Re: data-race in nf_tables_newtable / nf_tables_newtable

[Gabriel Ryan]

Ok, glad everything worked out then, thanks for sharing!

Best,

Gabe

On Mon, Aug 22, 2022 at 4:42 PM Florian Westphal <fw@xxxxxxxxx> wrote:
>
> Gabriel Ryan <gabe@xxxxxxxxxxxxxxx> wrote:
> > Hi Florian,
> >
> > I just looked at the lock event trace from our report and it looks
> > like two distinct commit mutexes were held when the race was
> > triggered. I think the race is probably on the table_handle variable
> > on net/netfilter/nf_tables_api.c:1221, and not the table->handle field
> > being written to.
>
> See
>
> https://urldefense.proofpoint.com/v2/url?u=https-3A__patchwork.ozlabs.org_project_netfilter-2Ddevel_patch_20220821085939.571378-2D1-2Dpablo-40netfilter.org_&d=DwIFAg&c=009klHSCxuh5AI1vNQzSO0KGjl4nbi2Q0M1QLJX9BeE&r=EyAJYRJu01oaAhhVVY3o8zKgZvacDAXd_PNRtaqACCo&m=tKybb4B0md9Rrb5d9AH-UN2fjr4L8y6LtRxywepWhKsWCyN0OWaRwAtDbGt10WQD&s=uJ4BtOvCcZOe6F17rDKQipDF0McLeOwaUL9jO3BgTCU&e=
>
> which makes table_handle per netns.