linux-kernel - Re: [PATCH 3/3] mmc: rtsx: Add SD Express mode support for RTS5261

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: <20201027193725.GA5579@lst.de>
Date:   Tue, 27 Oct 2020 20:37:25 +0100
From:   Christoph Hellwig <hch@....de>
To:     Ulf Hansson <ulf.hansson@...aro.org>
Cc:     冯锐 <rui_feng@...lsil.com.cn>,
        Christoph Hellwig <hch@....de>, Arnd Bergmann <arnd@...db.de>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        "linux-mmc@...r.kernel.org" <linux-mmc@...r.kernel.org>
Subject: Re: [PATCH 3/3] mmc: rtsx: Add SD Express mode support for RTS5261

On Tue, Oct 27, 2020 at 01:54:46PM +0100, Ulf Hansson wrote:
> > For information security purpose, some companies or business users set their notebook SD as "read only".
> > Because a lot of "read only" requirements from those companies or business users, notebook vendor controls reader write protect pin to achieve it.
> > Notebook BIOS might have option to choose "read only" or not.
> > This is why we think write protect is more important than speed.
> 
> I understand that it may be used, in some way or the other to provide
> a hint to the operating system to mount it in read-only mode.
> 
> Although, if there were a real security feature involved, the internal
> FW of the SD card would also monitor the switch, to support read-only
> mode. As I understand it, that's not the common case.

Yes.  "Security" that relies on the driver to fall back to a different
mode doesn't work.

> 
> > If you prefer to consistent behavior, I can ignore the write protect switch for SD express.
> 
> At this point, I prefer if you would ignore the write protect switch
> in the SD controller driver.

Same here.

> According to Christoph, it should be possible to support read-only
> mode via PCIe/NVMe. You may need to add some tweaks to support this in
> the PCIe controller driver, but I can't advise you how to exactly do
> this.

The NVMe driver already supports write protected namespaces.

I'll ask my contact in the JEDEC SD card working group if there was
any consideration of the read-only handling for classic SD vs NVMe.