Re: [PATCH RFC 09/22] block, cfq: replace CFQ with the BFQ-v0 I/O scheduler

[Austin S. Hemmelgarn]

On 2016-03-04 12:39, Christoph Hellwig wrote:
> On Sat, Mar 05, 2016 at 12:29:39AM +0700, Linus Walleij wrote:
> > Hi Tejun,
> >
> > I'm doing a summary of this discussion as a part of presenting
> > Linaro's involvement in Paolo's work. So I try to understand things.
>
> Btw, can someone explain why you guys waste so much time hacking and
> arguing about a legacy codebase (old request code and I/O schedulers)
> that everyone would really like to see disappear.  Why don't you
> spend your time on blk-mq where you have an entirely clean slate
> for scheduling?
1. This all started long before blk-mq hit mainline.
2. There's still a decent amount of block drivers that don't support 
blk-mq.  Last time I looked (around the time 4.4 came out), I saw the 
following that either obviously don't support it, or are ambiguous as to 
whether they support it or not.  Here's a list of just the ones I know 
are being used on existing systems running relatively recent kernel 
versions, not including any of the MTD stuff:
    * fd
    * MD
    * bcache
    * mmcblk
    * nbd
    * dasd
    * drbd
    * rbd
    * aoe
    * xvd (I know there were patches for this floating around, but I 
never saw if they got merged or not)