I got distracted by some other issues and lost track of this thread.
Can one of you please commit the fix if the discussion has reached a
conclusion? Thanks!
Regards,
Navdeep
On Wed, Jun 25, 2014 at 02:08:35PM +0400, Gleb Smirnoff wrote:
> Alan,
>
> On Tue, Jun 24, 2014 at 08:43:40AM -0600, A
Alan,
On Tue, Jun 24, 2014 at 08:43:40AM -0600, Alan Somers wrote:
A> That looks better. But I think there is one more possibility for a
A> leak. For multicast packets, IFP_TO_IA at line 263 will call
A> ifa_ref(), unless the the interface has no address assigned. How
A> about this patch inst
On Tue, Jun 24, 2014 at 3:08 AM, Gleb Smirnoff wrote:
> On Mon, Jun 23, 2014 at 10:44:58AM -0600, Alan Somers wrote:
> A> > On Fri, Jun 20, 2014 at 12:15:21PM -0700, Navdeep Parhar wrote:
> A> > N> Revision 264905 and 266860 that followed it seem to leak ifaddr
> A> > N> references. ifa_ifwithdst
On Mon, Jun 23, 2014 at 10:44:58AM -0600, Alan Somers wrote:
A> > On Fri, Jun 20, 2014 at 12:15:21PM -0700, Navdeep Parhar wrote:
A> > N> Revision 264905 and 266860 that followed it seem to leak ifaddr
A> > N> references. ifa_ifwithdstaddr and ifa_ifwithnet both install a
A> > N> reference on the
On Mon, Jun 23, 2014 at 2:52 AM, Gleb Smirnoff wrote:
> Navdeep,
>
> On Fri, Jun 20, 2014 at 12:15:21PM -0700, Navdeep Parhar wrote:
> N> Revision 264905 and 266860 that followed it seem to leak ifaddr
> N> references. ifa_ifwithdstaddr and ifa_ifwithnet both install a
> N> reference on the ifa
Navdeep,
On Fri, Jun 20, 2014 at 12:15:21PM -0700, Navdeep Parhar wrote:
N> Revision 264905 and 266860 that followed it seem to leak ifaddr
N> references. ifa_ifwithdstaddr and ifa_ifwithnet both install a
N> reference on the ifaddr returned to the caller but ip_output does not
N> release it, e
On 06/20/14 15:20, Alan Somers wrote:
...
>
> Do you have a test case that can reproduce the panic?
>
> -Alan
>
Just run some UDP traffic on head or stable/10 and watch the refcount
leak on the transmitter.
# netperf -H ... -t UDP_STREAM
I see these two do ifa_ref (once per packet).
On Fri, Jun 20, 2014 at 1:15 PM, Navdeep Parhar wrote:
> Revision 264905 and 266860 that followed it seem to leak ifaddr
> references. ifa_ifwithdstaddr and ifa_ifwithnet both install a
> reference on the ifaddr returned to the caller but ip_output does not
> release it, eventually leading to a p
Revision 264905 and 266860 that followed it seem to leak ifaddr
references. ifa_ifwithdstaddr and ifa_ifwithnet both install a
reference on the ifaddr returned to the caller but ip_output does not
release it, eventually leading to a panic when the refcount wraps over
to 0 and the ifaddr is freed w
