net/sched: etf: initialise the sock_exterr_skb before filling it
Subsystems (4)
PATCH 1/1 net/sched: etf: initialise the sock_exterr_skb before filling it
report_sock_error() clones the transmitted skb, overlays a struct sock_exterr_skb on its control block and fills in the seven struct sock_extended_err fields, but never clears addr_offset and port. skb_clone() copies cb[] verbatim, so the errqueue entry carries the TX skb's stale control block to the receive side, and the ipv4 error-queue reader trusts those bytes:
net/ipv4/ip_sockglue.c:556
sin->sin_addr.s_addr = *(__be32 *)(skb_network_header(skb) + serr->addr_offset); sin->sin_port = serr->port;
The gate that is supposed to guard that read only asks whether an address was requested at all:
net/ipv4/ip_sockglue.c:484
return serr->ee.ee_origin == SO_EE_ORIGIN_ICMP || serr->ee.ee_origin == SO_EE_ORIGIN_LOCAL || serr->port;
SO_EE_ORIGIN_TXTIME is neither ICMP nor LOCAL, so for sch_etf the gate is decided by serr->port alone -- by the stale byte pair. When it is non-zero the reader dereferences skb_network_header() + addr_offset with an arbitrary offset. On the TCP path cb[40..44) is not zero: the transmit path stored delivered_mstamp there (tcp_rate_skb_sent()), which is exactly where addr_offset/port sit in the arm64 struct layout.
Reached with an unmodified 7.3-rc4 (arm64, KASAN, in-tree modules): a TCP socket enables TX-timestamp error reporting (SO_TXTIME with SOF_TXTIME_REPORT_ERRORS), a sch_etf qdisc drops the packet, and a recvmsg() reaches ip_recv_error():
BUG: KASAN: slab-out-of-bounds in ip_recv_error+0x970/0xa0c Read of size 4 at addr ffff0000158ffdd5 by task A-227-A227-1_S2/113 ip_recv_error+0x970/0xa0c inet_recv_error+0x94/0xdc tcp_recvmsg+0x378/0x408 ... The buggy address is located 149 bytes to the right of allocated 704-byte region
Zero the whole overlay rather than the two fields individually: the errqueue entry is handed to the receive side as a self-contained record, which is what every other producer in the tree already does (sch_etf was the only error-queue producer that cleared neither field). This does not change what the in-tree selftest observes -- so_txtime.c reads only ee_origin/ee_errno/ee_code/ee_data/ee_info, never msg_name or sin_port.
---
Tested base
-----------
Detected on Linux 7.3-rc4 (commit 93f51579e7df), arm64 QEMU guest. Re-checked against mainline f47d223627f8 (v7.3-341): the affected code is unchanged between the two, so the bug is still present there. Checked against linux-next next-20261021 (commit 9a482184d910): the affected code is unchanged there as well, so the bug is still present in linux-next too.
Reproducer status
-----------------
A reproducer was written and executed on an arm64 QEMU guest (dummy-virt, KASAN kernel) against unmodified 7.3-rc4 in-tree modules; KASAN fired at the claimed sink. The reproducer source is not posted here per Documentation/process/security-bugs.rst ("do not publicly share a reproducer"); it is available on request.
Triggering conditions
---------------------CAP_NET_ADMIN to install the etf qdisc; the socket must opt into TX timestamp error reporting (SO_TXTIME + SOF_TXTIME_REPORT_ERRORS); a TCP flow with a non-empty retransmit queue. No out-of-tree module is needed.
Notes
-----
Built and checkpatch-clean: this patch was applied on top of 7.3-rc4 (commit 93f51579e7df) and a full arm64 kernel (defconfig) was built with it -- no new warnings or errors. scripts/checkpatch.pl reports 0 errors, 0 warnings.
net/sched/sch_etf.c | 1 + 1 file changed, 1 insertion(+)
net/sched/sch_etf.c +1 -0
@@ -147,6 +147,7 @@ static void report_sock_error(struct sk_buff *skb, u32 err, u8 code)return ; serr =SKB_EXT_ERR (clone); +memset (serr,0 ,sizeof (*serr)); serr->ee.ee_errno = err; serr->ee.ee_origin =SO_EE_ORIGIN_TXTIME ; serr->ee.ee_type =0 ;
--
2.39.5 (Apple Git-154)
[email protected]
AI reviewThis is an automated message. This series looks like a fix, but its commit messages seem to be missing some information:
Hi!
This is an automated message. This series looks like a fix, but its commit messages seem to be missing some information:
- How the issue was discovered, e.g. hit in production, hit during development, syzbot report, manual code inspection, LLM or static analysis tool scan.
Please do not repost the series just to address the above. Instead, reply to this email with the missing information, so that reviewers can take it into account. If the series needs another revision for other reasons, please include the information in the commit messages then.
The evaluation is done by an LLM so it may be wrong, if you think that is the case please reply and explain.
Hi, The issue was found by LLM-assisted review of this code path (static analysis), not from a production hit or a syzbot report. I then wrote a reproducer and ran it on an arm64 QEMU guest under KASAN; it fired at the claimed sink (ip_recv_error). The "Assisted-by: LLM" trailer reflects that assistance. Thanks.
Quoted 19 lines
2026年10月21日 下午11:58,[email protected] 写道:
Hi!
This is an automated message. This series looks like a fix, but its commit messages seem to be missing some information:
- How the issue was discovered, e.g. hit in production, hit during development, syzbot report, manual code inspection, LLM or static analysis tool scan.
Please do not repost the series just to address the above. Instead, reply to this email with the missing information, so that reviewers can take it into account. If the series needs another revision for other reasons, please include the information in the commit messages then.
The evaluation is done by an LLM so it may be wrong, if you think that is the case please reply and explain.
Quoted 74 lines · Dudu Lu
report_sock_error()clones the transmitted skb, overlays a structsock_exterr_skbon its control block and fills in the seven structsock_extended_errfields, but never clearsaddr_offsetand port.skb_clone()copies cb[] verbatim, so the errqueue entry carries the TX skb's stale control block to the receive side, and the ipv4 error-queue reader trusts those bytes:
net/ipv4/ip_sockglue.c:556sin->sin_addr.s_addr = *(__be32 *)(skb_network_header(skb) + serr->addr_offset); sin->sin_port = serr->port;The gate that is supposed to guard that read only asks whether an address was requested at all:
net/ipv4/ip_sockglue.c:484return serr->ee.ee_origin == SO_EE_ORIGIN_ICMP || serr->ee.ee_origin == SO_EE_ORIGIN_LOCAL || serr->port;
SO_EE_ORIGIN_TXTIMEis neither ICMP nor LOCAL, so forsch_etfthe gate is decided byserr->portalone -- by the stale byte pair. When it is non-zero the reader dereferencesskb_network_header()+addr_offsetwith an arbitrary offset. On the TCP path cb[40..44) is not zero: the transmit path storeddelivered_mstampthere (tcp_rate_skb_sent()), which is exactly where addr_offset/port sit in the arm64 struct layout.Reached with an unmodified 7.3-rc4 (arm64, KASAN, in-tree modules): a TCP socket enables TX-timestamp error reporting (
SO_TXTIMEwithSOF_TXTIME_REPORT_ERRORS), asch_etfqdisc drops the packet, and arecvmsg()reachesip_recv_error():BUG: KASAN: slab-out-of-bounds in ip_recv_error+0x970/0xa0c Read of size 4 at addr ffff0000158ffdd5 by task A-227-A227-1_S2/113 ip_recv_error+0x970/0xa0c inet_recv_error+0x94/0xdc tcp_recvmsg+0x378/0x408 ... The buggy address is located 149 bytes to the right of allocated 704-byte regionZero the whole overlay rather than the two fields individually: the errqueue entry is handed to the receive side as a self-contained record, which is what every other producer in the tree already does (
sch_etfwas the only error-queue producer that cleared neither field). This does not change what the in-tree selftest observes --so_txtime.creads only ee_origin/ee_errno/ee_code/ee_data/ee_info, nevermsg_nameorsin_port.Fixes 4b15c7075352 ("net/sched: Make etf report drops on error_queue")Assisted-by LLMSigned-off-by Dudu Lu <[email protected]>---
Tested base
-----------
Detected on Linux 7.3-rc4 (commit93f51579e7df), arm64 QEMU guest. Re-checked against mainlinef47d223627f8(v7.3-341): the affected code is unchanged between the two, so the bug is still present there. Checked against linux-next next-20261021 (commit9a482184d910): the affected code is unchanged there as well, so the bug is still present in linux-next too.Reproducer status
-----------------
A reproducer was written and executed on an arm64 QEMU guest (dummy-virt, KASAN kernel) against unmodified 7.3-rc4 in-tree modules; KASAN fired at the claimed sink. The reproducer source is not posted here perDocumentation/process/security-bugs.rst("do not publicly share a reproducer"); it is available on request.Triggering conditions
---------------------CAP_NET_ADMINto install the etf qdisc; the socket must opt into TX timestamp error reporting (SO_TXTIME+SOF_TXTIME_REPORT_ERRORS); a TCP flow with a non-empty retransmit queue. No out-of-tree module is needed.
Quoted 6 lines
Notes
-----
Built and checkpatch-clean: this patch was applied on top of 7.3-rc4 (commit93f51579e7df) and a full arm64 kernel (defconfig) was built with it -- no new warnings or errors.scripts/checkpatch.plreports 0 errors, 0 warnings.
This patch was applied to netdev/net.git (main) by Jakub Kicinski <[email protected]>:
Hello:
This patch was applied to netdev/net.git (main)
by Jakub Kicinski <[email protected]>:
Quoted 8 lines · On Wed, 21 Oct 2026 23:54:01 +0800 you wrote:
report_sock_error()clones the transmitted skb, overlays a structsock_exterr_skbon its control block and fills in the seven structsock_extended_errfields, but never clearsaddr_offsetand port.skb_clone()copies cb[] verbatim, so the errqueue entry carries the TX skb's stale control block to the receive side, and the ipv4 error-queue reader trusts those bytes:[...]
Here is the summary with links:
- [net] net/sched: etf: initialise the sock_exterr_skb before filling it
https://git.kernel.org/netdev/net/c/7076f684d513
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html