Re: Kernel fast memory registration API proposal [RFC]

[Chuck Lever <chuck.lever-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>]

On Jul 14, 2015, at 4:50 AM, Sagi Grimberg <sagig-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org> wrote:

> On 7/13/2015 5:16 PM, Chuck Lever wrote:
> 
>>> NFS really should be using something more similar to a scatterlist,
>>> as it maps pretty well to the sk_frags in the network layer as well.
>>> 
>>> Struct scatterlist is imprtant because it's the way the DMA mapping
>>> functions takes a multi-page argument, so ayone who wants to batch
>>> the S/G mapping calls needs it.
>> 
>> An excellent topic to bring up on linux-nfs-u79uwXL29TaiAVqoAR/hOA@public.gmane.org
>> 
>> In the meantime, I think rpcrdma.ko would have to be responsible for
>> converting struct page to struct scatterlist.
>> 
> 
> Fine by me, so I take it that you are OK with the proposed API?

I think I can make struct scatterlist work as well as struct struct
ib_fast_reg_page_list works now.

If Christoph is correct, the pay-off for this change would be when
the NFS upper layers are converted to use struct scatterlist instead
of struct page.

Until then, I don’t see a specific benefit for RPC/RDMA. But I will
go with the flow.


--
Chuck Lever



--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html