Re: [PATCH 6/7] generic/774: reduce file size

[Christoph Hellwig <hch@infradead.org>]

On Tue, Nov 11, 2025 at 09:13:27AM +0000, John Garry wrote:
> On 10/11/2025 18:27, Darrick J. Wong wrote:
> > From: Darrick J. Wong <djwong@kernel.org>
> > 
> > We've gotten complaints about this test taking hours to run and
> > producing stall warning on test VMs with a large number of cpu cores.  I
> > think this is due to the maximum atomic write unit being very large on
> > XFS where we can fall back to a software-based out of place write
> > implementation.
> > 
> > On the victim machine, the atomic write max is 4MB and there are 24
> > CPUs.  As a result, aw_bsize to be 1MB, so the file size is
> > 1MB * 24 * 2 * 100 == 4.8GB.  I set up a test machine with fast storage
> > and 24 CPUs, and the atomic writes poked along at 25MB/s and the total
> > runtime was 300s.  On spinning rust those stats will be much worse.
> > 
> > Let's try backing the file size off by 10x and see if that eases the
> > complaints.
> > 
> 
> The awu max for xfs is still unbounded (so the file size could still be
> huge). For ext4, it is limited by HW constraints - the largest HW awu max I
> heard about is 256KB. How about also limiting awu max to something sane,
> like 1MB?

Sounds fine to ne, as long as we document it as an arbitrary limit.