[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <aIt8BYa6Ti6SRh8C@infradead.org>
Date: Thu, 31 Jul 2025 07:21:57 -0700
From: Christoph Hellwig <hch@...radead.org>
To: alexjlzheng@...il.com
Cc: brauner@...nel.org, djwong@...nel.org, willy@...radead.org,
linux-xfs@...r.kernel.org, linux-fsdevel@...r.kernel.org,
linux-kernel@...r.kernel.org,
Jinliang Zheng <alexjlzheng@...cent.com>
Subject: Re: [PATCH v2] iomap: move prefaulting out of hot write path
On Thu, Jul 31, 2025 at 12:44:09AM +0800, alexjlzheng@...il.com wrote:
> From: Jinliang Zheng <alexjlzheng@...cent.com>
>
> Prefaulting the write source buffer incurs an extra userspace access
> in the common fast path. Make iomap_write_iter() consistent with
> generic_perform_write(): only touch userspace an extra time when
> copy_folio_from_iter_atomic() has failed to make progress.
This is probably a good thing to have, but I'm curous if you did see
it making a different for workloads?
> + /*
> + * Faults here on mmap()s can recurse into arbitrary
> + * filesystem code. Lots of locks are held that can
> + * deadlock. Use an atomic copy to avoid deadlocking
> + * in page fault handling.
We can and should use all 80 characters in a line for comments.
> + /*
> + * 'folio' is now unlocked and faults on it can be
> + * handled. Ensure forward progress by trying to
> + * fault it in now.
> + */
Same here.
I really wish we could find a way to share the core write loop between
at least iomap and generic_perform_write and maybe also the other copy
and pasters. But that's for another time..
Powered by blists - more mailing lists