[dm-devel] [RFC 0/3] Add support of iopoll for dm device

[Mike Snitzer]

On Sun, Nov 01 2020 at 10:14pm -0500,
JeffleXu <jefflexu at linux.alibaba.com> wrote:

> 
> On 10/27/20 2:53 AM, Mike Snitzer wrote:
> >What you detailed there isn't properly modeling what it needs to.
> >A given dm_target_io could result in quite a few bios (e.g. for
> >dm-striped we clone each bio for each of N stripes).  So the fan-out,
> >especially if then stacked on N layers of stacked devices, to all the
> >various hctx at the lowest layers is like herding cats.
> >
> >But the recursion in block core's submit_bio path makes that challenging
> >to say the least.  So much so that any solution related to enabling
> >proper bio-based IO polling is going to need a pretty significant
> >investment in fixing block core (storing __submit_bio()'s cookie during
> >recursion, possibly storing to driver provided memory location,
> >e.g. DM initialized bio->submit_cookie pointer to a blk_qc_t within a DM
> >clone bio's per-bio-data).
> >
> >SO __submit_bio_noacct would become:
> >
> >    retp = &ret;
> >    if (bio->submit_cookie)
> >           retp = bio->submit_cookie;
> >    *retp = __submit_bio(bio);
> 
> Sorry for the late reply. Exactly I missed this point before. IF you
> have not started working on this, I'd like to try to implement this as
> an RFC.

I did start on this line of development but it needs quite a bit more
work.  Even the pseudo code I provided above isn't useful in the context
of DM clone bios that have their own per-bio-data to assist with this
implementation.  Because the __submit_bio_noacct() recursive call
drivers/md/dm.c:__split_and_process_bio() makes is supplying the
original bio (modified to only point to remaining work).

But sure, I'm not opposed to you carrying this line of work forward.  I
can always lend a hand if you need help later or if you need to hand it
off to me.

> >I think you probably just got caught out by the recursive nature of the bio
> >submission path -- makes creating a graph of submitted bios and their
> >associated per-bio-data and generated cookies a mess to track (again,
> >like herding cats).
> >
> >Could also be you didn't quite understand the DM code's various
> >structures.
> >
> >In any case, the block core changes needed to make bio-based IO polling
> >work is the main limiting factor right now.
>
> Yes the logic is kind of subtle and maybe what I'm concerned here is
> really should be concerned at the coding phase.

Definitely, lots of little details and associations.

Mike