Re: about RTL8168 PMD on ARM SoC

[Thomas Monjalon]

When do you expect or desire to merge it into upstream DPDK?
It's always better to know the intended roadmap, thanks.


29/08/2022 17:40, Hau:
> Hi Honnappa,
> 
> Thanks for prompt reply.
> Currently our pmd driver still has other issues. After we fix it, we will 
> submit our code as RFC.
> 
> Thanks,
> Hau
> 
> From: Honnappa Nagarahalli [mailto:honnappa.nagaraha...@arm.com]
> Sent: Saturday, August 27, 2022 12:44 AM
> To: xing_wang <xing_w...@realsil.com.cn>; dev@dpdk.org
> Cc: dali_chen <dali_c...@realsil.com.cn>; howard_wang 
> <howard_w...@realsil.com.cn>; Ruifeng Wang <ruifeng.w...@arm.com>; Hau 
> <h...@realtek.com>; nd <n...@arm.com>
> Subject: RE: about RTL8168 PMD on ARM SoC
> 
> Hi Xing Wang,
>                Can you submit this code as RFC, it will enable us to provide 
> review comments?
> 
> Thanks,
> Honnappa
> 
> 
> From: 王星 <xing_w...@realsil.com.cn<mailto:xing_w...@realsil.com.cn>>
> Sent: Thursday, August 25, 2022 9:36 PM
> To: Honnappa Nagarahalli 
> <honnappa.nagaraha...@arm.com<mailto:honnappa.nagaraha...@arm.com>>; 
> dev@dpdk.org<mailto:dev@dpdk.org>
> Cc: 陈立 <dali_c...@realsil.com.cn<mailto:dali_c...@realsil.com.cn>>; 王颢 
> <howard_w...@realsil.com.cn<mailto:howard_w...@realsil.com.cn>>; Ruifeng Wang 
> <ruifeng.w...@arm.com<mailto:ruifeng.w...@arm.com>>; nd 
> <n...@arm.com<mailto:n...@arm.com>>; nd <n...@arm.com<mailto:n...@arm.com>>; 
> Hau <h...@realtek.com<mailto:h...@realtek.com>>
> Subject: 答复: about RTL8168 PMD on ARM SoC
> 
> Hi Honnappa,
> 
> I’m sorry I have to make a little correction,
> This issue was actually on our new 2.5 Gigabit RTL8125 series, the attachment 
> is latest r8125pmd,
> There is no essential difference w.r.t ARM platform though.
> 
> BRs,
> Xing Wang
> 
> 发件人: 王星
> 发送时间: 2022年8月26日 10:06
> 收件人: 'Honnappa Nagarahalli' 
> <honnappa.nagaraha...@arm.com<mailto:honnappa.nagaraha...@arm.com>>; 
> dev@dpdk.org<mailto:dev@dpdk.org>
> 抄送: 陈立 <dali_c...@realsil.com.cn<mailto:dali_c...@realsil.com.cn>>; 王颢 
> <howard_w...@realsil.com.cn<mailto:howard_w...@realsil.com.cn>>; Ruifeng Wang 
> <ruifeng.w...@arm.com<mailto:ruifeng.w...@arm.com>>; nd 
> <n...@arm.com<mailto:n...@arm.com>>; nd <n...@arm.com<mailto:n...@arm.com>>; 
> Hau <h...@realtek.com<mailto:h...@realtek.com>>
> 主题: 答复: about RTL8168 PMD on ARM SoC
> 
> Hi Honnappa,
> 
> The attachment is our current r8168pmd code for RTL8111/8168 Giga series
> (currently 8111G, 8111H are supported, we will add others later)
> I will contact the SoC vendor to consult you about this issue and let you 
> know some details about that SoC
> Thanks a lot!
> 
> BRs
> Xing Wang
> 发件人: Honnappa Nagarahalli [mailto:honnappa.nagaraha...@arm.com]
> 发送时间: 2022年8月25日 22:41
> 收件人: 王星 <xing_w...@realsil.com.cn<mailto:xing_w...@realsil.com.cn>>; 
> dev@dpdk.org<mailto:dev@dpdk.org>
> 抄送: 陈立 <dali_c...@realsil.com.cn<mailto:dali_c...@realsil.com.cn>>; 王颢 
> <howard_w...@realsil.com.cn<mailto:howard_w...@realsil.com.cn>>; Ruifeng Wang 
> <ruifeng.w...@arm.com<mailto:ruifeng.w...@arm.com>>; nd 
> <n...@arm.com<mailto:n...@arm.com>>; nd <n...@arm.com<mailto:n...@arm.com>>
> 主题: RE: about RTL8168 PMD on ARM SoC
> 
> Hello,
>                I cannot find many details of the SoC on the internet. Does it 
> use coherent IO? Depending on that, different barriers might be needed. Other 
> than this, I would not think it needs anything special.
> 
> If you could send an RFC to the DPDK mailing list, I am happy to review and 
> provide any feedback.
> 
> Thanks,
> Honnappa
> 
> 
> From: 王星 <xing_w...@realsil.com.cn<mailto:xing_w...@realsil.com.cn>>
> Sent: Wednesday, August 24, 2022 9:53 PM
> To: dev@dpdk.org<mailto:dev@dpdk.org>
> Cc: 陈立 <dali_c...@realsil.com.cn<mailto:dali_c...@realsil.com.cn>>; 王颢 
> <howard_w...@realsil.com.cn<mailto:howard_w...@realsil.com.cn>>
> Subject: about RTL8168 PMD on ARM SoC
> 
> Hi DPDK,
> 
> I am a pmd driver developer from Realtek NIC department,
> when I was porting r8168pmd already verified on x86 to an ARM64 SoC Unisoc: 
> UIS8650
> I found that after NIC Rx init (in general, Rx ring and buffers should have 
> been prepared for NIC to DMA read),
> the NIC status reg showed RDU (Rx Descriptor Unavailable), which means NIC 
> cannot read the proper desc content,
> 
> later I sended some packets to NIC hold by testpmd rx_only mode, HW internal 
> Rx packet counter can grow to some value, then stuck, 8168pmd Rx debug print 
> reported it received less packets than that value, and the print showed up 
> even some minutes later!
> 
> I doubt the phenomenon is caused by improper HW-based IO coherency support on 
> this ARM SoC,
> I have read the ARM SoC support list on DPDK website, to name it: NV 
> Bluefield, NXP DPAA, Marvell Octeon TX …
> 
> Does DPDK (or UIO/VFIO driver or hugetlb driver) need special HW IO cache 
> coherency support on ARM platform, say, ACE and Device side MMU etc?
> Should the SoC provide specialized UIO/VFIO driver or hugetlb driver and/or 
> specific DPDK lib to support such user mode DMA?
> Will you please give suggestions, thanks a lot!
> 
> BRs
> 
> ------Please consider the environment before printing this e-mail.
>