linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: David Hildenbrand <david@redhat.com>
To: Matthew Wilcox <willy@infradead.org>
Cc: linux-mm@kvack.org, linux-block@vger.kernel.org,
	Muchun Song <muchun.song@linux.dev>,
	Jane Chu <jane.chu@oracle.com>,
	Andres Freund <andres@anarazel.de>
Subject: Re: Direct I/O performance problems with 1GB pages
Date: Mon, 27 Jan 2025 17:09:57 +0100	[thread overview]
Message-ID: <f3710cc4-cbbf-4f1e-93a0-9eb6697df2d3@redhat.com> (raw)
In-Reply-To: <Z5euIf-OvrE1suWH@casper.infradead.org>

> 
>> If the workload doing a lot of single-page try_grab_folio_fast(), could it
>> do so on a larger area (multiple pages at once -> single refcount update)?
> 
> Not really.  This is memory that's being used as the buffer cache, so
> every thread in your database is hammering on it and pulling in exactly
> the data that it needs for the SQL query that it's processing.

Ouch.

> 
>> Maybe there is a link to the report you could share, thanks.
> 
> Andres shared some gists, but I don't want to send those to a
> mailing list without permission.  Here's the kernel part of the
> perf report:
> 
>      14.04%  postgres         [kernel.kallsyms]          [k] try_grab_folio_fast
>              |
>               --14.04%--try_grab_folio_fast
>                         gup_fast_fallback
>                         |
>                          --13.85%--iov_iter_extract_pages
>                                    bio_iov_iter_get_pages
>                                    iomap_dio_bio_iter
>                                    __iomap_dio_rw
>                                    iomap_dio_rw
>                                    xfs_file_dio_read
>                                    xfs_file_read_iter
>                                    __io_read
>                                    io_read
>                                    io_issue_sqe
>                                    io_submit_sqes
>                                    __do_sys_io_uring_enter
>                                    do_syscall_64
> 
> Now, since postgres is using io_uring, perhaps there could be a path
> which registers the memory with the iouring (doing the refcount/pincount
> dance once), and then use that pinned memory for each I/O.  Maybe that
> already exists; I'm not keeping up with io_uring development and I can't
> seem to find any documentation on what things like io_provide_buffers()
> actually do.

That's precisely what io-uring fixed buffers do :)

-- 
Cheers,

David / dhildenb



  reply	other threads:[~2025-01-27 16:10 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-26  0:46 Matthew Wilcox
2025-01-27 14:09 ` David Hildenbrand
2025-01-27 16:02   ` Matthew Wilcox
2025-01-27 16:09     ` David Hildenbrand [this message]
2025-01-27 16:20       ` David Hildenbrand
2025-01-27 16:56         ` Matthew Wilcox
2025-01-27 16:59           ` David Hildenbrand
2025-01-27 18:21       ` Andres Freund
2025-01-27 18:54         ` Jens Axboe
2025-01-27 19:07           ` David Hildenbrand
2025-01-27 21:32           ` Pavel Begunkov
2025-01-27 16:24     ` Keith Busch
2025-01-27 17:25   ` Andres Freund
2025-01-27 19:20     ` David Hildenbrand
2025-01-27 19:36       ` Andres Freund
2025-01-28  5:56 ` Christoph Hellwig
2025-01-28  9:47   ` David Hildenbrand
2025-01-29  6:03     ` Christoph Hellwig

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=f3710cc4-cbbf-4f1e-93a0-9eb6697df2d3@redhat.com \
    --to=david@redhat.com \
    --cc=andres@anarazel.de \
    --cc=jane.chu@oracle.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=muchun.song@linux.dev \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox