In the Linux kernel, the following vulnerability has been resolved: bpf: Make post-verification instruction rewrites killable After do_check()… (CVE-2026-98628)
Description
A vulnerability in the Linux kernel's BPF verifier was resolved by making post-verification instruction rewrites interruptible. Previously, after verification, certain instruction rewrite passes could take a long time and were not preemptible, causing a task to be unkillable during this period. The fix introduces cancellation and rescheduling points during these rewrites, allowing a killed loader to be terminated promptly.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
The Linux kernel's BPF verifier performs instruction rewrite passes after do_check() returns. Some passes patch or remove instructions one at a time, moving instruction and auxiliary-data arrays and adjusting branch offsets, resulting in quadratic time complexity relative to program length. A privileged loader can exploit this by submitting many unconditional jumps, causing the verifier to spend excessive time removing each jump separately. Since the post-verification work did not check for signals or reschedule, a pending SIGKILL could not terminate the task until completion. The patch makes bpf_patch_insn_data() and verifier_remove_insns() common cancellation and rescheduling points, allowing the process to be interrupted and killed promptly. Interrupted constant blinding is reported as -EINTR and propagated through JIT paths, improving responsiveness without reducing the quadratic cost of rewrites.
Potential Impact
The vulnerability could cause a privileged loader process to become unresponsive to termination signals during BPF program verification, potentially leading to denial of service by making the loader unkillable for an extended period. The fix improves system responsiveness by allowing the verification process to be interrupted and terminated promptly. There is no indication of code execution or privilege escalation impact.
Mitigation Recommendations
A fix has been implemented in the Linux kernel to make the post-verification instruction rewrites interruptible and killable. Users should update to a kernel version that includes this patch. No additional mitigation is required as the patch addresses the issue directly.
Technical Details
- Gcve Source
- db.gcve.eu
- Osv Id
- GHSA-mq6p-2mw2-mp6g
- Osv Schema Version
- 1.4.0
- Aliases
- ["CVE-2026-98628"]
Threat ID: 6ad520ad2cdf04f656993a63
Added to database: 10/18/2026, 19:40:29 UTC
Last enriched: 10/18/2026, 21:34:15 UTC
Last updated: 10/19/2026, 00:08:23 UTC
Views: 2
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.