netdev - Re: [PATCH] net: ethernet: ti: ale: ensure vlan/mdb deleted when no members

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:   Thu, 28 Nov 2019 11:38:04 +0200
From:   Ilias Apalodimas <ilias.apalodimas@...aro.org>
To:     Grygorii Strashko <grygorii.strashko@...com>
Cc:     netdev@...r.kernel.org, "David S . Miller" <davem@...emloft.net>,
        Andrew Lunn <andrew@...n.ch>, Sekhar Nori <nsekhar@...com>,
        Ivan Khoronzhuk <ivan.khoronzhuk@...aro.org>,
        linux-kernel@...r.kernel.org, linux-omap@...r.kernel.org
Subject: Re: [PATCH] net: ethernet: ti: ale: ensure vlan/mdb deleted when no
 members

On Thu, Nov 28, 2019 at 10:21:27AM +0200, Ilias Apalodimas wrote:
> On Wed, Nov 27, 2019 at 05:59:05PM +0200, Grygorii Strashko wrote:
> > The recently updated ALE APIs cpsw_ale_del_mcast() and
> > cpsw_ale_del_vlan_modify() have an issue and will not delete ALE entry even
> > if VLAN/mcast group has no more members. Hence fix it here and delete ALE
> > entry if !port_mask.
> > 
> > The issue affected only new cpsw switchdev driver.
> > 
> > Fixes: e85c14370783 ("net: ethernet: ti: ale: modify vlan/mdb api for switchdev")
> > Signed-off-by: Grygorii Strashko <grygorii.strashko@...com>
> > ---
> >  drivers/net/ethernet/ti/cpsw_ale.c | 14 ++++++++++----
> >  1 file changed, 10 insertions(+), 4 deletions(-)
> > 
> > diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c
> > index 929f3d3354e3..a5179ecfea05 100644
> > --- a/drivers/net/ethernet/ti/cpsw_ale.c
> > +++ b/drivers/net/ethernet/ti/cpsw_ale.c
> > @@ -396,12 +396,14 @@ int cpsw_ale_del_mcast(struct cpsw_ale *ale, const u8 *addr, int port_mask,
> >  	if (port_mask) {
> >  		mcast_members = cpsw_ale_get_port_mask(ale_entry,
> >  						       ale->port_mask_bits);
> > -		mcast_members &= ~port_mask;
> > -		cpsw_ale_set_port_mask(ale_entry, mcast_members,
> > +		port_mask = mcast_members & ~port_mask;
> > +	}
> > +
> > +	if (port_mask)
> > +		cpsw_ale_set_port_mask(ale_entry, port_mask,
> >  				       ale->port_mask_bits);
> > -	} else {
> > +	else
> >  		cpsw_ale_set_entry_type(ale_entry, ALE_TYPE_FREE);
> > -	}
> 
> The code assumed calls cpsw_ale_del_mcast() should have a port mask '0' when
> deleting an entry. Do we want to have 'dual' functionality on it? 
> This will delete mcast entries if port mask is 0 or port mask matches exactly
> what's configured right?
> 

Deleting the ALE entry if the port_mask matches execlty what's configured makes
sense. Can we change it to something that doesn't change the function argument?

I think something like: 
mcast_members = 0;
if (port_mask) {
	mcast_members = cpsw_ale_get_port_mask(ale_entry,
											ale->port_mask_bits);
	mcast_members &= ~port_mask;
}
if (mcast_members)
	cpsw_ale_set_port_mask(ale_entry, mcast_members, ....)
else
	cpsw_ale_set_entry_type(....)

is more readable?

Thanks
/Ilias