Linux-ARM-Kernel Archive on
 help / color / Atom feed
From: Linus Torvalds <>
To: Waiman Long <>
Cc: linux-arch <>,,
	Davidlohr Bueso <>,, Tim Chen <>,
	Arnd Bergmann <>,
	Linux-sh list <>,
	Peter Zijlstra <>,,
	the arch/x86 maintainers <>,
	Will Deacon <>,
	Linux List Kernel Mailing <>,
	Ingo Molnar <>, Borislav Petkov <>,
	"H. Peter Anvin" <>,,,
	Thomas Gleixner <>,,
	Andrew Morton <>,
Subject: Re: [PATCH v2 2/2] locking/rwsem: Optimize down_read_trylock()
Date: Tue, 12 Feb 2019 11:58:03 -0800
Message-ID: <> (raw)
In-Reply-To: <>

On Mon, Feb 11, 2019 at 11:31 AM Waiman Long <> wrote:
> Modify __down_read_trylock() to make it generate slightly better code
> (smaller and maybe a tiny bit faster).

This looks good, but I would ask you to try one slightly different approach.

Instead of this:

>        long tmp = atomic_long_read(&sem->count);
>        while (tmp >= 0) {
>                if (atomic_long_try_cmpxchg_acquire(&sem->count, &tmp,
>                                        tmp + RWSEM_ACTIVE_READ_BIAS)) {
>                        return 1;
>                }
>        }

try doing this instead:

        long tmp = 0;

        do {
                if (atomic_long_try_cmpxchg_acquire(&sem->count, &tmp,
                                        tmp + RWSEM_ACTIVE_READ_BIAS)) {
                        return 1;
        } while (tmp >= 0);
        return 0;

because especially when it comes to locking, it's usually better to
just *guess* that the lock is unlocked, than it is to actually read
from the line to see what the state is.

Often - but certainly not always - the lock is the first access to the
target cacheline, and assuming the trylock is successful (which I
think is the case we want to optimize for), we're much better off
causing that first access to be a read-for-ownership, rather than a

Because if you first read from the line, and then do a cmpxchg, and if
the line was not in the cache, your cache coherency protocol will
generally go through two states: first shared (for the initial read)
and then exclusive-dirty (for the cmpxchg).

Now, this is obviously very micro-architecture dependent, and in fact
the microarchitecture could even see the "predict fallthrough to a
cmpxchg with the same address" and turn the first read into a
read-for-ownership, but we've done this at some point before, and the
"guess unlocked" was actually the one that performed better.

Of course, the downside is that it might be worse when the guess is
incorrect - either because of a nested read lock or due to an actual
conflict with a write, but on the whole those *should* be the rare
cases, and not the cases where we necessarily optimize for latency of
the operation.



linux-arm-kernel mailing list

  parent reply index

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-02-11 19:31 [PATCH v2 0/2] locking/rwsem: Remove arch specific rwsem files Waiman Long
2019-02-11 19:31 ` [PATCH v2 1/2] " Waiman Long
2019-02-11 19:31 ` [PATCH v2 2/2] locking/rwsem: Optimize down_read_trylock() Waiman Long
2019-02-12 13:24   ` Peter Zijlstra
2019-02-12 13:25     ` Peter Zijlstra
2019-02-12 18:36       ` Waiman Long
2019-02-12 18:38         ` Waiman Long
2019-02-12 19:58   ` Linus Torvalds [this message]
2019-02-12 21:21     ` Waiman Long
2019-02-13  7:45       ` Ingo Molnar
2019-02-13 15:33         ` Waiman Long

Reply instructions:

You may reply publically 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:

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

  git send-email \
    --in-reply-to='' \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

Linux-ARM-Kernel Archive on

Archives are clonable:
	git clone --mirror linux-arm-kernel/git/0.git
	git clone --mirror linux-arm-kernel/git/1.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-arm-kernel linux-arm-kernel/ \
	public-inbox-index linux-arm-kernel

Example config snippet for mirrors

Newsgroup available over NNTP:

AGPL code for this site: git clone