linux-kernel - Re: [PATCH] i2c: aspeed: Avoid accessing freed buffers during i2c transfers.

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>] [day] [month] [year] [list]

Message-ID: <CACPK8Xf6YssamEmHB5XDf8JYk+_=hnG8Yzqn4kCikseqg6rqOA@mail.gmail.com>
Date:   Mon, 31 Jul 2023 04:21:18 +0000
From:   Joel Stanley <joel@....id.au>
To:     Lixue Liang <lianglixuehao@....com>,
        Eddie James <eajames@...ux.ibm.com>
Cc:     brendan.higgins@...ux.dev, benh@...nel.crashing.org,
        p.zabel@...gutronix.de, linux-i2c@...r.kernel.org,
        openbmc@...ts.ozlabs.org, linux-arm-kernel@...ts.infradead.org,
        linux-aspeed@...ts.ozlabs.org, linux-kernel@...r.kernel.org,
        Lixue Liang <lianglixue@...atwall.com.cn>
Subject: Re: [PATCH] i2c: aspeed: Avoid accessing freed buffers during i2c transfers.

On Fri, 28 Jul 2023 at 12:40, Lixue Liang <lianglixuehao@....com> wrote:
>
> From: Lixue Liang <lianglixue@...atwall.com.cn>
>
> After waiting for the transmission timeout, the I2C controller will
> continue to transmit data when the bus is idle. Clearing bus->msg will
> avoid kernel panic when accessing the freed msg->buf in
> aspeed_i2c_master_irq.
>
> Signed-off-by: Lixue Liang <lianglixue@...atwall.com.cn>
> ---
>  drivers/i2c/busses/i2c-aspeed.c | 2 ++
>  1 file changed, 2 insertions(+)
>
> diff --git a/drivers/i2c/busses/i2c-aspeed.c b/drivers/i2c/busses/i2c-aspeed.c
> index 2e5acfeb76c8..c83057497e26 100644
> --- a/drivers/i2c/busses/i2c-aspeed.c
> +++ b/drivers/i2c/busses/i2c-aspeed.c
> @@ -713,6 +713,8 @@ static int aspeed_i2c_master_xfer(struct i2c_adapter *adap,
>                 spin_lock_irqsave(&bus->lock, flags);
>                 if (bus->master_state == ASPEED_I2C_MASTER_PENDING)
>                         bus->master_state = ASPEED_I2C_MASTER_INACTIVE;
> +
> +               bus->msgs = NULL;

Eddie, is this the same issue you were debugging?

>                 spin_unlock_irqrestore(&bus->lock, flags);
>
>                 return -ETIMEDOUT;
> --
> 2.27.0
>