[Drbd-dev] [PATCH v5 01/11] block: make generic_make_request handle arbitrarily sized bios

[Mike Snitzer]

On Tue, Aug 11 2015 at  1:36pm -0400,
Martin K. Petersen <martin.petersen at oracle.com> wrote:

> >>>>> "Mike" == Mike Snitzer <snitzer at redhat.com> writes:
> 
> Mike> DM-thinp processes discards internally before it passes them down
> Mike> (if configured to do so).  If a discard is smaller than the
> Mike> granularity of a thinp block (whose size is configurable) or if
> Mike> the start and end of the discard's extent is misaligned (relative
> Mike> to the thinp blocks mapped to the logical extent) then the discard
> Mike> won't actually discard partial thinp blocks.
> 
> That's fine. You can throw away anything you don't like as long as
> discard_zeroes_data=0.

Correct, dm-thinp sets discard_zeroes_data=0

> But I don't understand why having an artificial cap at 2GB fixes
> things. Other than making it less likely for you to receive a runt by
> virtue of being aligned to a power of two.

That is the benefit.  And when coupled with the new default max_discard
of 64K (pending change from Jens for 4.3) this 2GB upper limit really
isn't such a big deal.  Unless I'm missing something...