Re: better block swap batching and a different take on swap_ops v2

[Baoquan He <baoquan.he@linux.dev>]

On 06/01/26 at 04:50pm, Christoph Hellwig wrote:
> On Mon, Jun 01, 2026 at 09:29:30PM +0800, Baoquan He wrote:
> > > this series makes use of the swap_iocb for block as well so that it
> > > doesn't do inefficient single-bio I/O, and then rebases the swap_ops
> > > from Baoquan on top of the now very different method structure.
> > 
> > What tree is this series based on? I tried the latest mainline,
> > linux-next main branch and Andrew's mm-unstable, all failed. 
> 
> The base is:
> 
> commit e1af79f3291a268adf4e149e1faba3052743e898 (akpm/mm-unstable)
> Author: Joshua Hahn <joshua.hahnjy@gmail.com>
> Date:   Fri May 29 13:27:54 2026 -0700
> 
>     mm/nodemask: correctly describe nodemask operation return types

Thanks. And seems the patchset pulled back via b4 only contains the
patch 1~5/8 and patch 8/8, it misses 6~7/8. Just in case if someone else
also use b4 and meets the similar issue when applying patches. This
could be related to the email sending tool or its settings.

> 
> 
> > And by the way, usually we add version number in each patch's subject as
> > "[PATCH v2] xxx ".
> 
> That does sometimes happen but is rather unusual if you take a quick
> look over lkml.  It also isn't the default for any normal patch
> sending tool.

Really, what I have seen are all this kind of format, but surely I could
see too few. Anyway, it doesn't matter much. Thanks.