From e0edf25f141a7acc03748902bcb43974bd885353 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Thu, 24 Sep 2026 08:18:41 -0400 Subject: [PATCH] net/sched: cls_flower: exact-match ERSPAN key when no mask supplied Commit 292207809486 ("net: sched: fix erspan_opt settings in cls_flower") removed the unconditional memset of the erspan_metadata mask in fl_set_erspan_opt(). On the mask-parse pass, when userspace supplies the encap option key but no mask (TCA_FLOWER_KEY_ENC_OPTS without TCA_FLOWER_KEY_ENC_OPTS_MASK), the !depth early return leaves the mask zeroed -- a wildcard. A filter keyed on an ERSPAN index (v1), or dir/hwid (v2), then matches every ERSPAN packet's values instead of exact-matching. Default the match fields on the no-mask path the same way the with-mask branches default them: v1 index to 0xff, v2 dir and hwid to their exact-match masks. The union cannot just be memset to 0xff -- that would re-mask the overlapping timestamp/sgt bytes of the v2 header that the commit deliberately left to 0. A mask and a key that declare different ERSPAN versions describe different union layouts, so the mask would silently wildcard the key's matched fields. Reject that combination. Have fl_set_erspan_opt() report the parsed key's version back through a u8 pointer so the mask pass uses it without re-reading the union. This is a follow-up to commit fee10655709c ("net/sched: cls_flower: validate mask pointer after nla_next()"). Conditions to recreate the bug: add an ERSPAN encap-opts flower filter with a key but no mask (raw netlink, TCA_FLOWER_KEY_ENC_OPTS present, TCA_FLOWER_KEY_ENC_OPTS_MASK absent); dump the filter and observe the match mask is zero (wildcard) instead of exact for the supplied fields. For the mismatch case, pair a v1 key with a v2 mask (or vice versa) and observe the request is accepted with a mask of the other version. Closes: https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20260826025123.62758-1-ljp1205831794@gmail.com Link: https://lore.kernel.org/netdev/20260826025123.62758-1-ljp1205831794@gmail.com/ Reported-by: Sashiko (nipa) Link: https://lore.kernel.org/netdev/179003833955.2160803.1077681716391858800@kernel.org/ Suggested-by: Xin Long Signed-off-by: Jamal Hadi Salim Reviewed-by: Simon Horman Reviewed-by: Xin Long Link: https://patch.msgid.link/QDISC-4CSL.v2.20260924114800@mojatatu.com Signed-off-by: Jakub Kicinski --- net/sched/cls_flower.c | 39 ++++++++++++++++++++++++++++++++++----- 1 file changed, 34 insertions(+), 5 deletions(-) diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 1cefea571efd..0c4beff18d68 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -1358,17 +1358,29 @@ static int fl_set_vxlan_opt(const struct nlattr *nla, struct fl_flow_key *key, static int fl_set_erspan_opt(const struct nlattr *nla, struct fl_flow_key *key, int depth, int option_len, - struct netlink_ext_ack *extack) + struct netlink_ext_ack *extack, u8 *key_ver) { struct nlattr *tb[TCA_FLOWER_KEY_ENC_OPT_ERSPAN_MAX + 1]; struct erspan_metadata *md; int err; md = (struct erspan_metadata *)&key->enc_opts.data[key->enc_opts.len]; - md->version = 1; - if (!depth) + if (!depth) { + /* No mask attribute supplied: exact-match every field the + * key set, keyed off the key's own version so the v2 dir/hwid + * bytes are masked without touching the overlapping + * timestamp/sgt bytes of the union (see 292207809486). + */ + md->version = *key_ver; + if (md->version == 1) { + memset(&md->u.index, 0xff, sizeof(md->u.index)); + } else if (md->version == 2) { + md->u.md2.dir = 1; + set_hwid(&md->u.md2, 0xff); + } return sizeof(*md); + } if (nla_type(nla) != TCA_FLOWER_KEY_ENC_OPTS_ERSPAN) { NL_SET_ERR_MSG(extack, "Non-erspan option type for mask"); @@ -1387,6 +1399,22 @@ static int fl_set_erspan_opt(const struct nlattr *nla, struct fl_flow_key *key, if (tb[TCA_FLOWER_KEY_ENC_OPT_ERSPAN_VER]) md->version = nla_get_u8(tb[TCA_FLOWER_KEY_ENC_OPT_ERSPAN_VER]); + else + md->version = 1; + + /* The key and mask must describe the same union layout: the + * per-version default below only exact-matches the fields of its own + * version, so a mask version that differs from the key's would + * silently wildcard the key's matched fields. A mask without a + * version keeps the historical v1 default. + */ + if (option_len && md->version != *key_ver) { + NL_SET_ERR_MSG(extack, "Tunnel key erspan option ver mismatch"); + return -EINVAL; + } + + if (!option_len) + *key_ver = md->version; if (md->version == 1) { if (!option_len && !tb[TCA_FLOWER_KEY_ENC_OPT_ERSPAN_INDEX]) { @@ -1509,6 +1537,7 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, { const struct nlattr *nla_enc_key, *nla_opt_key, *nla_opt_msk = NULL; int err, option_len, key_depth, msk_depth = 0; + u8 key_ver = 1; err = nla_validate_nested_deprecated(tb[TCA_FLOWER_KEY_ENC_OPTS], TCA_FLOWER_KEY_ENC_OPTS_MAX, @@ -1607,7 +1636,7 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, key->enc_opts.dst_opt_type = IP_TUNNEL_ERSPAN_OPT_BIT; option_len = fl_set_erspan_opt(nla_opt_key, key, key_depth, option_len, - extack); + extack, &key_ver); if (option_len < 0) return option_len; @@ -1618,7 +1647,7 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, mask->enc_opts.dst_opt_type = IP_TUNNEL_ERSPAN_OPT_BIT; option_len = fl_set_erspan_opt(nla_opt_msk, mask, msk_depth, option_len, - extack); + extack, &key_ver); if (option_len < 0) return option_len;