{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2026-43385","assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","state":"PUBLISHED","assignerShortName":"Linux","dateReserved":"2026-05-01T14:12:56.006Z","datePublished":"2026-05-08T14:21:32.007Z","dateUpdated":"2026-08-05T12:27:43.989Z"},"containers":{"cna":{"providerMetadata":{"orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux","dateUpdated":"2026-08-05T12:27:43.989Z"},"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nnet: Fix rcu_tasks stall in threaded busypoll\n\nI was debugging a NIC driver when I noticed that when I enable\nthreaded busypoll, bpftrace hangs when starting up. dmesg showed:\n\n  rcu_tasks_wait_gp: rcu_tasks grace period number 85 (since boot) is 10658 jiffies old.\n  rcu_tasks_wait_gp: rcu_tasks grace period number 85 (since boot) is 40793 jiffies old.\n  rcu_tasks_wait_gp: rcu_tasks grace period number 85 (since boot) is 131273 jiffies old.\n  rcu_tasks_wait_gp: rcu_tasks grace period number 85 (since boot) is 402058 jiffies old.\n  INFO: rcu_tasks detected stalls on tasks:\n  00000000769f52cd: .N nvcsw: 2/2 holdout: 1 idle_cpu: -1/64\n  task:napi/eth2-8265  state:R  running task     stack:0     pid:48300 tgid:48300 ppid:2      task_flags:0x208040 flags:0x00004000\n  Call Trace:\n   <TASK>\n   ? napi_threaded_poll_loop+0x27c/0x2c0\n   ? __pfx_napi_threaded_poll+0x10/0x10\n   ? napi_threaded_poll+0x26/0x80\n   ? kthread+0xfa/0x240\n   ? __pfx_kthread+0x10/0x10\n   ? ret_from_fork+0x31/0x50\n   ? __pfx_kthread+0x10/0x10\n   ? ret_from_fork_asm+0x1a/0x30\n   </TASK>\n\nThe cause is that in threaded busypoll, the main loop is in\nnapi_threaded_poll rather than napi_threaded_poll_loop, where the\nlatter rarely iterates more than once within its loop. For\nrcu_softirq_qs_periodic inside napi_threaded_poll_loop to report its\nqs state, the last_qs must be 100ms behind, and this can't happen\nbecause napi_threaded_poll_loop rarely iterates in threaded busypoll,\nand each time napi_threaded_poll_loop is called last_qs is reset to\nlatest jiffies.\n\nThis patch changes so that in threaded busypoll, last_qs is saved\nin the outer napi_threaded_poll, and whether busy_poll_last_qs\nis NULL indicates whether napi_threaded_poll_loop is called for\nbusypoll. This way last_qs would not reset to latest jiffies on\neach invocation of napi_threaded_poll_loop."}],"metrics":[{"cvssV3_1":{"version":"3.1","vectorString":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","baseScore":7.5,"baseSeverity":"HIGH"},"scenarios":[{"lang":"en","value":"AV:N - In a reasonable high-impact deployment, threaded NAPI busy-poll is enabled on a network-facing NIC/forwarder, and received packets schedule the NAPI kthread into the vulnerable polling loop. The netlink knob is an administrative deployment gate, but a remote sender needs no local access once that mode is in use.\nAC:L - There is no attacker-controlled race or fragile timing requirement; once the NAPI thread is in threaded busy-poll mode, repeated loop invocations reliably fail to report RCU Tasks quiescent states. Packet delivery to the configured NAPI is sufficient to activate the path.\nPR:N - The highest-impact scenario is an already-configured network service/device where the attacker only sends packets and needs no account or credentials. Enabling threaded busy-poll itself requires init-namespace CAP_NET_ADMIN, but that is a deployment precondition rather than an exploit privilege in this scenario.\nUI:N - No victim user action is required after the affected NAPI mode is configured. Network traffic alone can keep the vulnerable kthread path active.\nS:U - The impact remains within the kernel and host networking/RCU scheduling authority. There is no VM escape, sandbox boundary crossing, or separate security authority affected.\nC:N - The bug is an RCU Tasks quiescent-state starvation/stall, not memory corruption or an information disclosure path. I found no read primitive or exposure of protected data.\nI:N - The flaw does not provide an arbitrary write, control-flow hijack, or data modification primitive. It only prevents timely RCU Tasks grace-period progress.\nA:H - The vulnerable loop can indefinitely stall RCU Tasks grace periods, causing kernel operations such as tracing/BPF setup or teardown to hang and producing persistent RCU stall reports. This is a kernel-level hang/deadlock-style availability failure."}]}],"affected":[{"product":"Linux","vendor":"Linux","defaultStatus":"unaffected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["net/core/dev.c"],"versions":[{"version":"c18d4b190a46651726c9a952667c74d2deb33c28","lessThan":"52459201d0df3fdbb1d281738b7b772e2cacb49c","status":"affected","versionType":"git"},{"version":"c18d4b190a46651726c9a952667c74d2deb33c28","lessThan":"1a86a1f7d88996085934139fa4c063b6299a2dd3","status":"affected","versionType":"git"}]},{"product":"Linux","vendor":"Linux","defaultStatus":"affected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["net/core/dev.c"],"versions":[{"version":"6.19","status":"affected"},{"version":"0","lessThan":"6.19","status":"unaffected","versionType":"semver"},{"version":"6.19.9","lessThanOrEqual":"6.19.*","status":"unaffected","versionType":"semver"},{"version":"7.0","lessThanOrEqual":"*","status":"unaffected","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"operator":"OR","negate":false,"cpeMatch":[{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"6.19","versionEndExcluding":"6.19.9"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"6.19","versionEndExcluding":"7.0"}]}]}],"references":[{"url":"https://git.kernel.org/stable/c/52459201d0df3fdbb1d281738b7b772e2cacb49c"},{"url":"https://git.kernel.org/stable/c/1a86a1f7d88996085934139fa4c063b6299a2dd3"}],"title":"net: Fix rcu_tasks stall in threaded busypoll","x_generator":{"engine":"bippy-1.2.0"}}}}