linux-kernel - Re: [PATCH v3 1/7] mmc: sdhci_am654: Add tuning algorithm for delay chain

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: <f4392550-bbec-4c26-a5de-ce29e9f34551@intel.com>
Date: Tue, 19 Mar 2024 08:35:10 +0200
From: Adrian Hunter <adrian.hunter@...el.com>
To: Judith Mendez <jm@...com>
Cc: Andrew Davis <afd@...com>, linux-mmc@...r.kernel.org,
 linux-kernel@...r.kernel.org, Ulf Hansson <ulf.hansson@...aro.org>
Subject: Re: [PATCH v3 1/7] mmc: sdhci_am654: Add tuning algorithm for delay
 chain

On 18/03/24 16:04, Judith Mendez wrote:
> On 3/14/24 9:18 AM, Adrian Hunter wrote:
>> On 8/03/24 02:57, Judith Mendez wrote:
>>> @@ -290,10 +297,12 @@ static void sdhci_am654_set_clock(struct sdhci_host *host, unsigned int clock)
>>>         regmap_update_bits(sdhci_am654->base, PHY_CTRL4, mask, val);
>>>   -    if (timing > MMC_TIMING_UHS_SDR25 && clock >= CLOCK_TOO_SLOW_HZ)
>>> +    if (timing > MMC_TIMING_UHS_SDR25 && clock >= CLOCK_TOO_SLOW_HZ) {
>>>           sdhci_am654_setup_dll(host, clock);
>>> -    else
>>> +        sdhci_am654->dll_enable = true;
>>> +    } else {
>>>           sdhci_am654_setup_delay_chain(sdhci_am654, timing);
>>
>> V2 patch had here:
>>
>>         sdhci_am654->dll_enable = false;
>>
>> Was its removal intended?
> 
> I did remove on purpose since it did not seem to be necessary.

I suspect it is necessary because ->set_clock() can be called in
when the timing has changed (e.g. recovery resets and reinitializes
the card device, or the card changes etc.) but it seems like
dll_enable would be stuck as always true once it is set to true.