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;
Message-ID: <0b73df73-d5e8-32a8-1495-63596b256392@linux.alibaba.com>
Date:   Sat, 22 Jan 2022 00:48:32 +0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:91.0)
 Gecko/20100101 Thunderbird/91.5.0
Subject: Re: [RFC PATCH net-next] net/smc: Introduce receive queue flow
 control support
Content-Language: en-US
To:     Tony Lu <tonylu@linux.alibaba.com>
Cc:     kgraul@linux.ibm.com, davem@davemloft.net, kuba@kernel.org,
        linux-s390@vger.kernel.org, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org
References: <20220120065140.5385-1-guangguan.wang@linux.alibaba.com>
 <YelwGOBhjBFsVPxA@TonyMac-Alibaba>
From:   Guangguan Wang <guangguan.wang@linux.alibaba.com>
In-Reply-To: <YelwGOBhjBFsVPxA@TonyMac-Alibaba>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 2022/1/20 22:22, Tony Lu wrote:>>  #include "smc_ib.h"
>>  
>> -#define SMC_RMBS_PER_LGR_MAX	255	/* max. # of RMBs per link group */
>> +#define SMC_RMBS_PER_LGR_MAX	32	/* max. # of RMBs per link group. Correspondingly,
>> +					 * SMC_WR_BUF_CNT should not be less than 2 *
>> +					 * SMC_RMBS_PER_LGR_MAX, since every connection at
>> +					 * least has two rq/sq credits in average, otherwise
>> +					 * may result in waiting for credits in sending process.
>> +					 */
> 
> This gives a fixed limit for per link group connections. Using tunable
> knobs to control this for different workload would be better. It also
> reduce the completion of free slots in the same link group and link.
> 

It is a good idea, but I find a patch (https://lore.kernel.org/linux-s390/20220114054852.38058-7-tonylu@linux.alibaba.com/) where you have already done this idea.