Re: [PATCH for-next 1/8] RDMA/erdma: Probe the erdma RoCEv2 device

[Zhu Yanjun <yanjun.zhu@linux.dev>]

在 2024/11/28 3:07, Cheng Xu 写道:
> 
> 
> On 11/26/24 11:36 PM, Zhu Yanjun wrote:
>> 在 2024/11/26 7:59, Boshi Yu 写道:
>>> Currently, the erdma driver supports both the iWARP and RoCEv2 protocols.
>>> The erdma driver reads the ERDMA_REGS_DEV_PROTO_REG register to identify
>>> the protocol used by the erdma device. Since each protocol requires
>>> different ib_device_ops, we introduce the erdma_device_ops_iwarp and
>>> erdma_device_ops_rocev2 for iWARP and RoCEv2 protocols, respectively.
>>>
>> Hi, Boshi
>>
>>  From alomost all of the RDMA users, it seems that RoCEv2 protocol is required while very few users request iWARP protocol in their production hosts. But in the erdma HW/FW, this iWARP protocol is supported. Is this iWARP protocol for some special use cases in the Ali-Cloud environment?
>>
> 
> Hi, Yanjun,
> 
> iWarp & TCP have some good features, such as Tail Loss Probe(TLP), selective ACK and
> out-of-order DDP, so that can get better performance in large-scale lossy network. The

Thanks for your reply. The design purpose of RDMA is for big bandwidth 
and low latency network, that is, a loss-less network. And most users 
work in this loss-less network. I can not image why rdma should work in 
this kind of large-scale lossy network.

Anyway
Reviewed-by: Zhu Yanjun <yanjun.zhu@linux.dev>

Zhu Yanjun

> current usage scenarios include HPC industrial simulation, redis, PolarDB and big data
> in public cloud environment.
> 
> Thanks,
> Cheng Xu
> 
>> I am just curious about this, not to be against this patch series. If this is related with the Ali-Cloud security, you can ignore this.
>>
>> Thanks,
>> Zhu Yanjun
>>
>