From mboxrd@z Thu Jan 1 00:00:00 1970 From: AndyLiebman@aol.com Subject: Re: Frequent Oops on Shutdown 2.6.10 Date: Wed, 23 Feb 2005 09:51:44 EST Message-ID: Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="-----------------------------1109170304" Cc: terryg@axian.com, netdev@oss.sgi.com, davem@davemloft.net, akpm@osdl.org, AndyLiebman@aol.com To: herbert@gondor.apana.org.au, yoshfuji@linux-ipv6.org Sender: netdev-bounce@oss.sgi.com Errors-to: netdev-bounce@oss.sgi.com List-Id: netdev.vger.kernel.org -------------------------------1109170304 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit Hello, Yoshi Should I bother to apply this patch, or should I wait for you to make this last change? What did you think about my comment that the Oops only occurred when the Ethernet cable had been unplugged during operation? Regards, Andy Liebman On Wed, Feb 23, 2005 at 06:35:55PM +0900, YOSHIFUJI Hideaki / ?$B5HF#1QL@ wrote: > > What do you think of this? Thanks, this looks great. There is just one technical detail to patch up. > -int snmp6_unregister_dev(struct inet6_dev *idev) > +int snmp6_free_dev(struct inet6_dev *idev) > { > snmp6_mib_free((void **)idev->stats.icmpv6); > return 0; > } You need to check whether icmpv6[0] is NULL either here or in snmp6_mib_free. Otherwise when snmp6_alloc_dev fails we'll wind up here and then call free_percpu on a pair of NULL pointers. Cheers, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt -------------------------------1109170304 Content-Type: text/html; charset="US-ASCII" Content-Transfer-Encoding: quoted-printable

Hello, Yoshi

Should I bother to apply this patch, or should I wait for you to make t= his=20 last change? What did you think about my comment that the Oops only occurred= =20 when the Ethernet cable had been unplugged during operation?

Regards,

Andy Liebman

<= FONT=20 style=3D"BACKGROUND-COLOR: transparent" face=3DArial color=3D#000000 size= =3D2>On Wed,=20 Feb 23, 2005 at 06:35:55PM +0900, YOSHIFUJI Hideaki / ?$B5HF#1QL@=20 wrote:
>
> What do you think of this?

Thanks, this loo= ks=20 great. There is just one technical detail
to patch up.

>= ;=20 -int snmp6_unregister_dev(struct inet6_dev *idev)
> +int=20 snmp6_free_dev(struct inet6_dev *idev)
> {
> = ;=20 snmp6_mib_free((void **)idev->stats.icmpv6);
> = ;=20 return 0;
> }

You need to check whether icmpv6[0= ] is=20 NULL either here or in
snmp6_mib_free. Otherwise when snmp6_alloc= _dev=20 fails we'll
wind up here and then call free_percpu on a pair of NULL=20 pointers.

Cheers,
--
Visit Openswan at=20 http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~}=20 <herbert@gondor.apana.org.au>
Home Page:=20 http://gondor.apana.org.au/~herbert/
PGP Key:=20 http://gondor.apana.org.au/~herbert/pubkey.txt

-------------------------------1109170304--