/spi/spi-s3c64xx.c b/drivers/spi/spi-s3c64xx.c
> index 273aa02322d9..886722fb40ea 100644
> --- a/drivers/spi/spi-s3c64xx.c
> +++ b/drivers/spi/spi-s3c64xx.c
> @@ -568,6 +568,7 @@ static int s3c64xx_wait_for_pio(struct s3c64xx_spi_driver_data *sdd,
>  
>  	val = msecs_to_loops(ms);
>  	do {
> +		cpu_relax();

Shouldn't this be just readl_poll_timeout()? Or the syntax would be too
complicated?

>  		status = readl(regs + S3C64XX_SPI_STATUS);
>  	} while (RX_FIFO_LVL(status, sdd) < xfer->len && --val);
>  

Best regards,
Krzysztof