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: <995aeae7-86f5-494a-3f7a-1fcc32f27181@linux.alibaba.com>
Date:   Fri, 4 Aug 2023 11:59:03 +0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:102.0)
 Gecko/20100101 Thunderbird/102.14.0
Subject: Re: [RFC PATCH net-next 1/6] net/smc: support smc release version
 negotiation in clc handshake
To:     Simon Horman <horms@kernel.org>
Cc:     wenjia@linux.ibm.com, jaka@linux.ibm.com, kgraul@linux.ibm.com,
        tonylu@linux.alibaba.com, davem@davemloft.net, edumazet@google.com,
        kuba@kernel.org, pabeni@redhat.com, alibuda@linux.alibaba.com,
        guwen@linux.alibaba.com, linux-s390@vger.kernel.org,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org
References: <20230803132422.6280-1-guangguan.wang@linux.alibaba.com>
 <20230803132422.6280-2-guangguan.wang@linux.alibaba.com>
 <ZMvnIszqS4ZpkYHj@kernel.org>
Content-Language: en-US
From:   Guangguan Wang <guangguan.wang@linux.alibaba.com>
In-Reply-To: <ZMvnIszqS4ZpkYHj@kernel.org>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

Hi Simon,

Thanks for the review.

Before this patch set, the ini pointer is NULL when sending accept clc msg. This patch set
has changed the ini pointer to non-NULL value both when sending accept clc msg and when
sending confirm clc msg. And the ini pointer in smc_clc_send_confirm_accept will not be NULL
any more.

I will remove the ini NULL check in the next version.
if (ini && clc->hdr.type == SMC_CLC_CONFIRM) => if (clc->hdr.type == SMC_CLC_CONFIRM)

Thanks,
Guangguan Wang

On 2023/8/4 01:42, Simon Horman wrote:
> On Thu, Aug 03, 2023 at 09:24:17PM +0800, Guangguan Wang wrote:
> 
> ...
> 
> Hi Guangguan Wang,
> 
>> @@ -1063,7 +1063,7 @@ static int smc_clc_send_confirm_accept(struct smc_sock *smc,
>>  				memcpy(clc_v2->r1.eid, eid, SMC_MAX_EID_LEN);
>>  			len = SMCR_CLC_ACCEPT_CONFIRM_LEN_V2;
>>  			if (first_contact) {
>> -				smc_clc_fill_fce(&fce, &len);
>> +				smc_clc_fill_fce(&fce, &len, ini->release_ver);
> 
> Here ini is dereferenced...
> 
> 
>>  				fce.v2_direct = !link->lgr->uses_gateway;
>>  				memset(&gle, 0, sizeof(gle));
>>  				if (ini && clc->hdr.type == SMC_CLC_CONFIRM) {
> 
> ... but here it is assumed that ini may be NULL.
> 
> This seems inconsistent.
> 
> As flagged by Smatch.
> 
> ...
>