Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <191c7661-c47f-5aba-97b9-ff698bda4bc7@acm.org>
Date:   Mon, 22 May 2023 06:31:44 -0700
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.11.0
Subject: Re: [PATCH] scsi: Let scsi_execute_cmd() mark args->sshdr as invalid
Content-Language: en-US
To:     John Garry <john.g.garry@oracle.com>,
        Juergen Gross <jgross@suse.com>,
        "Martin K. Petersen" <martin.petersen@oracle.com>
Cc:     linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org,
        "James E.J. Bottomley" <jejb@linux.ibm.com>, stable@vger.kernel.org
References: <20230511123432.5793-1-jgross@suse.com>
 <yq1ttwbsoii.fsf@ca-mkp.ca.oracle.com>
 <6614f626-d174-03d0-0993-79e6f6169b71@suse.com>
 <9d356278-c826-dacf-cbe0-79f512b7970e@oracle.com>
 <60aeffe4-b31d-4ea3-d4ea-f50ae25e0316@suse.com>
 <74879c87-689f-6a8e-a177-8bde4c9c4e51@oracle.com>
 <fb0efbd1-a54f-09d6-bd27-6f665b461e58@acm.org>
 <554bfa20-2228-8655-09e2-492cbfa183fa@oracle.com>
 <c9f0bc23-d5c1-23ba-2752-d89be9fef04a@acm.org>
 <611e1210-d89b-9046-ac3f-68a89af6159e@oracle.com>
 <ff04d098-17cc-42c5-cf72-2128fb43114e@acm.org>
 <73e3da3d-ed90-6c38-3c8c-13653bd89944@oracle.com>
From:   Bart Van Assche <bvanassche@acm.org>
In-Reply-To: <73e3da3d-ed90-6c38-3c8c-13653bd89944@oracle.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 5/22/23 02:55, John Garry wrote:
> On 19/05/2023 18:39, Bart Van Assche wrote:
>>        *args->resid = scmd->resid_len;
>> -    if (args->sense)
>> -        memcpy(args->sense, scmd->sense_buffer, SCSI_SENSE_BUFFERSIZE);
>> +    if (args->sense) {
>> +        *args->sense = scmd->sense_buffer;
>> +        scmd->sense_buffer = NULL;
> 
> I think that you will agree that this is not a good pattern to follow. 
> We cannot have SCSI core allocating the sense buffer but a driver 
> freeing it.

Why not? Something similar can happen anywhere in the kernel anywhere 
reference counting is used.

Bart.