On Thu, Aug 30, 2012 at 06:07:45PM -0400, Vivek Goyal wrote:

On Wed, Aug 29, 2012 at 10:13:45AM -0700, Kent Overstreet wrote:

[..]

quoted

quoted

Performance aside, punting submission to per device worker in case of deep
stack usage sounds cleaner solution to me.

Agreed, but performance tends to matter in the real world. And either
way the tricky bits are going to be confined to a few functions, so I
don't think it matters that much.

If someone wants to code up the workqueue version and test it, they're
more than welcome...

Here is one quick and dirty proof of concept patch. It checks for stack
depth and if remaining space is less than 20% of stack size, then it
defers the bio submission to per queue worker.

Given that we are working around stack depth issues in the
filesystems already in several places, and now it seems like there's
a reason to work around it in the block layers as well, shouldn't we
simply increase the default stack size rather than introduce
complexity and performance regressions to try and work around not
having enough stack?

I mean, we can deal with it like the ia32 4k stack issue was dealt
with (i.e. ignore those stupid XFS people, that's an XFS bug), or
we can face the reality that storage stacks have become so complex
that 8k is no longer a big enough stack for a modern system....

Cheers,

Dave.
-- 
Dave Chinner
david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org