lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Mon, 8 Apr 2024 09:25:39 -0700
From: Breno Leitao <leitao@...ian.org>
To: Eric Dumazet <edumazet@...gle.com>
Cc: kuba@...nel.org, davem@...emloft.net, pabeni@...hat.com,
	linux-kernel@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: [PATCH net-next 3/3] net: dql: Optimize stall information
 population

On Thu, Apr 04, 2024 at 06:36:00PM +0200, Eric Dumazet wrote:
> On Thu, Apr 4, 2024 at 5:00 PM Breno Leitao <leitao@...ian.org> wrote:
> >
> > When Dynamic Queue Limit (DQL) is set, it always populate stall
> > information through dql_queue_stall().  However, this information is
> > only necessary if a stall threshold is set, stored in struct
> > dql->stall_thrs.
> >
> > dql_queue_stall() is cheap, but not free, since it does have memory
> > barriers and so forth.
> >
> > Do not call dql_queue_stall() if there is no stall threshold set, and
> > save some CPU cycles.
> >
> > Signed-off-by: Breno Leitao <leitao@...ian.org>
> > ---
> >  include/linux/dynamic_queue_limits.h | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/include/linux/dynamic_queue_limits.h b/include/linux/dynamic_queue_limits.h
> > index 9980df0b7247..869afb800ea1 100644
> > --- a/include/linux/dynamic_queue_limits.h
> > +++ b/include/linux/dynamic_queue_limits.h
> > @@ -137,7 +137,9 @@ static inline void dql_queued(struct dql *dql, unsigned int count)
> >
> >         dql->num_queued += count;
> >
> > -       dql_queue_stall(dql);
> > +       /* Only populate stall information if the threshold is set */
> > +       if (READ_ONCE(dql->stall_thrs))
> > +               dql_queue_stall(dql);
> 
> Note that this field is not in the first cache line of 'struct dql',
> we will have some false sharing.
> 
> Perhaps we could copy it in a hole of the first cache line (used by producers).

That is a good point. Let me move stall_thrs to the first hole.

Thanks

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ