Re: btrfs I/O completion cleanup and single device I/O optimizations v2

[Christoph Hellwig <hch@lst.de>]

On Tue, Sep 06, 2022 at 03:12:01PM +0200, David Sterba wrote:
> Let me point out one example, the 5/11 "btrfs: remove
> bioc->stripes_pending". The subject gives a false idea what's going on.
> There was a member removed, yes but the whole logic regarding bios is
> changed, it's now using the bio chaining, as mentioned in the first
> paragraph at least. Other patches state in words what the code does, not
> explaining why.

Maybe I'm just too familiar with the block code to not agree with you,
as to me these changes are totally obvious.  But as I often point out:
if an experienced developers asks questions that seem obvious to me
my changelogs are not good enough.  But asking for that after the series
has been out for a month and a half is a bit silly.  Please comment
on these patches in a somewhat timely fashion if you think the
changelogs are not good, and I will update them in a timely fashion.