Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Mon, 13 Sep 2021 10:02:10 -0300
From:   Jason Gunthorpe <jgg@ziepe.ca>
To:     Florian Eckert <fe@dev.tdt.de>
Cc:     peterhuewe@gmx.de, jarkko@kernel.org,
        Eckert.Florian@googlemail.com, linux-integrity@vger.kernel.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 1/1] tpm/tpm_i2c_infineon: Fix init endian vendor check
Message-ID: <20210913130210.GD3544071@ziepe.ca>
References: <20210913120521.18572-1-fe@dev.tdt.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210913120521.18572-1-fe@dev.tdt.de>
Precedence: bulk

On Mon, Sep 13, 2021 at 02:05:21PM +0200, Florian Eckert wrote:
> On my embedded system I use this tpm infineon chip via i2c bus.
> The system is a MIPS architecture and therefore works in big endian mode.
> 
> The problem is, that the chip type is not correctly recognized,
> because the vendor ID is wrongly aligned in the memory.
> 
> By declaring the vendor ID variable as a `__le32` type, the TPM chip is
> then correctly recognized by the driver and feels then responsible.
> 
> The device works than as expected.
> 
> Signed-off-by: Florian Eckert <fe@dev.tdt.de>
> ---
> v2:
> * use variable type instead of le32_to_cpus function call
>  drivers/char/tpm/tpm_i2c_infineon.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

And if you do this it need to be made sparse clean/etc

Jason