netdev - Re: [net-next PATCH v3 01/14] net: dsa: qca8k: cache match data to speed up access

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] [thread-next>] [day] [month] [year] [list]

Date:   Sun, 24 Jul 2022 22:27:13 +0200
From:   Christian Marangi <ansuelsmth@...il.com>
To:     Vladimir Oltean <olteanv@...il.com>
Cc:     Andrew Lunn <andrew@...n.ch>,
        Vivien Didelot <vivien.didelot@...il.com>,
        Florian Fainelli <f.fainelli@...il.com>,
        "David S. Miller" <davem@...emloft.net>,
        Eric Dumazet <edumazet@...gle.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>, Jens Axboe <axboe@...nel.dk>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        linux-kernel@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: [net-next PATCH v3 01/14] net: dsa: qca8k: cache match data to
 speed up access

On Mon, Jul 25, 2022 at 01:30:31AM +0300, Vladimir Oltean wrote:
> On Sat, Jul 23, 2022 at 04:18:32PM +0200, Christian Marangi wrote:
> > Using of_device_get_match_data is expensive. Cache match data to speed
> 
> 'is expensive' sounds like it's terribly expensive. It's just more than
> it needs to be.
>

Ok will reword.

> > up access and rework user of match data to use the new cached value.
> > 
> > Signed-off-by: Christian Marangi <ansuelsmth@...il.com>
> > ---
> >  drivers/net/dsa/qca/qca8k.c | 28 ++++++++++------------------
> >  drivers/net/dsa/qca/qca8k.h |  1 +
> >  2 files changed, 11 insertions(+), 18 deletions(-)
> > 
> > diff --git a/drivers/net/dsa/qca/qca8k.c b/drivers/net/dsa/qca/qca8k.c
> > index 1cbb05b0323f..212b284f9f73 100644
> > --- a/drivers/net/dsa/qca/qca8k.c
> > +++ b/drivers/net/dsa/qca/qca8k.c
> > @@ -3168,6 +3155,11 @@ qca8k_sw_probe(struct mdio_device *mdiodev)
> >  	if (ret)
> >  		return ret;
> >  
> > +	/* Cache match data in priv struct.
> > +	 * Match data is already checked in read_switch_id.
> > +	 */
> > +	priv->info = of_device_get_match_data(priv->dev);
> > +
> 
> So why don't you set priv->info right before calling qca8k_read_switch_id(),
> then?
> 

The idea was to make the read_switch_id a function to check if the
switch is compatible... But yhea now that i think about it doesn't
really make sense.

(Just for reference I just sent v4 as I got a report from kernel test
bot... it's really just this series with a change in 0002 patch that set
the struct for ops as a pointer... didn't encounter this with gcc but it
seems kernel test bot use some special config...)

> >  	priv->ds = devm_kzalloc(&mdiodev->dev, sizeof(*priv->ds), GFP_KERNEL);
> >  	if (!priv->ds)
> >  		return -ENOMEM;
> > diff --git a/drivers/net/dsa/qca/qca8k.h b/drivers/net/dsa/qca/qca8k.h
> > index ec58d0e80a70..0b990b46890a 100644
> > --- a/drivers/net/dsa/qca/qca8k.h
> > +++ b/drivers/net/dsa/qca/qca8k.h
> > @@ -401,6 +401,7 @@ struct qca8k_priv {
> >  	struct qca8k_mdio_cache mdio_cache;
> >  	struct qca8k_pcs pcs_port_0;
> >  	struct qca8k_pcs pcs_port_6;
> > +	const struct qca8k_match_data *info;
> >  };
> >  
> >  struct qca8k_mib_desc {
> > -- 
> > 2.36.1
> > 
> 

-- 
	Ansuel