linux-kernel - Re: [PATCH] mmc: sdhci_am654: Add workaround for card detect debounce timer

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Wed, 5 Aug 2020 18:46:28 +0530
From:   Faiz Abbas <faiz_abbas@...com>
To:     Adrian Hunter <adrian.hunter@...el.com>,
        <linux-kernel@...r.kernel.org>, <linux-mmc@...r.kernel.org>
CC:     <ulf.hansson@...aro.org>
Subject: Re: [PATCH] mmc: sdhci_am654: Add workaround for card detect debounce
 timer

Hi Adrian,

On 05/08/20 3:16 pm, Adrian Hunter wrote:
> On 5/08/20 11:22 am, Faiz Abbas wrote:
>> Hi Adrian,
>>
>> On 05/08/20 1:44 pm, Adrian Hunter wrote:
>>> On 30/07/20 2:41 am, Faiz Abbas wrote:
>>>> There is a one time delay because of a card detect debounce timer in the
>>>> controller IP. This timer runs as soon as power is applied to the module
>>>> regardless of whether a card is present or not and any writes to
>>>> SDHCI_POWER_ON will return 0 before it expires. This timeout has been
>>>> measured to be about 1 second in am654x and j721e.
>>>>
>>>> Write-and-read-back in a loop on SDHCI_POWER_ON for a maximum of
>>>> 1.5 seconds to make sure that the controller actually powers on.
>>>>
>>>> Signed-off-by: Faiz Abbas <faiz_abbas@...com>
>>>> ---
>>>>  drivers/mmc/host/sdhci_am654.c | 21 +++++++++++++++++++++
>>>>  1 file changed, 21 insertions(+)
>>>>
>>>> diff --git a/drivers/mmc/host/sdhci_am654.c b/drivers/mmc/host/sdhci_am654.c
>>>> index 1718b9e8af63..55cff9de2f3e 100644
>>>> --- a/drivers/mmc/host/sdhci_am654.c
>>>> +++ b/drivers/mmc/host/sdhci_am654.c
>>>> @@ -272,6 +272,7 @@ static void sdhci_j721e_4bit_set_clock(struct sdhci_host *host,
>>>>  	sdhci_set_clock(host, clock);
>>>>  }
>>>>  
>>>> +#define MAX_POWER_ON_TIMEOUT	1500 /* ms */
>>>>  static void sdhci_am654_write_b(struct sdhci_host *host, u8 val, int reg)
>>>>  {
>>>>  	unsigned char timing = host->mmc->ios.timing;
>>>> @@ -291,6 +292,26 @@ static void sdhci_am654_write_b(struct sdhci_host *host, u8 val, int reg)
>>>>  	}
>>>>  
>>>>  	writeb(val, host->ioaddr + reg);
>>>> +	if (reg == SDHCI_POWER_CONTROL && (val & SDHCI_POWER_ON)) {
>>>> +		/*
>>>> +		 * Power on will not happen until the card detect debounce
>>>> +		 * timer expires. Wait at least 1.5 seconds for the power on
>>>> +		 * bit to be set
>>>> +		 */
>>>
>>> Can you use readb_poll_timeout() here?
>>>
>>
>> The loop is write -> readback -> check for set bit -> write again and so on until timeout
>> so poll_timeout() calls will not work.
> 
> I mentioned it because pedantically you need to check the condition
> again after a timeout.  Alternatively, the read_poll_timeout macro

Ideally, the timeout will never happen because the internal timer always expires at
1 second. The actual time spent in the loop can be anything less than 1 second
depending on when clocks were enabled

> can be used with a function, something like below (not even compile
> tested!)
> 
> static u8 write_power_on(struct sdhci_host *host, u8 val, int reg)
> {
> 	writeb(val, host->ioaddr + reg);
> 	usleep_range(1000, 10000);
> 	return readb(host->ioaddr + reg);
> }
> 
> #define MAX_POWER_ON_TIMEOUT	1500000 /* us */
> static void sdhci_am654_write_b(struct sdhci_host *host, u8 val, int reg)
> {
> 	unsigned char timing = host->mmc->ios.timing;
> 	u8 pwr;
> 
> 	if (reg == SDHCI_HOST_CONTROL) {
> 		switch (timing) {
> 		/*
> 		 * According to the data manual, HISPD bit
> 		 * should not be set in these speed modes.
> 		 */
> 		case MMC_TIMING_SD_HS:
> 		case MMC_TIMING_MMC_HS:
> 		case MMC_TIMING_UHS_SDR12:
> 		case MMC_TIMING_UHS_SDR25:
> 			val &= ~SDHCI_CTRL_HISPD;
> 		}
> 	}
> 
> 	writeb(val, host->ioaddr + reg);
> 
> 	/*
> 	 * Power on will not happen until the card detect debounce
> 	 * timer expires. Wait at least 1.5 seconds for the power on
> 	 * bit to be set
> 	 */
> 	if (reg == SDHCI_POWER_CONTROL && (val & SDHCI_POWER_ON) &&
> 	    read_poll_timeout(write_power_on, pwr, (pwr & SDHCI_POWER_ON), 0,
> 			      MAX_POWER_ON_TIMEOUT, false, host, val, reg))
> 			dev_warn(mmc_dev(host->mmc), "Power on failed\n");
> 			return;
> 		}
> 	}
> }
> 

Looks good. Let me add this in v2.

Thanks,
Faiz