netdev - RE: [PATCH iwl-next v3 06/10] ice: Flesh out implementation of support for SRIOV on bonded interface

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Fri, 9 Jun 2023 16:32:27 +0000
From: "Ertman, David M" <david.m.ertman@...el.com>
To: Simon Horman <simon.horman@...igine.com>
CC: "intel-wired-lan@...ts.osuosl.org" <intel-wired-lan@...ts.osuosl.org>,
	"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
	"daniel.machon@...rochip.com" <daniel.machon@...rochip.com>
Subject: RE: [PATCH iwl-next v3 06/10] ice: Flesh out implementation of
 support for SRIOV on bonded interface

> -----Original Message-----
> From: Simon Horman <simon.horman@...igine.com>
> Sent: Friday, June 9, 2023 2:01 AM
> To: Ertman, David M <david.m.ertman@...el.com>
> Cc: intel-wired-lan@...ts.osuosl.org; netdev@...r.kernel.org;
> daniel.machon@...rochip.com
> Subject: Re: [PATCH iwl-next v3 06/10] ice: Flesh out implementation of
> support for SRIOV on bonded interface
> 
> On Thu, Jun 08, 2023 at 11:06:14AM -0700, Dave Ertman wrote:
> 
> ...
> 

...

> > +
> > +	/* add parent if none were free */
> > +	if (!n_prt) {
> 
> Hi Dave,
> 
> I suppose this can't happen.
> But if aggnode->num_children is 0 then n_prt will be uninitialised here.

If aggnode->num_children is 0, something very wrong (to the point of the device failing probe)
would have to have happened.  But, you are correct, it would be better to initialize n_prt to NULL.
That is a nice catch!

> 
> > +		u16 num_nodes_added;
> > +		u32 first_teid;
> > +		int status;

...

> > +	/* find parent in primary tree */
> > +	pi = hw->port_info;
> > +	tc_node = ice_sched_get_tc_node(pi, tc);
> > +	if (!tc_node) {
> > +		dev_warn(dev, "Failure to find TC node in failover tree\n");
> > +		goto resume_reclaim;
> > +	}
> > +
> > +	aggnode = ice_sched_get_agg_node(pi, tc_node,
> ICE_DFLT_AGG_ID);
> > +	if (!aggnode) {
> > +		dev_warn(dev, "Failure to find aggreagte node in failover
> tree\n");
> > +		goto resume_reclaim;
> > +	}
> > +
> > +	aggl = ice_sched_get_agg_layer(hw);
> > +	vsil = ice_sched_get_vsi_layer(hw);
> > +
> > +	for (n = aggl + 1; n < vsil; n++)
> > +		num_nodes[n] = 1;
> > +
> > +	for (n = 0; n < aggnode->num_children; n++) {
> > +		n_prt = ice_sched_get_free_vsi_parent(hw, aggnode-
> >children[n],
> > +						      num_nodes);
> > +		if (n_prt)
> > +			break;
> > +	}
> > +
> > +	/* if no free parent found - add one */
> > +	if (!n_prt) {
> 
> Likewise, here too.

Actually, this code was refactored out into a subfunction in patch 10/10 for
this series.  In the subfunction, n_prt is initialized to NULL (for purposes of
using it as a return value.  I need to move that refactoring back into this
patch since there are multiple uses this far back.  Thanks for pointing this
one out to me!

I will send this change out in patch set v4!

DaveE

> 
> > +		u16 num_nodes_added;
> > +		u32 first_teid;