Re: [PATCH 4.4 72/88] block: dont deal with discard limit in blkdev_issue_discard()

From: Mike Snitzer
Date: Fri Jan 11 2019 - 10:23:25 EST


On Fri, Jan 11 2019 at 10:17am -0500,
Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:

> On Fri, Jan 11, 2019 at 10:06:05AM -0500, Mike Snitzer wrote:
> > On Fri, Jan 11 2019 at 9:35am -0500,
> > Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> >
> > > On Fri, Jan 11, 2019 at 09:25:39AM -0500, Mike Snitzer wrote:
> > > > On Fri, Jan 11 2019 at 9:08am -0500,
> > > > Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> > > >
> > > > > 4.4-stable review patch. If anyone has any objections, please let me know.
> > > > >
> > > > > ------------------
> > > > >
> > > > > From: Ming Lei <ming.lei@xxxxxxxxxx>
> > > > >
> > > > > commit 744889b7cbb56a64f957e65ade7cb65fe3f35714 upstream.
> > > >
> > > > Hi Greg,
> > > >
> > > > Please also pick up this commit:
> > > > 89f5fa47476eda56402e29fff3c5097f5c2a1e19 ("dm: call blk_queue_split() to
> > > > impose device limits on bios")
> > >
> > > That's going to be hard as the dependancy for that patch is not here in
> > > 4.4.y, and this patch itself isn't even in anything older than 4.19.y.
> >
> > Right, I quickly replied to this thread with followup of the 3 prereq
> > patches needed to get 89f5fa47476 to apply.
> >
> > > So why add it here to 4.4.y only?
> >
> > Because you're looking to pull in a commit into 4.4 that causes problems
> > elsewhere.
> >
> > > Can you send the needed patch series to the stable@ mailing list for the
> > > different stable trees if this needs to get into them?
> >
> > I'll try to get to that, but it is low priority for me. And in the
> > meantime.. DM will be broken in 4.4 if you take 744889b7cb.. wheeeee.
>
> Ok, then I should drop this patch from 4.4, I can do that. Looks like
> it's not in 4.9 either, so that's another good reason to not take it
> here as well.
>
> Any objection to just dropping it?

I'd prefer that for sure. I don't think the empty discard issue that
commit 744889b7cbb56a64f957e65ade7cb65fe3f35714 addresses is worth the
other DM breakage. But I'm biased ;)

Mike