[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <874lr3vzg4.fsf@weeman.i-did-not-set--mail-host-address--so-tickle-me>
Date: Fri, 13 Oct 2017 11:29:15 -0400
From: Vivien Didelot <vivien.didelot@...oirfairelinux.com>
To: David Laight <David.Laight@...LAB.COM>,
'Florian Fainelli' <f.fainelli@...il.com>,
"netdev\@vger.kernel.org" <netdev@...r.kernel.org>
Cc: Andrew Lunn <andrew@...n.ch>,
"kernel\@savoirfairelinux.com" <kernel@...oirfairelinux.com>,
"linux-kernel\@vger.kernel.org" <linux-kernel@...r.kernel.org>,
"David S. Miller" <davem@...emloft.net>
Subject: RE: [PATCH net-next 5/5] net: dsa: split dsa_port's netdev member
Hi again,
Vivien Didelot <vivien.didelot@...oirfairelinux.com> writes:
>>> How about using:
>>>
>>> union {
>>> struct net_device *master;
>>> struct net_device *slave;
>>> } netdev;
>> ...
>>
>> You can remove the 'netdev' all the compilers support unnamed unions.
>
> There are issues with older GCC versions, see the commit 42275bd8fcb3
> ("switchdev: don't use anonymous union on switchdev attr/obj structs")
>
> That's why I kept it in the v2 I sent.
At the same time, I can see that struct sk_buff uses anonym union a lot.
It seems weird that one raised a compiler issue for switchdev but not
for skbuff.h... Do you think it is viable to drop the name here then?
I'd be happy to respin a v3 if this sounds safe.
Thanks,
Vivien
Powered by blists - more mailing lists