CVE-2026-98644: Linux
EPSS: 0.2% probability of exploitation activity in the next 30 days.
In the Linux kernel, the following vulnerability has been resolved: bpf: Avoid soft lockup in __htab_map_lookup_and_delete_batch() __htab_map_lookup_and_delete_batch() has no rescheduling point. The batch count bounds how many entries are copied out, not how many buckets are visited, so one BPF_MAP_LOOKUP_BATCH call can walk the map end to end. The empty-bucket fast path is worse: it stays inside a single rcu_read_lock() / bpf_disable_instrumentation() section for any run of consecutive empty buckets. That holds up on small maps, but it falls apart at scale. On a 144-CPU arm64 host running a CONFIG_PREEMPT_NONE kernel, periodic BPF_MAP_LOOKUP_BATCH calls against an LRU hash map with 16,777,216 buckets held a CPU inside the batch op for 77+ seconds and triggered the soft lockup watchdog. Commit 75134f16e7dd ("bpf: Add schedule points in batch ops") fixed this same problem in the generic batch ops, but not in this htab-native path, which every htab-based hash map variant uses for its lookup[_and_delete] batch ops. Complete that fix here. Leave the critical section after 64 consecutive empty buckets, call cond_resched_tasks_rcu_qs(), and resume at the saved bucket cursor. No locks are held at that point, and resuming from the cursor is already the function's behavior for non-empty buckets. Add the same call to the per-bucket loop after copy_to_user(), where every lock has been dropped. cond_resched_rcu() is not enough here: sleeping with bpf_prog_active elevated makes tracing programs on that CPU silently skip their invocations. Plain cond_resched() is not enough either. It is a no-op under PREEMPT and PREEMPT_LAZY, the only models arm64 and x86 have offered since commit 7dadeaa6e851 ("sched: Further restrict the preemption modes"). It is also never a Tasks RCU quiescent state, in any model: the reschedule counts as a preemption. The walking task stays a holdout and stalls every synchronize_rcu_tasks() caller, ftrace and BPF trampoline teardown included, until the syscall returns [1]. cond_resched_tasks_rcu_qs() is the usual tool for that [2]. It reports the quiescent state at each yield and still reschedules as cond_resched() does on PREEMPT_NONE and PREEMPT_VOLUNTARY kernels.
What to do
The record states a fixed version for each range of affected versions. Update to the one for the range your version is in: a later release on another branch may not have the fix.
- Linux, from 5.6, before 5.15.222: fixed in 5.15.222
- Linux, from 5.16, before 6.1.189: fixed in 6.1.189
- Linux, from 6.2, before 6.6.158: fixed in 6.6.158
- Linux, from 6.7, before 6.12.112: fixed in 6.12.112
- Linux, from 6.13, before 6.18.55: fixed in 6.18.55
- Linux, from 6.19, before 7.2.9: fixed in 7.2.9
Affected products
- Linux Linux: from 5.6, before 5.15.222 (fixed in 5.15.222); from 5.16, before 6.1.189 (fixed in 6.1.189); from 6.2, before 6.6.158 (fixed in 6.6.158); from 6.7, before 6.12.112 (fixed in 6.12.112); from 6.13, before 6.18.55 (fixed in 6.18.55); from 6.19, before 7.2.9 (fixed in 7.2.9)
Published 2026-10-18. Last modified 2026-10-19.