linux-kernel - Re: [PATCH] spi: atmel: Fix CS and initialization bug

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:   Tue, 29 Jun 2021 18:07:41 +0100
From:   Mark Brown <broonie@...nel.org>
To:     Dan.Sneddon@...rochip.com
Cc:     linux-spi@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
        linux-kernel@...r.kernel.org, Tudor.Ambarus@...rochip.com,
        Nicolas.Ferre@...rochip.com, alexandre.belloni@...tlin.com,
        Ludovic.Desroches@...rochip.com
Subject: Re: [PATCH] spi: atmel: Fix CS and initialization bug

On Tue, Jun 29, 2021 at 05:01:57PM +0000, Dan.Sneddon@...rochip.com wrote:
> On 6/29/21 9:47 AM, Mark Brown wrote:

>  >In what way does it do that?  I can't tell what the patch is supposed >to
>  >do.

> The SPI_MASTER_GPIO_SS flag has to be set so that the set_cs function 
> gets called even when using gpio cs pins.

This all needs to be clear in the changelog.

>  >> -	enable =3D (!!(spi->mode & SPI_CS_HIGH) =3D=3D enable);
>  >> =20
>  >> -	if (enable) {
>  >> +	if ((enable && (spi->mode & SPI_CS_HIGH))
>  >> +			|| (!enable && !(spi->mode & SPI_CS_HIGH))) {

>  >This looks especially suspicious.

> It's due to the fact that the spi core tells set_cs if the cs should be 
> high or low, not active or disabled.  This logic is to convert from 
> high/low to active/disabled.

spi_set_cs() handles SPI_CS_HIGH...  this looks like a separate existing
driver bug, it should just be ignoring SPI_CS_HIGH if it's providing a
set_cs() operation and letting the core implement SPI_CS_HIGH for it.  I
only checked breifly but it looks like spi-atmel is trying to use the
core support for chipselect handling here.

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