Skip to contentSkip to conversation

net: garp: reload skb header pointers after pskb_may_pull()

PATCH Applied David Carlier 2 participants 2 messages Jun 4 – Jun 9

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 net: garp: reload skb header pointers after pskb_may_pull()

garp_pdu_parse_attr() keeps a pointer into the skb linear area across pskb_may_pull(skb, ga->len), and garp_pdu_parse_msg() dereferences gm on every loop iteration even though the nested parse may pull again. pskb_may_pull() can reallocate the skb head, which would leave those pointers stale.

This is not reachable today: GARP PDUs arrive via the 802.2 LLC SAP path, where llc_fixup_skb() already pulls and trims the whole payload into the linear area, so the inner pulls never reallocate. Reload ga after the pull and snapshot gm->attrtype into a local anyway, to harden the parser and match the skb_header_pointer() discipline used by mrp.c.

No functional change.

Assisted-by Claude <[email protected]>
---
 net/802/garp.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
net/802/garp.c +4 -1
index c7a39f298ad6..56b934ec1aae 100644
@@ -452,6 +452,7 @@ static int garp_pdu_parse_attr(struct garp_applicant *app, struct sk_buff *skb,
 
 	if (!pskb_may_pull(skb, ga->len))
 		return -1;
+	ga = (struct garp_attr_hdr *)skb->data;
 	skb_pull(skb, ga->len);
 	dlen = ga->len - sizeof(*ga);
 
@@ -492,6 +493,7 @@ static int garp_pdu_parse_attr(struct garp_applicant *app, struct sk_buff *skb,
 static int garp_pdu_parse_msg(struct garp_applicant *app, struct sk_buff *skb)
 {
 	const struct garp_msg_hdr *gm;
+	u8 attrtype;
 
 	if (!pskb_may_pull(skb, sizeof(*gm)))
 		return -1;
@@ -499,9 +501,10 @@ static int garp_pdu_parse_msg(struct garp_applicant *app, struct sk_buff *skb)
 	if (gm->attrtype == 0)
 		return -1;
 	skb_pull(skb, sizeof(*gm));
+	attrtype = gm->attrtype;
 
 	while (skb->len > 0) {
-		if (garp_pdu_parse_attr(app, skb, gm->attrtype) < 0)
+		if (garp_pdu_parse_attr(app, skb, attrtype) < 0)
 			return -1;
 		if (garp_pdu_parse_end_mark(skb) < 0)
 			break;
Signature · 1 line
2.53.0
#1 lore raw
This patch was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Hello:

This patch was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Quoted 6 lines · On Thu, 4 Jun 2026 15:19:22 +0100 you wrote:

garp_pdu_parse_attr() keeps a pointer into the skb linear area across pskb_may_pull(skb, ga->len), and garp_pdu_parse_msg() dereferences gm on every loop iteration even though the nested parse may pull again. pskb_may_pull() can reallocate the skb head, which would leave those pointers stale.

Quoted 7 lines

This is not reachable today: GARP PDUs arrive via the 802.2 LLC SAP path, where llc_fixup_skb() already pulls and trims the whole payload into the linear area, so the inner pulls never reallocate. Reload ga after the pull and snapshot gm->attrtype into a local anyway, to harden the parser and match the skb_header_pointer() discipline used by mrp.c.

[...]

Here is the summary with links:
- net: garp: reload skb header pointers after pskb_may_pull()

    https://git.kernel.org/netdev/net-next/c/ab71cf79e2f8

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html