From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-98549][IMPORTANT] tcp: prevent collapsing skbs across boundary in rtx queue Date: Sun, 18 Oct 2026 22:57:28 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-98549 X-AL-KERNEL-Priority: IMPORTANT X-AL-KERNEL-Severity: IMPORTANT X-AL-KERNEL-Base-Severity: IMPORTANT X-AL-KERNEL-KPANIC: NO X-AL-KERNEL-ActionableScore: 7 X-AL-KERNEL-ActionableScore-Lower: 6 X-AL-KERNEL-Commit: 1ae7aee395245874d797890380d9192bd858fb7e List-Id: CVE: CVE-2026-98549 Priority: IMPORTANT AL-KERNEL base severity: IMPORTANT KPANIC flag: NO Patch: tcp: prevent collapsing skbs across boundary in rtx queue Commit: 1ae7aee395245874d797890380d9192bd858fb7e Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=1ae7aee395245874d797890380d9192bd858fb7e Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98549 Analysis date: Sun, 18 Oct 2026 22:57:28 -0400 ActionableScore: 7 ActionableScore lower bound: 6 Actionable bucket: Actionable Moderate / Strong Important candidate requiring manual review Manual review required: YES Summary: A TCP retransmission collapse can merge data across a device encryption boundary, causing PSP protected data to be retransmitted in cleartext or TLS offloaded connections to stall persistently. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-98549 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98549 The original announcement does not normally provide a security severity estimate, CVSS assessment, or enough information to determine whether the reported kernel bug represents a practically relevant security issue. This report was generated by AL-KERNEL, an AI-assisted Linux kernel vulnerability analysis system developed by Alexander Larkin. It combines an autonomous classifier with LLM-assisted technical analysis and a separate ActionableScore mechanism. The purpose of this report is to prioritize Linux kernel CVEs before manual review, identify cases that require prompt investigation, and support automatic closure of issues that are unlikely to have meaningful security impact. Published priority for this report: IMPORTANT Manual review required: YES A detailed explanation of the methodology and priority rules is included at the end of this message. ====================================================================== AL-KERNEL CLASSIFICATION RESULT ====================================================================== CVE-2026-98549 IMPORTANT https://www.kernelcve.org/list/?q=CVE-2026-98549 CHECK WITH IMPACT FROM ORIG NN LOW | AEGIS IMPACT MODERATE_KPANIC AEGIS_AS=5 LEGACY=IMPORTANT RESULT=DIFF INDEPENDENT_LEGACY=IMPORTANT INDEPENDENT_RESULT=DIFF TEMP_KPANIC_MERGE=IMPORTANT_HARD_VETO PRE_MERGE_LEGACY=IMPORTANT (final Legacy and Aegis differ; FINAL: IMPORTANT, Aegis: MODERATE_KPANIC) Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:A/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:H';CWE-319;CWE-693;**CWE-400;Other CVSS 'AV:A/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:L';BEST CVSS score: '6.8';DESCR 'A TCP skb collapse can cross the boundary where device encryption becomes active because the collapse fence may be absent from the retransmit queue when the write queue is empty. With PSP, the merged skb can retain decrypted equal to zero from the earlier skb and retransmit data queued after psp_sock_assoc_set_tx in cleartext. With TLS device offload, the merged skb can cross the offload start marker and be repeatedly dropped by the software fallback or affected drivers, causing the connection to stall. For the CVSS the PR:N is selected because a network peer or on path attacker does not need privileges on the affected host. AV:A and AC:H reflect that practical exploitation requires access to the relevant network path, enabled device encryption, an encryption transition with earlier data still unacknowledged, and a retransmit or SACK collapse event. Impact includes disclosure of application data expected to be protected and denial of service for affected connections. No memory corruption or privilege escalation primitive is identified. Automated impact decision path. The original SimpleNet classifier selected MODERATE_KPANIC. The primary LLM estimated paranoid CVSS 6.8 and the production ActionableScore returned 7 with recommendation Actionable Moderate / Strong Important candidate requiring manual review. The production rules then evaluated CVSS thresholds and ActionableScore thresholds. Recorded rule transitions were CVSS and initial triage stage MODERATE_KPANIC to IMPORTANT because the CVSS score, vector or bug class threshold rule was applied. final reconciliation stage IMPORTANT to IMPORTANT because a specific evidence based protection or manual review veto was applied and KP specialist did not establish a qualifying panic. The legacy KP specialist produced a review artifact and its decision was processed by the applicable KP rules. The legacy FD or DH downgrade specialist produced a review artifact and its decision was processed by the applicable downgrade and preservation rules. Exact triggered rule markers were INCREASED_TO_HIGH_BASED_ON_GUESSCVSS and TEMP_KPANIC_MERGE_IMPORTANT_VETO and KPANIC_UNSUPPORTED_REMOVED. The independent weaker Aegis shadow evaluated CVSS 5.9 and ActionableScore 5 with primitive class CLASS_A and manual review REQUIRED. Its independently reconciled impact was MODERATE_KPANIC. Cross model aggregation changed the production result under an enabled evidence gated merge rule. The final reported impact is IMPORTANT.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 7) MAYBE REMOTE NETWORK SKB HARDWARE INCREASED_TO_HIGH_BASED_ON_GUESSCVSS TEMP_KPANIC_MERGE_IMPORTANT_VETO KPANIC_UNSUPPORTED_REMOVED - - checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=7 ActionableScoreLower=6 PrimitiveClass=CLASS_A RealUAF=NO HostAdminRequired=NO NamespaceScopedPrivilege=NO ImportantCandidate=YES ManualReview=REQUIRED AutoDowngradeAllowed=NO RequiresHostUID0=NO AuthorizationDecisionBypass=NO ProtectedResourceImpact=NONE B4PendingReview=NO B4PendingConfirmed=NO 1. ActionableScore * Raw conservative score: 6 * Conservative score: 6 * Raw paranoid score: 7 * Paranoid score: 7 * No floor or cap was applied. * Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**:: 2. Signal breakdown Conservative positive signals: * +2 Remote reachable or directly network-triggerable. TCP retransmission and SACK processing consume peer-controlled ACK and SACK behavior after device encryption is enabled. The attacker needs no privilege on the affected host. * +2 Strong attacker-influenced semantic reinterpretation. Data queued before and after the encryption transition can be merged while retaining the earlier skb encryption metadata. The resulting skb is treated as belonging to the wrong encryption domain. * +1 Confidentiality impact. The commit directly establishes that PSP data queued after `psp_sock_assoc_set_tx()` can be retransmitted in cleartext. * +1 Availability impact. With TLS device offload, the malformed cross-boundary skb is repeatedly rebuilt and dropped, causing the affected connection to stall. Paranoid addition: * +1 Practical security-consequence chain. An on-path attacker can induce retransmission through packet loss, while a TCP peer can influence ACK and SACK processing. The patch establishes the remaining chain from collapse through cleartext PSP transmission or persistent TLS connection failure. Important non-awarded signals: * No UAF, OOB access, heap corruption, refcount error, or arbitrary kernel write is established. * No privilege escalation, kernel code execution, credential modification, or host-wide panic is established. * No additional integrity point is awarded. The relevant integrity failure is the encryption-domain metadata mismatch already counted as semantic reinterpretation. * No extra availability point is awarded for repeatability. Rebuilding and dropping the same skb is part of the already counted connection stall. * No common-networking-path point is added because the affected PSP and TLS device-offload configurations are narrower than ordinary TCP operation. * No hard-race penalty is applied. The issue requires a state sequence, but the patch does not establish a narrow or uncontrollable timing race. * NEW52 B5-R does not apply because there is no retry-accumulated heap OOB write. * NEW53 B3-PENDING does not apply because no shared-backing write is established. * V32 and V34 do not apply because checksum metadata does not cause a header write. * B4-PENDING does not apply because no authorization DENY-to-ALLOW decision is bypassed. 3. Reachability analysis The evaluated attacker is a remote TCP peer or an on-path network actor. Once the affected host uses PSP or TLS device offload, retransmission or SACK-driven collapse can reach the vulnerable operation without host credentials. For confidentiality exploitation, an on-path actor can observe cleartext retransmissions and can induce packet loss that causes retransmission. A malicious TCP peer can directly influence ACK and SACK behavior, although that peer is normally already the intended recipient of the application data. Host UID 0 and host-administrator capabilities are not required by the network attacker. Namespace capabilities are not part of the trigger. Device encryption must already be enabled by the local endpoint, but this is a deployment prerequisite rather than an attacker privilege. PSP or TLS device offload support and a transition while earlier data remains unacknowledged are required. The write queue must be empty while the retransmit queue still contains an earlier skb, followed by retransmission or SACK shifting. These conditions increase complexity but do not establish a hard, one-shot race. 4. Primitive and severity interpretation The exact primitive is an active security-state boundary violation. TCP combines data from opposite sides of a device-encryption transition while preserving the earlier skb `decrypted` state. This is not memory corruption. It is a security-relevant semantic reinterpretation of active structured packet state. For PSP, the demonstrated consequence is retransmission of data in cleartext even though it was queued after encryption became active. For TLS device offload, the demonstrated consequence is a persistent per-connection stall because software fallback and affected drivers reject the cross-boundary skb on every retransmission. The classification is CLASS_A because a concrete confidentiality boundary is defeated and protected traffic is transmitted under the wrong encryption state. The patch does not support stronger claims such as arbitrary packet modification, host compromise, kernel panic, or privilege escalation. ImportantCandidate is YES because the cleartext transmission is a demonstrated non-crash security consequence across an encryption trust boundary. AutoDowngradeAllowed is NO because automatic Moderate downgrading would discard this confirmed confidentiality primitive. The likely operational severity is Important-candidate for PSP deployments and actionable Moderate for the TLS connection-stall branch. 5. One-sentence report phrase A TCP retransmission collapse can merge data across a device encryption boundary, causing PSP protected data to be retransmitted in cleartext or TLS offloaded connections to stall persistently. 6. Manual review recommendation MANUAL CHECK REQUIRED The PSP path demonstrates cleartext transmission of data expected to be encrypted, while the exact exposure and attacker positioning depend on deployment topology and device-offload configuration. 7. Call-site confidence Call-site confidence: medium ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: tcp: prevent collapsing skbs across boundary in rtx queue Commit: 1ae7aee395245874d797890380d9192bd858fb7e Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=1ae7aee395245874d797890380d9192bd858fb7e Commit description: tcp_write_collapse_fence() sets TCP_SKB_CB(skb)->eor = 1 on tcp_write_queue_tail(sk) to prevent skbs queued after a switch to device encryption from being collapsed into earlier skbs. The fence is a no-op if all earlier data has already been transmitted when the switch happens: sk->sk_write_queue is empty. The not yet acknowledged earlier skbs wait in sk->tcp_rtx_queue with eor 0. On a subsequent retransmit or SACK shift, tcp_retrans_try_collapse() or tcp_shift_skb_data() can then merge an skb queued after the switch into one queued before it. Both users of the fence are affected: - psp: devices only encrypt skbs with skb->decrypted set. The merged skb keeps decrypted = 0 from the earlier skb, so merged data sent after psp_sock_assoc_set_tx() is retransmitted in cleartext. - tls device offload: the merged skb straddles the start marker set in tls_set_device_offload(). The software fallback (fill_sg_in() returns -EINVAL) and the mlx5, nfp and funeth drivers cannot handle such an skb and drop it. Every retransmit rebuilds the same skb, so the connection stalls. Fix this in two places, for defense in depth: 1. Fall back to tcp_rtx_queue_tail(sk) in tcp_write_collapse_fence() when tcp_write_queue_tail(sk) is NULL. 2. Check !skb_cmp_decrypted(to, from) in tcp_skb_can_collapse(), as tcp_skb_can_collapse_rx() does on receive. skb_shift(), which both collapse paths call, already has a DEBUG_NET_WARN_ON_ONCE() for this condition. Fixes: e8f6979 ("net/tls: Add generic NIC offload infrastructure") Cc: stable@vger.kernel.org Signed-off-by: Willem de Bruijn Reviewed-by: Eric Dumazet Reviewed-by: Daniel Zahka Link: https://patch.msgid.link/20260924154427.953800-1-willemdebruijn.kernel@gmail.com Signed-off-by: Jakub Kicinski Signed-off-by: Greg Kroah-Hartman Changed files: include/net/tcp.h Diff excerpt: Not included in this email. See the upstream URL for the full patch. Full patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=1ae7aee395245874d797890380d9192bd858fb7e ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-98549 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-98549 The original CVE announcement normally does not include a security-level estimate. In particular, it may not contain a CVSS assessment, an impact level, or enough information to determine whether the reported bug is a practically relevant security issue. One purpose of this parallel CVE list is to provide that missing technical and prioritization information. The original goal of the AL-KERNEL project was to prioritize Linux kernel CVE analysis automatically before manual review. The system can also help identify non-security issues that may be suitable for automatic closure. This report was generated by AL-KERNEL, an AI-assisted Linux kernel vulnerability analysis system developed by Alexander Larkin. The first analysis stage combines an autonomous classifier with additional LLM-based analysis. The autonomous classifier runs locally on a CPU and is based on a backpropagation neural network. Together, these mechanisms produce a technical vulnerability description, identify likely weakness types, estimate CVSS severity, and provide input for ActionableScore. Two CVSS estimates are retained because incomplete kernel vulnerability information often permits more than one defensible interpretation: Conservative CVSS vector: AV:A/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:L The Best / paranoid CVSS vector: AV:A/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:H The Best / paranoid CVSS score: 6.8 The conservative vector represents a lower-impact interpretation. The Best/paranoid vector intentionally represents a plausible upper-bound interpretation and should not automatically be treated as demonstrated real-world impact. CVSS may also need to be adjusted for a particular Linux deployment, because actual reachability, privileges, enabled kernel configuration, hardware, namespaces, exposed device nodes, and other environmental conditions can differ significantly between systems. A separate ActionableScore mechanism evaluates practical remediation urgency. Its analysis may include reachability, attack prerequisites, subsystem exposure, memory-corruption characteristics, denial-of-service reliability, and possible confidentiality, integrity, or privilege-escalation impact. Conservative ActionableScore: 6 Paranoid ActionableScore: 7 The final base severity is taken directly from the second tab-separated field of the AL-KERNEL classification result. ActionableScore does not replace or independently override that final AL-KERNEL decision, and if ActionableScore adjusted impact level of ALKERNEL, then you would see self-readable flags above like INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHEREQTHAN7. For an AL-KERNEL result of MODERATE, this report uses the following additional presentation split: ActionableScore below 5 -> MODERATE REGULAR ActionableScore 5 or more -> MODERATE 7.0 The distinction between MODERATE REGULAR and MODERATE 7.0 makes it possible to identify Moderate issues that should receive manual analysis and fixes before lower-priority MODERATE REGULAR issues. In many cases, MODERATE REGULAR fixes may wait for a later rebase or routine update. There is one override in which MODERATE REGULAR becomes MODERATE 7.0 even when the ActionableScore is below 5. When the AL-KERNEL result contains the KPANIC flag, a MODERATE result is always presented as MODERATE 7.0. The KPANIC flag selected with few regexps without usage of AI at all, so it helps to detect cases when Kernel Crash happens and similar (to filter False-Negative results from the LLM usage). KPANIC indicates that a reliable kernel crash, kernel panic, or similarly serious kernel availability impact was identified by the classification workflow. AL-KERNEL base severity for this report: IMPORTANT KPANIC detected for this report: NO Published priority for this report (same as in Subject): IMPORTANT These results are intended to support engineering triage. They are machine-generated estimates, and cases marked for manual review should be validated by a human security engineer before final disposition. For more info read docs linked from here: https://kernelcve.org/ (and you can submit you own patch there to generate such a report for non-existant CVE-id yet). Note that in many cases this AI tool selects higher severity, than real is (means you can expect Importants instead of Moderate 7.0 or Moderates 7.0 instead of regular Moderates). If you see such cases, please use reply email interface to add additional manual analyses info to this particular CVE. And please, please, let me know when you see Lows instead of Importants or Important instead of Low (because particular for such cases I need to tune this AI tool to make it better for this one and next similar). My contact email for such notifications is alexanjelausa@gmail.com (and both send reply to CVE record itself too and see "reply" button below for howto reply).