On Mon, Sep 18, 2017 at 01:27:52PM -0700, Greg Rose wrote:
> On 09/18/2017 11:15 AM, Ben Pfaff wrote:
> >On Mon, Sep 18, 2017 at 10:58:28AM -0700, Greg Rose wrote:
> >>On 09/12/2017 08:37 PM, ychen wrote:
> >>>in function nla_alloc_flow_actions(), there is a check if action length is
> >>>greater than MAX_ACTIONS_BUFSIZE(32k), then kernel datapath flow will not
> >>>be installed, and packets will droppped.
> >>>but in function xlate_actions(), there is such clause:
> >>>if (nl_attr_oversized(ctx.odp_actions->size)) {
> >>> /* These datapath actions are too big for a Netlink attribute, so
> >>> we
> >>> * can't hand them to the kernel directly. dpif_execute() can
> >>> execute
> >>> * them one by one with help, so just mark the result as
> >>> SLOW_ACTION to
> >>> * prevent the flow from being installed. */
> >>> COVERAGE_INC(xlate_actions_oversize);
> >>> ctx.xout->slow |= SLOW_ACTION;
> >>> }
> >>>and in function nl_attr_oversized(), the clause is like this:
> >>>return payload_size > UINT16_MAX - NLA_HDRLEN;
> >>>
> >>>
> >>>so we can see that in user space, max action length is almost 64K, but in
> >>>kernel space, max action length is only 32K.
> >>>my question is: why the max action length is different? packet will drop
> >>>when its action length exceeds 32K, but packet can excute in slow path
> >>>when its action length exceeds 64K?
> >>
> >>It's a kernel limitation.
> >>
> >>http://www.spinics.net/lists/netdev/msg431592.html
> >
> >It sounds like the userspace limit, then, should also be 32 kB (or
> >possibly 16 kB). I guess we should fix that.
> >
>
> Correct, the user space limit should be 32KB. That's what it is in iproute2.
OVS supports Linux < 4.9 as well, so should we stick with 16 kB (or
detect the kernel version or limit somehow)?
http://www.spinics.net/lists/netdev/msg431592.html
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
