Re: [PATCH v2 5/5] xfs: replace zero range flush with folio batch

[Christoph Hellwig <hch@infradead.org>]

On Tue, Feb 10, 2026 at 02:19:53PM -0500, Brian Foster wrote:
> This adds a bit more code in the same area as this series. It doesn't
> seem terrible so far, but it was one reason I was wondering if this
> perhaps warranted splitting off its own callback. The behavior for zero
> range here is unique enough from standard read/write ops such that might
> be a readability improvement. Since I'm not the only person with that
> thought, I'll take a proper look and see if it's worth a prototype to go
> along with the mapping behavior change..

Yeah, a separate handler might be cleaner.  Note that in
xfs_buffered_write_iomap_begin, the zeroing special case is also
a fairly large part of the function.

One thing that might help here is my old idea of splitting looking up the
existing mapping, and creating a new mapping for dirtied data into
separate nested iomap iterations, preferably with a way to pass along the
mapping for in-place updates.  That way all this zeroing special casing
would go away, as we'd only perform an action if the read side lookup
returned an extent.  But so far this is just a crazy idea in my mind,
and I'm not really sure how well it would work in practice.