linux-kernel - Re: [PATCH 03/10] spi: Add flag for no TX after a RX in the same Chip Select

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Fri, 10 Sep 2021 15:44:42 +0100
From:   Mark Brown <broonie@...nel.org>
To:     Lucas tanure <tanureal@...nsource.cirrus.com>
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        "Rafael J . Wysocki" <rafael@...nel.org>,
        Sanjay R Mehta <sanju.mehta@....com>,
        Nehal Bakulchandra Shah <Nehal-Bakulchandra.shah@....com>,
        linux-kernel@...r.kernel.org, linux-spi@...r.kernel.org,
        patches@...nsource.cirrus.com
Subject: Re: [PATCH 03/10] spi: Add flag for no TX after a RX in the same
 Chip Select

On Thu, Sep 09, 2021 at 11:51:21AM +0100, Lucas tanure wrote:
> On 9/8/21 1:37 PM, Mark Brown wrote:
> > On Wed, Sep 08, 2021 at 12:34:44PM +0100, Lucas Tanure wrote:
> > > Some controllers can't write to the bus after a read without
> > > releasing the chip select, so add flag and a check in spi core

> > Nothing you've added ever reads this flag and I'm not sure what anything
> > would be able to constructively do with it so why add the flag?  I don't
> > understand what the use case is.

> __spi_validate checks this flag and makes sure the message can be received
> by the controller.
> __spi_validate can't fix the message, so it only rejects the message.

Given that this is hardware that can't possibly work how useful is that
validation?  It's a fairly unusual thing for devices to do in the first
place, only applies if using the native chip select (which your patch
doesn't check for) and I am not sure that this is a general enough
pattern in controllers to have generic support for.  I suspect that a
lot of controllers with similar restrictions will be even more limited
than this, for example only supporting one or two transfers with limits
on the data, so it's not clear to me how useful this capability would
be.

Download attachment "signature.asc" of type "application/pgp-signature" (489 bytes)