linux-kernel - Re: [PATCH 36/64] scsi: ibmvscsi: Avoid multi-field memset() overflow by aiming at srp

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]

Message-ID: <3ffbcf75-166e-5802-1d8e-9c7739961b80@linux.ibm.com>
Date:   Fri, 30 Jul 2021 12:11:02 -0700
From:   Tyrel Datwyler <tyreld@...ux.ibm.com>
To:     "Martin K. Petersen" <martin.petersen@...cle.com>,
        Kees Cook <keescook@...omium.org>
Cc:     linux-hardening@...r.kernel.org,
        "Gustavo A. R. Silva" <gustavoars@...nel.org>,
        Keith Packard <keithpac@...zon.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        linux-kernel@...r.kernel.org, linux-wireless@...r.kernel.org,
        netdev@...r.kernel.org, dri-devel@...ts.freedesktop.org,
        linux-staging@...ts.linux.dev, linux-block@...r.kernel.org,
        linux-kbuild@...r.kernel.org, clang-built-linux@...glegroups.com,
        Brian King <brking@...ux.vnet.ibm.com>
Subject: Re: [PATCH 36/64] scsi: ibmvscsi: Avoid multi-field memset() overflow
 by aiming at srp

On 7/28/21 8:35 PM, Martin K. Petersen wrote:
> 
> Kees,
> 
>> For example, change it to:
>>
>> +	BUILD_BUG_ON(sizeof(evt_struct->iu.srp) != SRP_MAX_IU_LEN);
>> +	memset(&evt_struct->iu.srp, 0x00, sizeof(evt_struct->iu.srp));
>>  	srp_cmd = &evt_struct->iu.srp.cmd;
>> -	memset(srp_cmd, 0x00, SRP_MAX_IU_LEN);
> 
>> For the moment, I'll leave the patch as-is unless you prefer having
>> the BUILD_BUG_ON(). :)
> 
> I'm OK with the BUILD_BUG_ON(). Hopefully Tyrel or Brian will chime in.
> 

All the other srp structs are at most 64 bytes and the size of the union is
explicitly set to SRP_MAX_IU_LEN by the last field of the union.

union srp_iu {
        struct srp_login_req login_req;
        struct srp_login_rsp login_rsp;
        struct srp_login_rej login_rej;
        struct srp_i_logout i_logout;
        struct srp_t_logout t_logout;
        struct srp_tsk_mgmt tsk_mgmt;
        struct srp_cmd cmd;
        struct srp_rsp rsp;
        u8 reserved[SRP_MAX_IU_LEN];
};

So, in my mind if SRP_MAX_IU_LEN ever changes so does the size of the union
making the BUILD_BUG_ON() superfluous. But it doesn't really hurt anything either.

-Tyrel