From: "Yin, Fengwei" <fengwei.yin@intel.com>
To: "willy@infradead.org" <willy@infradead.org>,
"surenb@google.com" <surenb@google.com>
Cc: "linux-mm@kvack.org" <linux-mm@kvack.org>,
"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
"Agrawal, Punit" <punit.agrawal@bytedance.com>
Subject: Re: [PATCH 6/6] mm: Run the fault-around code under the VMA lock
Date: Mon, 10 Apr 2023 04:53:19 +0000 [thread overview]
Message-ID: <1c700db59114617ca0a7b6e40754a6ea0dbb86e0.camel@intel.com> (raw)
In-Reply-To: <ZCxA+DYkzVWbLAod@casper.infradead.org>
On Tue, 2023-04-04 at 16:23 +0100, Matthew Wilcox wrote:
> On Tue, Apr 04, 2023 at 02:58:50PM +0100, Matthew Wilcox (Oracle)
> wrote:
> > The map_pages fs method should be safe to run under the VMA lock
> > instead
> > of the mmap lock. This should have a measurable reduction in
> > contention
> > on the mmap lock.
>
> https://github.com/antonblanchard/will-it-scale/pull/37/files should
> be a good microbenchmark to report numbers from. Obviously real-
> world
> benchmarks will be more compelling.
>
Test result in my side with page_fault4 of will-it-scale in thread
mode is:
15274196 (without the patch) -> 17291444 (with the patch)
13.2% improvement on a Ice Lake with 48C/96T + 192G RAM + ext4
filesystem.
The perf showed the mmap_lock contention reduced a lot:
(Removed the grandson functions of do_user_addr_fault())
latest linux-next with the patch:
51.78%--do_user_addr_fault
|
|--49.09%--handle_mm_fault
|--1.19%--lock_vma_under_rcu
--1.09%--down_read
latest linux-next without the patch:
73.65%--do_user_addr_fault
|
|--28.65%--handle_mm_fault
|--17.22%--down_read_trylock
|--10.92%--down_read
|--9.20%--up_read
--7.30%--find_vma
My understanding is down_read_trylock, down_read and up_read all are
related with mmap_lock. So the mmap_lock contention reduction is quite
obvious.
For functional testing, our 0day already cherry-picked the patchset and
the testing is ongoing. If any problem hit during testing, 0day will
report it out. Thanks.
Regards
Yin, Fengwei
next prev parent reply other threads:[~2023-04-10 4:53 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-04 13:58 [PATCH 0/6] Avoid the mmap lock for fault-around Matthew Wilcox (Oracle)
2023-04-04 13:58 ` [PATCH 1/6] mm: Allow per-VMA locks on file-backed VMAs Matthew Wilcox (Oracle)
2023-04-07 17:54 ` Suren Baghdasaryan
2023-04-07 20:12 ` Matthew Wilcox
2023-04-07 20:26 ` Suren Baghdasaryan
2023-04-07 21:36 ` Matthew Wilcox
2023-04-07 22:40 ` Suren Baghdasaryan
2023-04-07 22:49 ` Suren Baghdasaryan
2023-04-14 18:02 ` Suren Baghdasaryan
2023-04-04 13:58 ` [PATCH 2/6] mm: Move FAULT_FLAG_VMA_LOCK check from handle_mm_fault() Matthew Wilcox (Oracle)
2023-04-04 13:58 ` [PATCH 3/6] mm: Move FAULT_FLAG_VMA_LOCK check into handle_pte_fault() Matthew Wilcox (Oracle)
2023-04-04 13:58 ` [PATCH 4/6] mm: Move FAULT_FLAG_VMA_LOCK check down in handle_pte_fault() Matthew Wilcox (Oracle)
2023-04-04 13:58 ` [PATCH 5/6] mm: Move the FAULT_FLAG_VMA_LOCK check down from do_pte_missing() Matthew Wilcox (Oracle)
2023-04-04 13:58 ` [PATCH 6/6] mm: Run the fault-around code under the VMA lock Matthew Wilcox (Oracle)
2023-04-04 15:23 ` Matthew Wilcox
2023-04-07 18:20 ` Suren Baghdasaryan
2023-04-10 4:53 ` Yin, Fengwei [this message]
2023-04-10 14:11 ` Matthew Wilcox
2023-04-12 7:35 ` 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=1c700db59114617ca0a7b6e40754a6ea0dbb86e0.camel@intel.com \
--to=fengwei.yin@intel.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=punit.agrawal@bytedance.com \
--cc=surenb@google.com \
--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;
as well as URLs for NNTP newsgroup(s).