| Message ID | 21037748d4f9d8ff486151f4c09083bcf12d5df8.1723036486.git.petrm@nvidia.com (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 75bab45e6b2da379fe2ebda48ed35f8ce371a2ef |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | net: nexthop: Increase weight to u16 | expand |
On Wed, Aug 07, 2024 at 04:13:46PM +0200, Petr Machata wrote: > There are many unpatched kernel versions out there that do not initialize > the reserved fields of struct nexthop_grp. The issue with that is that if > those fields were to be used for some end (i.e. stop being reserved), old > kernels would still keep sending random data through the field, and a new > userspace could not rely on the value. > > In this patch, use the existing NHA_OP_FLAGS, which is currently inbound > only, to carry flags back to the userspace. Add a flag to indicate that the > reserved fields in struct nexthop_grp are zeroed before dumping. This is > reliant on the actual fix from commit 6d745cd0e972 ("net: nexthop: > Initialize all fields in dumped nexthops"). > > Signed-off-by: Petr Machata <petrm@nvidia.com> Reviewed-by: Ido Schimmel <idosch@nvidia.com> [...] > diff --git a/include/uapi/linux/nexthop.h b/include/uapi/linux/nexthop.h > index dd8787f9cf39..f4f060a87cc2 100644 > --- a/include/uapi/linux/nexthop.h > +++ b/include/uapi/linux/nexthop.h > @@ -33,6 +33,9 @@ enum { > #define NHA_OP_FLAG_DUMP_STATS BIT(0) > #define NHA_OP_FLAG_DUMP_HW_STATS BIT(1) > > +/* Response OP_FLAGS. */ > +#define NHA_OP_FLAG_RESP_GRP_RESVD_0 BIT(31) /* Dump clears resvd fields. */ (also cleared in notifications)
diff --git a/include/uapi/linux/nexthop.h b/include/uapi/linux/nexthop.h index dd8787f9cf39..f4f060a87cc2 100644 --- a/include/uapi/linux/nexthop.h +++ b/include/uapi/linux/nexthop.h @@ -33,6 +33,9 @@ enum { #define NHA_OP_FLAG_DUMP_STATS BIT(0) #define NHA_OP_FLAG_DUMP_HW_STATS BIT(1) +/* Response OP_FLAGS. */ +#define NHA_OP_FLAG_RESP_GRP_RESVD_0 BIT(31) /* Dump clears resvd fields. */ + enum { NHA_UNSPEC, NHA_ID, /* u32; id for nexthop. id == 0 means auto-assign */ diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c index 6b9787ee8601..23caa13bf24d 100644 --- a/net/ipv4/nexthop.c +++ b/net/ipv4/nexthop.c @@ -865,7 +865,7 @@ static int nla_put_nh_group_stats(struct sk_buff *skb, struct nexthop *nh, } static int nla_put_nh_group(struct sk_buff *skb, struct nexthop *nh, - u32 op_flags) + u32 op_flags, u32 *resp_op_flags) { struct nh_group *nhg = rtnl_dereference(nh->nh_grp); struct nexthop_grp *p; @@ -874,6 +874,8 @@ static int nla_put_nh_group(struct sk_buff *skb, struct nexthop *nh, u16 group_type = 0; int i; + *resp_op_flags |= NHA_OP_FLAG_RESP_GRP_RESVD_0; + if (nhg->hash_threshold) group_type = NEXTHOP_GRP_TYPE_MPATH; else if (nhg->resilient) @@ -934,10 +936,12 @@ static int nh_fill_node(struct sk_buff *skb, struct nexthop *nh, if (nh->is_group) { struct nh_group *nhg = rtnl_dereference(nh->nh_grp); + u32 resp_op_flags = 0; if (nhg->fdb_nh && nla_put_flag(skb, NHA_FDB)) goto nla_put_failure; - if (nla_put_nh_group(skb, nh, op_flags)) + if (nla_put_nh_group(skb, nh, op_flags, &resp_op_flags) || + nla_put_u32(skb, NHA_OP_FLAGS, resp_op_flags)) goto nla_put_failure; goto out; } @@ -1050,7 +1054,9 @@ static size_t nh_nlmsg_size(struct nexthop *nh) sz += nla_total_size(4); /* NHA_ID */ if (nh->is_group) - sz += nh_nlmsg_size_grp(nh); + sz += nh_nlmsg_size_grp(nh) + + nla_total_size(4) + /* NHA_OP_FLAGS */ + 0; else sz += nh_nlmsg_size_single(nh);
There are many unpatched kernel versions out there that do not initialize the reserved fields of struct nexthop_grp. The issue with that is that if those fields were to be used for some end (i.e. stop being reserved), old kernels would still keep sending random data through the field, and a new userspace could not rely on the value. In this patch, use the existing NHA_OP_FLAGS, which is currently inbound only, to carry flags back to the userspace. Add a flag to indicate that the reserved fields in struct nexthop_grp are zeroed before dumping. This is reliant on the actual fix from commit 6d745cd0e972 ("net: nexthop: Initialize all fields in dumped nexthops"). Signed-off-by: Petr Machata <petrm@nvidia.com> --- Notes: v2: - Move the new OP_FLAG to bit 31 to make in/out confusion less likely - Add a comment to the flag include/uapi/linux/nexthop.h | 3 +++ net/ipv4/nexthop.c | 12 +++++++++--- 2 files changed, 12 insertions(+), 3 deletions(-)