All of lore.kernel.org
 help / color / mirror / Atom feed
From: "hch@infradead.org" <hch@infradead.org>
To: "Darrick J. Wong" <djwong@kernel.org>
Cc: Dave Chinner <david@fromorbit.com>,
	"hch@infradead.org" <hch@infradead.org>,
	Matthew Wilcox <willy@infradead.org>,
	Trond Myklebust <trondmy@hammerspace.com>,
	"bfoster@redhat.com" <bfoster@redhat.com>,
	"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
	"trondmy@kernel.org" <trondmy@kernel.org>,
	"axboe@kernel.dk" <axboe@kernel.dk>,
	"linux-xfs@vger.kernel.org" <linux-xfs@vger.kernel.org>
Subject: Re: [PATCH] iomap: Address soft lockup in iomap_finish_ioend()
Date: Wed, 5 Jan 2022 05:42:48 -0800	[thread overview]
Message-ID: <YdWgWC8FDwv0z7ze@infradead.org> (raw)
In-Reply-To: <20220104231230.GG31606@magnolia>

On Tue, Jan 04, 2022 at 03:12:30PM -0800, Darrick J. Wong wrote:
> As I see it, the problem here is that we're spending too much time
> calling iomap_finish_page_writeback over and over and over, right?
> 
> If we have a single page with a single mapping that fits in a single
> bio, that means we call bio_add_page once, and on the other end we call
> iomap_finish_page_writeback once.

iomap_finish_page_writeback is called once per page, and the folio
equivalent will be called once per folio, yes.

But usually call bio_add_page mutliple times, due to the silly one block
at a time loop in iomap_writepage_map.  But that is someting we can
easily fix.

  parent reply	other threads:[~2022-01-05 13:42 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-30 19:35 [PATCH] iomap: Address soft lockup in iomap_finish_ioend() trondmy
2021-12-30 21:24 ` Jens Axboe
2021-12-30 22:25   ` Trond Myklebust
2021-12-30 22:27     ` Jens Axboe
2021-12-30 22:55       ` Trond Myklebust
2021-12-31  1:42 ` Matthew Wilcox
2021-12-31  6:16   ` Trond Myklebust
2022-01-01  3:55     ` Dave Chinner
2022-01-01 17:39       ` Trond Myklebust
2022-01-03 22:03         ` Dave Chinner
2022-01-04  0:04           ` Trond Myklebust
2022-01-04  1:22             ` Dave Chinner
2022-01-04  3:01               ` Trond Myklebust
2022-01-04  7:08               ` hch
2022-01-04 18:08                 ` Matthew Wilcox
2022-01-04 18:14                   ` hch
2022-01-04 19:22                     ` Darrick J. Wong
2022-01-04 21:52                       ` Dave Chinner
2022-01-04 23:12                         ` Darrick J. Wong
2022-01-05  2:10                           ` Dave Chinner
2022-01-05 13:56                             ` Brian Foster
2022-01-05 22:04                               ` Dave Chinner
2022-01-06 16:44                                 ` Brian Foster
2022-01-10  8:18                                   ` Dave Chinner
2022-01-10 17:45                                     ` Brian Foster
2022-01-10 18:11                                       ` hch
2022-01-11 14:33                                       ` Trond Myklebust
2022-01-05 13:42                           ` hch [this message]
2022-01-04 21:16                 ` Dave Chinner
2022-01-05 13:43                   ` hch
2022-01-05 22:34                     ` Dave Chinner
2022-01-05  2:09               ` Trond Myklebust
2022-01-05 20:45                 ` Trond Myklebust
2022-01-05 22:48                   ` Dave Chinner
2022-01-05 23:29                     ` Trond Myklebust
2022-01-06  0:01                     ` Darrick J. Wong
2022-01-09 23:09                       ` Dave Chinner
2022-01-06 18:36                     ` Trond Myklebust
2022-01-06 18:38                       ` Trond Myklebust
2022-01-06 20:07                       ` Brian Foster
2022-01-07  3:08                         ` Trond Myklebust
2022-01-07 15:15                           ` Brian Foster
2022-01-09 23:34                       ` Dave Chinner
2022-01-10 23:37                       ` Dave Chinner
2022-01-11  0:08                         ` Dave Chinner
2022-01-13 17:01                         ` Trond Myklebust
2022-01-17 17:24                           ` Trond Myklebust
2022-01-17 17:36                             ` Darrick J. Wong
2022-01-04 13:36         ` Brian Foster
2022-01-04 19:23           ` Darrick J. Wong
2022-01-05  2:31 ` [iomap] f5934dda54: BUG:sleeping_function_called_from_invalid_context_at_fs/iomap/buffered-io.c kernel test robot
2022-01-05  2:31   ` kernel test robot

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=YdWgWC8FDwv0z7ze@infradead.org \
    --to=hch@infradead.org \
    --cc=axboe@kernel.dk \
    --cc=bfoster@redhat.com \
    --cc=david@fromorbit.com \
    --cc=djwong@kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-xfs@vger.kernel.org \
    --cc=trondmy@hammerspace.com \
    --cc=trondmy@kernel.org \
    --cc=willy@infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.