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;
Subject: Re: [PATCH] scsi: target: pscsi: Fix possible null-pointer
 dereference in pscsi_complete_cmd()
To:     Tuo Li <islituo@gmail.com>, martin.petersen@oracle.com
Cc:     linux-scsi@vger.kernel.org, target-devel@vger.kernel.org,
        linux-kernel@vger.kernel.org, baijiaju1990@gmail.com,
        TOTE Robot <oslab@tsinghua.edu.cn>
References: <20210807134651.245436-1-islituo@gmail.com>
From:   Bodo Stroesser <bostroesser@gmail.com>
Message-ID: <dea07ecc-7700-5ee7-aa40-2d4455dc6c3f@gmail.com>
Date:   Mon, 9 Aug 2021 12:36:03 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.11.0
MIME-Version: 1.0
In-Reply-To: <20210807134651.245436-1-islituo@gmail.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 07.08.21 15:46, Tuo Li wrote:
> The return value of transport_kmap_data_sg() is assigned to the variable
> buf:
>    buf = transport_kmap_data_sg(cmd);
> 
> And then it is checked:
>    if (!buf) {
> 
> This indicates that buf can be NULL. However, it is dereferenced in the
> following statements:
>    if (!(buf[3] & 0x80))
>      buf[3] |= 0x80;
>    if (!(buf[2] & 0x80))
> 	buf[2] |= 0x80;
> 
> To fix these possible null-pointer dereferences, dereference buf only when
> it is not NULL.
> 
> Reported-by: TOTE Robot <oslab@tsinghua.edu.cn>
> Signed-off-by: Tuo Li <islituo@gmail.com>
> ---
>   drivers/target/target_core_pscsi.c | 14 +++++++-------
>   1 file changed, 7 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/target/target_core_pscsi.c b/drivers/target/target_core_pscsi.c
> index 2629d2ef3970..560815729182 100644
> --- a/drivers/target/target_core_pscsi.c
> +++ b/drivers/target/target_core_pscsi.c
> @@ -620,14 +620,14 @@ static void pscsi_complete_cmd(struct se_cmd *cmd, u8 scsi_status,
>   			buf = transport_kmap_data_sg(cmd);
>   			if (!buf) {
>   				; /* XXX: TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE */
> -			}
> -
> -			if (cdb[0] == MODE_SENSE_10) {
> -				if (!(buf[3] & 0x80))
> -					buf[3] |= 0x80;
>   			} else {
> -				if (!(buf[2] & 0x80))
> -					buf[2] |= 0x80;
> +				if (cdb[0] == MODE_SENSE_10) {
> +					if (!(buf[3] & 0x80))
> +						buf[3] |= 0x80;
> +				} else {
> +					if (!(buf[2] & 0x80))
> +						buf[2] |= 0x80;
> +				}
>   			}
>   
>   			transport_kunmap_data_sg(cmd);
> 

I'm wondering whether we should better put the
transport_kunmap_data_sg into the else-branch of the if (!buf)?
AFAICS, calling it after transport_kmap_data_sg failed does not
cause problems, but I feel it would be cleaner.

Otherwise it looks good to me.