Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 30 Mar 2021 11:52:12 +0300
From:   Dan Carpenter <dan.carpenter@oracle.com>
To:     David Mosberger-Tang <davidm@egauge.net>
Cc:     Ajay.Kathat@microchip.com, Claudiu.Beznea@microchip.com,
        kvalo@codeaurora.org, linux-wireless@vger.kernel.org,
        kernel-janitors@vger.kernel.org
Subject: Re: [PATCH] wilc1000: fix a loop timeout condition
Message-ID: <20210330085212.GA2088@kadam>
References: <YFS5gx/gi70zlIaO@mwanda>
 <cd087f6b-5f99-3bce-0015-ccf1a82625f2@microchip.com>
 <37239c87142346dcba616cc63c64294dc274983b.camel@egauge.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <37239c87142346dcba616cc63c64294dc274983b.camel@egauge.net>
User-Agent: Mutt/1.9.4 (2018-02-28)
Precedence: bulk

On Mon, Mar 29, 2021 at 12:47:15PM -0600, David Mosberger-Tang wrote:
> On Fri, 2021-03-19 at 16:09 +0000, Ajay.Kathat@microchip.com wrote:
> > On 19/03/21 8:17 pm, Dan Carpenter wrote:
> > > If the loop fails, the "while(trials--) {" loop will exit with "trials"
> > > set to -1.  The test for that expects it to end with "trials" set to 0
> > > so the warning message will not be printed.
> > > 
> > > Fix this by changing from a post-op to a pre-op.  This does mean that
> > > we only make 99 attempts instead of 100 but that's okay.
> > > 
> > > Fixes: f135a1571a05 ("wilc1000: Support chip sleep over SPI")
> > > Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
> > 
> > Thanks Dan.
> 
> Good catch, but wouldn't it be better to fix the time-out check
> condition instead?  Something a long the lines of:
> 
> --- drivers/net/wireless/microchip/wilc1000/wlan.c~	2021-03-29 12:44:52.066039259 -0600
> +++ drivers/net/wireless/microchip/wilc1000/wlan.c	2021-03-29 12:40:29.176365116 -0600
> @@ -457,7 +457,7 @@
>  	u32 wakeup_reg, wakeup_bit;
>  	u32 to_host_from_fw_reg, to_host_from_fw_bit;
>  	u32 from_host_to_fw_reg, from_host_to_fw_bit;
> -	u32 trials = 100;
> +	int trials = 100;
>  	int ret;
>  
>  	if (wilc->io_type == WILC_HIF_SDIO) {
> @@ -483,7 +483,7 @@
>  		if ((reg & to_host_from_fw_bit) == 0)
>  			break;
>  	}
> -	if (!trials)
> +	if (trials < 0)
>  		pr_warn("FW not responding\n");
>  
>  	/* Clear bit 1 */
> 
> 
> This way, the loop could actually get executed the number of times
> indicated by the initialization of "trial" before issuing a warning
> message.

Those numbers are just made up...  It doesn't matter either way.

regards,
dan carpenter