From: Peter Xu <peterx@redhat.com>
To: Ryan Roberts <ryan.roberts@arm.com>
Cc: David Hildenbrand <david@redhat.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
linux-doc@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>,
Jonathan Corbet <corbet@lwn.net>,
Mike Kravetz <mike.kravetz@oracle.com>,
Hugh Dickins <hughd@google.com>,
"Matthew Wilcox (Oracle)" <willy@infradead.org>,
Yin Fengwei <fengwei.yin@intel.com>,
Yang Shi <shy828301@gmail.com>, Zi Yan <ziy@nvidia.com>
Subject: Re: [PATCH mm-unstable v1] mm: add a total mapcount for large folios
Date: Thu, 10 Aug 2023 13:15:32 -0400 [thread overview]
Message-ID: <ZNUbNDiciFefJngZ@x1n> (raw)
In-Reply-To: <155bd03e-b75c-4d2d-a89d-a12271ada71b@arm.com>
On Thu, Aug 10, 2023 at 11:48:27AM +0100, Ryan Roberts wrote:
> > For PTE-mapped THP, it might be a bit bigger noise, although I doubt it is
> > really significant (judging from my experience on managing PageAnonExclusive
> > using set_bit/test_bit/clear_bit when (un)mapping anon pages).
> >
> > As folio_add_file_rmap_range() indicates, for PTE-mapped THPs we should be
> > batching where possible (and Ryan is working on some more rmap batching).
>
> Yes, I've just posted [1] which batches the rmap removal. That would allow you
> to convert the per-page atomic_dec() into a (usually) single per-large-folio
> atomic_sub().
>
> [1] https://lore.kernel.org/linux-mm/20230810103332.3062143-1-ryan.roberts@arm.com/
Right, that'll definitely make more sense, thanks for the link; I'd be very
happy to read more later (finally I got some free time recently..). But
then does it mean David's patch can be attached at the end instead of
proposed separately and early?
I was asking mostly because I read it as a standalone patch first, and
honestly I don't know the effect. It's based on not only the added atomic
ops itself, but also the field changes.
For example, this patch moves Hugh's _nr_pages_mapped into the 2nd tail
page, I think it means for any rmap change of any small page of a huge one
we'll need to start touching one more 64B cacheline on x86. I really have
no idea what does it mean for especially a large SMP: see 292648ac5cf1 on
why I had an impression of that. But I've no enough experience or clue to
prove it a problem either, maybe would be interesting to measure the time
needed for some pte-mapped loops? E.g., something like faulting in a thp,
then measure the split (by e.g. mprotect() at offset 1M on a 4K?) time it
takes before/after this patch.
When looking at this, I actually found one thing that is slightly
confusing, not directly relevant to your patch, but regarding the reuse of
tail page 1 on offset 24 bytes. Current it's Hugh's _nr_pages_mapped,
and you're proposing to replace it with the total mapcount:
atomic_t _nr_pages_mapped; /* 88 4 */
Now my question is.. isn't byte 24 of tail page 1 used for keeping a
poisoned mapping? See prep_compound_tail() where it has:
p->mapping = TAIL_MAPPING;
While here mapping is, afaict, also using offset 24 of the tail page 1:
struct address_space * mapping; /* 24 8 */
I hope I did a wrong math somewhere, though.
--
Peter Xu
next prev parent reply other threads:[~2023-08-10 17:15 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-09 8:32 David Hildenbrand
2023-08-09 15:45 ` Zi Yan
2023-08-09 19:07 ` Ryan Roberts
2023-08-09 19:17 ` David Hildenbrand
2023-08-10 10:40 ` Ryan Roberts
2023-08-10 11:14 ` David Hildenbrand
2023-08-10 11:27 ` David Hildenbrand
2023-08-10 11:32 ` David Hildenbrand
2023-08-10 11:35 ` Ryan Roberts
2023-08-09 19:21 ` Matthew Wilcox
2023-08-09 19:26 ` David Hildenbrand
2023-08-10 3:14 ` Yin Fengwei
2023-08-09 21:23 ` Peter Xu
2023-08-10 3:25 ` Matthew Wilcox
2023-08-10 8:37 ` David Hildenbrand
2023-08-10 21:48 ` Peter Xu
2023-08-10 21:54 ` Matthew Wilcox
2023-08-10 21:59 ` David Hildenbrand
2023-08-11 15:03 ` Peter Xu
2023-08-11 15:14 ` Zi Yan
2023-08-11 15:17 ` David Hildenbrand
2023-08-10 8:59 ` David Hildenbrand
2023-08-10 10:48 ` Ryan Roberts
2023-08-10 17:15 ` Peter Xu [this message]
2023-08-10 17:47 ` David Hildenbrand
2023-08-10 19:02 ` Ryan Roberts
2023-08-10 20:57 ` Peter Xu
2023-08-10 21:48 ` Matthew Wilcox
2023-08-10 22:27 ` David Hildenbrand
2023-08-11 15:18 ` Peter Xu
2023-08-11 15:32 ` David Hildenbrand
2023-08-11 15:58 ` Peter Xu
2023-08-11 16:08 ` David Hildenbrand
2023-08-11 16:11 ` Zi Yan
2023-08-11 22:18 ` Peter Xu
2023-08-10 22:16 ` David Hildenbrand
2023-08-10 3:24 ` Yin Fengwei
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=ZNUbNDiciFefJngZ@x1n \
--to=peterx@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=corbet@lwn.net \
--cc=david@redhat.com \
--cc=fengwei.yin@intel.com \
--cc=hughd@google.com \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mike.kravetz@oracle.com \
--cc=ryan.roberts@arm.com \
--cc=shy828301@gmail.com \
--cc=willy@infradead.org \
--cc=ziy@nvidia.com \
/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