linux-kernel - Re: [PATCH v2 2/2] mmc: sdhci-npcm: Add NPCM SDHCI driver

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]

Date:   Fri, 17 Mar 2023 11:37:54 -0700
From:   Guenter Roeck <linux@...ck-us.net>
To:     Andy Shevchenko <andy.shevchenko@...il.com>
Cc:     Tomer Maimon <tmaimon77@...il.com>, ulf.hansson@...aro.org,
        avifishman70@...il.com, tali.perry1@...il.com, joel@....id.au,
        venture@...gle.com, yuenn@...gle.com, benjaminfair@...gle.com,
        adrian.hunter@...el.com, skhan@...uxfoundation.org,
        davidgow@...gle.com, pbrobinson@...il.com, gsomlo@...il.com,
        briannorris@...omium.org, arnd@...db.de, krakoczy@...micro.com,
        openbmc@...ts.ozlabs.org, linux-mmc@...r.kernel.org,
        devicetree@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 2/2] mmc: sdhci-npcm: Add NPCM SDHCI driver

On 3/17/23 10:36, Andy Shevchenko wrote:
> On Fri, Mar 17, 2023 at 4:16 PM Guenter Roeck <linux@...ck-us.net> wrote:
>>
>> On Mon, Dec 05, 2022 at 10:53:51AM +0200, Tomer Maimon wrote:
>>> Add Nuvoton NPCM BMC sdhci-pltfm controller driver.
>>>
>>> Signed-off-by: Tomer Maimon <tmaimon77@...il.com>
>>
>> I still don't see this driver in the upstream kernel, or in linux-next.
>>
>> Couple of comments:
>>
>> - devm ordering does not really matter here. The devm resource
>>    is the clock, it does not depend on local data, and it will be
>>    released last, so that is ok.
> 
> Not sure. Strictly speaking this is the problem. If you leave a clock
> going on in a wrong period of time it (theoretically) might break your
> hardware once and forever. Similar discussion about power, clock and
> reset signals has been held for camera sensors.
> 

In general I agree, but not here. The remove function (sdhci_pltfm_unregister)
does call clk_disable_unprepare(), so the clock isn't left running.

Also, I think it is worthwhile to point out that exactly the same sequence
(sdhci_pltfm_init followed by devm_clk_get and cleanup/removal with
sdhci_pltfm_unregister) is shared among several sdhci drivers (including
the memory leak I pointed out, but only in the aspeed driver).

On a higher level I do agree that the sdhci platform code is in need of cleanup,
but I don't think it is appropriate to tie such a cleanup to this driver
submission.

Note that I don't really care much, I just realized that this patch is stuck
when I tried to test booting from SD drive with qemu.

Guenter