Re: [PATCH] sata_sil: add Large Block Transfer support

[Tejun Heo <tj@kernel.org>]

Hello, Robert.

Robert Hancock wrote:
> Tejun Heo wrote:
>>
>> ioread/iowrite?
> 
> We know the register's always MMIO on this controller, so it's slightly
> more optimal to avoid the conditional in there.

Yeah, by a small margin but I think the consensus was to use
ioread/write no matter what if it's mapped using iomap.  I don't think
we guarantee iomapped address == mmio address after all.

>> I know it's not specific to this change but does mb() really make
>> sense here?  I don't think we need any barrier here.
> 
> Not sure. Documentation/memory-barriers.txt seems rather unfortunately
> vague on whether MMIO writes are strongly ordered with respect to memory
> writes. I seem to recall some debate on this a while ago, did it ever
> get resolved?

I remember that thread too but don't really any definite conclusion.
Eh... I hate barriers lying around without exact explanation.  :-(

Thanks.

-- 
tejun