linux-kernel - Re: [PATCH 1/2] mmc: sdhci-uhs2: Fix calling incorrect sdhci_set

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] [thread-next>] [day] [month] [year] [list]
Message-ID: <416be416-014c-4efb-9f85-8f7023dcdc3f@intel.com>
Date: Wed, 3 Sep 2025 14:14:56 +0300
From: Adrian Hunter <adrian.hunter@...el.com>
To: Ben Chuang <benchuanggli@...il.com>
CC: <ulf.hansson@...aro.org>, <victor.shih@...esyslogic.com.tw>,
	<ben.chuang@...esyslogic.com.tw>, <HL.Liu@...esyslogic.com.tw>,
	<SeanHY.Chen@...esyslogic.com.tw>, <linux-mmc@...r.kernel.org>,
	<linux-kernel@...r.kernel.org>, <stable@...r.kernel.org>
Subject: Re: [PATCH 1/2] mmc: sdhci-uhs2: Fix calling incorrect
 sdhci_set_clock() function

On 02/09/2025 09:32, Ben Chuang wrote:
> On Tue, Sep 2, 2025 at 12:50 AM Adrian Hunter <adrian.hunter@...el.com> wrote:
>>
>> On 01/09/2025 15:07, Adrian Hunter wrote:
>>> On 01/09/2025 12:40, Ben Chuang wrote:
>>>> From: Ben Chuang <ben.chuang@...esyslogic.com.tw>
>>>>
>>>> Fix calling incorrect sdhci_set_clock() in __sdhci_uhs2_set_ios() when the
>>>> vendor defines its own sdhci_set_clock().
>>>>
>>>> Fixes: 10c8298a052b ("mmc: sdhci-uhs2: add set_ios()")
>>>> Cc: stable@...r.kernel.org # v6.13+
>>>> Signed-off-by: Ben Chuang <ben.chuang@...esyslogic.com.tw>
>>>> ---
>>>>  drivers/mmc/host/sdhci-uhs2.c | 5 ++++-
>>>>  1 file changed, 4 insertions(+), 1 deletion(-)
>>>>
>>>> diff --git a/drivers/mmc/host/sdhci-uhs2.c b/drivers/mmc/host/sdhci-uhs2.c
>>>> index 0efeb9d0c376..704fdc946ac3 100644
>>>> --- a/drivers/mmc/host/sdhci-uhs2.c
>>>> +++ b/drivers/mmc/host/sdhci-uhs2.c
>>>> @@ -295,7 +295,10 @@ static void __sdhci_uhs2_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
>>>>      else
>>>>              sdhci_uhs2_set_power(host, ios->power_mode, ios->vdd);
>>>>
>>>> -    sdhci_set_clock(host, host->clock);
>>>> +    if (host->ops->set_clock)
>>>> +            host->ops->set_clock(host, host->clock);
>>>> +    else
>>>> +            sdhci_set_clock(host, host->clock);
>>>
>>> host->ops->set_clock is not optional.  So this should just be:
>>>
>>>       host->ops->set_clock(host, host->clock);
>>>
> 
> I will update it. Thank you.
> 
>>
>> Although it seems we are setting the clock in 2 places:
>>
>>         sdhci_uhs2_set_ios()
>>                 sdhci_set_ios_common()
>>                         host->ops->set_clock(host, ios->clock)
>>               __sdhci_uhs2_set_ios
>>                         sdhci_set_clock(host, host->clock)
>>
>> Do we really need both?
>>
> 
> We only need one sdhci_set_clock() in __sdhci_uhs2_set_ios() for the
> UHS-II card interface detection sequence.
> Refer to Section 3.13.2, "Card Interface Detection Sequence" of the SD
> Host Controller Standard Spec. Ver. 7.00,
> First set the VDD1 power on and VDD2 power on, then enable the SD clock supply.
> 
> Do I need to add a separate patch or add it in the same patch like this?
> 
> diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
> index 3a17821efa5c..bd498b1bebce 100644
> --- a/drivers/mmc/host/sdhci.c
> +++ b/drivers/mmc/host/sdhci.c
> @@ -2369,7 +2369,8 @@ void sdhci_set_ios_common(struct mmc_host *mmc,
> struct mmc_ios *ios)
>                 sdhci_enable_preset_value(host, false);
> 
>         if (!ios->clock || ios->clock != host->clock) {
> -               host->ops->set_clock(host, ios->clock);
> +               if (!mmc_card_uhs2(host->mmc))
> +                       host->ops->set_clock(host, ios->clock);
>                 host->clock = ios->clock;
> 
>                 if (host->quirks & SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK &&

It can be a separate patch, but the whole of

	if (!ios->clock || ios->clock != host->clock) {
		etc
	}

needs to move from sdhci_set_ios_common() into
sdhci_set_ios() like further below.  Note, once that is done, you need
to add "host->clock = ios->clock;" to __sdhci_uhs2_set_ios()
like:
	host->ops->set_clock(host, ios->clock);
	host->clock = ios->clock;


diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index 3a17821efa5c..ac7e11f37af7 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -2367,23 +2367,6 @@ void sdhci_set_ios_common(struct mmc_host *mmc, struct mmc_ios *ios)
 		(ios->power_mode == MMC_POWER_UP) &&
 		!(host->quirks2 & SDHCI_QUIRK2_PRESET_VALUE_BROKEN))
 		sdhci_enable_preset_value(host, false);
-
-	if (!ios->clock || ios->clock != host->clock) {
-		host->ops->set_clock(host, ios->clock);
-		host->clock = ios->clock;
-
-		if (host->quirks & SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK &&
-		    host->clock) {
-			host->timeout_clk = mmc->actual_clock ?
-						mmc->actual_clock / 1000 :
-						host->clock / 1000;
-			mmc->max_busy_timeout =
-				host->ops->get_max_timeout_count ?
-				host->ops->get_max_timeout_count(host) :
-				1 << 27;
-			mmc->max_busy_timeout /= host->timeout_clk;
-		}
-	}
 }
 EXPORT_SYMBOL_GPL(sdhci_set_ios_common);
 
@@ -2410,6 +2393,23 @@ void sdhci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 
 	sdhci_set_ios_common(mmc, ios);
 
+	if (!ios->clock || ios->clock != host->clock) {
+		host->ops->set_clock(host, ios->clock);
+		host->clock = ios->clock;
+
+		if (host->quirks & SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK &&
+		    host->clock) {
+			host->timeout_clk = mmc->actual_clock ?
+						mmc->actual_clock / 1000 :
+						host->clock / 1000;
+			mmc->max_busy_timeout =
+				host->ops->get_max_timeout_count ?
+				host->ops->get_max_timeout_count(host) :
+				1 << 27;
+			mmc->max_busy_timeout /= host->timeout_clk;
+		}
+	}
+
 	if (host->ops->set_power)
 		host->ops->set_power(host, ios->power_mode, ios->vdd);
 	else