linux-kernel - Re: [PATCH v4 2/4] mmc: starfive: Add sdio/emmc driver support

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: <TY3P286MB26113FF6ABD3CBA616DEC6DC98959@TY3P286MB2611.JPNP286.PROD.OUTLOOK.COM>
Date:   Tue, 11 Apr 2023 02:04:14 +0800
From:   Shengyu Qu <wiagn233@...look.com>
To:     William Qiu <william.qiu@...rfivetech.com>,
        linux-riscv@...ts.infradead.org, devicetree@...r.kernel.org,
        linux-mmc@...r.kernel.org
Cc:     Rob Herring <robh+dt@...nel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
        Jaehoon Chung <jh80.chung@...sung.com>,
        Ulf Hansson <ulf.hansson@...aro.org>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v4 2/4] mmc: starfive: Add sdio/emmc driver support

Hello William,

> On 2023/3/29 0:08, Shengyu Qu wrote:
>> Hello William,
>>
>> Sorry for making noise about this, but seems deleted voltage swtich function
>>
>> doesn't help about this. But there's still a problem about eMMC speed. Currently
>>
>> only about 20MB/s maximum reading speed could be reached when using eMMC
>>
>> on VF2, any idea about this?
>>
>> Best regards,
>>
>> Shengyu
>>
>> 在 2023/3/28 0:01, Shengyu Qu 写道:
>>> Hello William,
>>>
>>> I'm digging into downstream mmc driver these days and found a problem
>>>
>>> that current version mainline driver doesn't has a voltage switch function for
>>>
>>> it. Downstream older version has one but was deleted in this commit [1].
>>>
>>> It was deleted since vf2's SD slot doesn't have 1.8V input but commiter forgot
>>>
>>> that vf2's eMMC slot has a proper 1.8V input.
>>>
>>> So could you add voltage switch function for mainline? I've met a eMMC speed
>>>
>>> problem possibly due to it.
>>>
>>> Best regards,
>>>
>>> Shengyu
>>>
> 
> Hi Shengyu,
> 
> Sorry for the late reply.
> 
> First of all, I will consider adding voltage switch function, but the implementation
> method is to configure the pmic register configuration in dts, and the implementation
> interface will use the voltage switch function in dw_mmc.c.
> 
> As for speed, the main reason for the low rate is the clock of JH7110 and the
> associated IO driving strength, in this limit, the maximum reading speed I tested was
> about 50Mb/s.
> 
> I will try to reproduce your problem and try to solve it. Thanks for suggestions.
> 
> Best regards,
> William
I found out the reason and fixed that. dmwci driver needs vqmmc supply
configured in device tree and a successful voltage change to actually
enable 1.8v mode, even 1.8 supply actually already physically exists.
So to solve this problem, I wrote AXP15060 driver and device tree
bindings basing on -upstream branch and gets over 75MB/s read speed. The
driver series is already under review here:

https://lore.kernel.org/all/TY3P286MB26117891DFB2DD615A7C54EF98969@TY3P286MB2611.JPNP286.PROD.OUTLOOK.COM/

Best regards,
Shengyu
Download attachment "OpenPGP_0xE3520CC91929C8E7.asc" of type "application/pgp-keys" (6869 bytes)

Download attachment "OpenPGP_signature" of type "application/pgp-signature" (834 bytes)