Re: [PATCH 5/5] dm thin: unroll issue_discard() to create longer discard bio chains

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri, May 06, 2016 at 06:12:16PM +0200, Christoph Hellwig wrote:
> On Thu, May 05, 2016 at 11:54:25AM -0400, Mike Snitzer wrote:
> > From: Joe Thornber <ejt@xxxxxxxxxx>
> > 
> > There is little benefit to doing this but it does structure DM thinp's
> > code to more cleanly use the __blkdev_issue_discard() interface --
> > particularly in passdown_double_checking_shared_status().
> 
> As-is I think it makes the code a whole lot more convoluted, and especially
> the structure that's just used to communicated 4 parameters between
> functions which don't even use all of them isn't very helpful.

I started coding it inline, and then factored it out because I thought
it was cleaner.  I'm also likely to lift it into a separate file for
use in dm-cache too.

> This is a useful helper I think, but passing tc and bio directly would
> make it even more obvious.

y, this is exactly what I had.

> Both the plugging and the bi_error handling would benefit from being
> moved into process_prepared_discard_passdown and handled in the common
> path.

But the plugging and error handling are called from two places.
process_prepared_discard_passdown() and
passdown_double_checking_shared_status().

- Joe

(I agree about the confusing comment, we'll remove)

--
dm-devel mailing list
dm-devel@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/dm-devel



[Index of Archives]     [DM Crypt]     [Fedora Desktop]     [ATA RAID]     [Fedora Marketing]     [Fedora Packaging]     [Fedora SELinux]     [Yosemite Discussion]     [KDE Users]     [Fedora Docs]

  Powered by Linux