KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf

Status: moderation: reported on 2026/11/20 00:00
Subsystems: net
Labels: race:benign prio:low
[Documentation on labels]
Reported-by: syzbot+3812d1cc59d0b30cb5f0@syzkaller.appspotmail.com
First crash: 2d22h, last: 2d03h
✨ AI-generated root cause summary:

Commit 378b7708194f ("sched: Make migrate_{en,dis}able() inline") moved __migrate_disable() and __migrate_enable() from kernel/sched/core.c to include/linux/sched.h. In kernel/sched/, KCSAN sanitization was disabled (KCSAN_SANITIZE := n), but moving the helpers inline into the header exposed plain C accesses of p->migration_disabled to KCSAN instrumentation at callsites throughout the kernel (such as BPF helpers). In __migrate_disable(), guard(preempt)() disables preemption but leaves hardware interrupts enabled; an interrupt on the same CPU can trigger softirq processing (e.g. net_rx_action -> packet_rcv -> bpf_prog_run_clear_cb) under the same current task, causing an unannotated read of p->migration_disabled while the interrupted task is in the middle of writing to it.

▶ ▼ ✨ Full AI root cause analysis
Based on kernel commit 666539eecdf4:

Root Cause Analysis

1. Functions, Structs, and Files Involved

  • __migrate_disable() and __migrate_enable(): include/linux/sched.h:2479-2497 (and lines 2444-2477)
  • struct task_struct::migration_disabled: include/linux/sched.h:948
  • Callers: bpf_prog_run_array() (include/linux/bpf.h:2796 called from trace_call_bpf()), bpf_prog_run_pin_on_cpu() / bpf_prog_run_clear_cb() (include/linux/filter.h:836, 1119 called from run_filter() / packet_rcv()).
  • Commit introducing the issue: 378b7708194f ("sched: Make migrate_{en,dis}able() inline").

2. Technical Mechanism and Trigger

In commit 378b7708194f, migrate_disable() and migrate_enable() were moved inline into include/linux/sched.h (as __migrate_disable() and __migrate_enable()) so that BPF hooks and other fast paths could avoid function call overhead. Previously, these functions resided in kernel/sched/core.c, where KCSAN instrumentation was disabled by KCSAN_SANITIZE := n in kernel/sched/Makefile to avoid noise on scheduler internal fast paths.

Once inlined into include/linux/sched.h, the logic became part of every caller's compilation unit (including BPF tracepoint handlers in kernel/trace/bpf_trace.c and network packet receive filters in net/packet/af_packet.c), which are compiled with KCSAN instrumentation enabled.

The data race occurs as follows:

  1. Task Context (CPU 0): A user task (PID 3710) executes a syscall (getsockopt) which allocates memory (vzalloc()) and fires a tracepoint (trace_kmalloc()). The tracepoint invokes trace_call_bpf() -> bpf_prog_run_array(), which calls migrate_disable() -> __migrate_disable(). In __migrate_disable():

    static inline void __migrate_disable(void)
    {
        struct task_struct *p = current;
    
        if (p->migration_disabled) {
            p->migration_disabled++;
            return;
        }
    
        guard(preempt)();
        this_rq_pinned()++;
        p->migration_disabled = 1; /* Line 2496: plain 2-byte write */
    }
    

    At line 2494, guard(preempt)() disables preemption (preempt_disable()), but hardware interrupts remain enabled. At line 2496, the task performs a plain, unannotated 2-byte write to p->migration_disabled. Under KCSAN, a watchpoint is installed around this write, inserting a small delay (udelay()).

  2. Interrupt / Softirq Context (CPU 0): While the interrupted task is paused at line 2496, a hardware interrupt arrives on CPU 0. Upon interrupt exit, __irq_exit_rcu() processes pending softirqs (invoke_softirq() -> __do_softirq() -> net_rx_action()). In softirq context, current is still the interrupted task (PID 3710). The network receive path processes an incoming packet: packet_rcv() -> run_filter() -> bpf_prog_run_clear_cb() -> bpf_prog_run_pin_on_cpu() -> migrate_disable(). In __migrate_disable():

    if (p->migration_disabled) /* Line 2483: plain 2-byte read */
    

    The softirq performs a plain, unannotated 2-byte read of p->migration_disabled on the same current task while the watchpoint on the write at line 2496 is active.

  3. Sanitizer Detection: KCSAN detects a data race between the write at include/linux/sched.h:2496 and the read at include/linux/sched.h:2483 due to the lack of concurrency annotations (data_race() or READ_ONCE() / WRITE_ONCE()). In production environments without KCSAN, this access pattern is benign because 2-byte aligned operations are naturally atomic on x86, and the softirq strictly runs to completion and pairs its migrate_disable() with migrate_enable() before execution returns to the interrupted task context.

▶ ▼ ✨ AI Jobs (2)
ID Workflow Result Correct Bug Created Started Finished Revision
4c053cb7 assessment-security DenialOfService: ❌ Exploitable: ❌ FilesystemTrigger: ❌ FuzzingArtifact: ❌ MemoryCorruption: ❌ NetworkNeedsSetup: ❌ NetworkTrigger: ❌ PeripheralTrigger: ❌ RemoteTrigger: ❌ Unprivileged: ✅ UserNamespace: ✅ VMGuestTrigger: ❌ VMHostTrigger: ❌ ❓ KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf 2026/11/19 08:14 2026/11/19 08:14 2026/11/19 08:53 2006bdcc
ec6f519f assessment-kcsan Benign: ✅ ❓ KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf 2026/11/19 08:06 2026/11/19 08:06 2026/11/19 08:42 2006bdcc

Sample crash report:
==================================================================
BUG: KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf

read to 0xffff88810244ebf8 of 2 bytes by interrupt on cpu 0:
 __migrate_disable include/linux/sched.h:2483 [inline]
 migrate_disable include/linux/sched.h:2512 [inline]
 bpf_prog_run_pin_on_cpu include/linux/filter.h:836 [inline]
 bpf_prog_run_clear_cb+0x53/0x290 include/linux/filter.h:1119
 run_filter net/packet/af_packet.c:2111 [inline]
 packet_rcv+0x26e/0xba0 net/packet/af_packet.c:2184
 deliver_skb+0x10b/0x170 net/core/dev.c:2564
 deliver_ptype_list_skb net/core/dev.c:2579 [inline]
 __netif_receive_skb_core+0x13e0/0x19a0 net/core/dev.c:6391
 __netif_receive_skb_list_core+0x112/0x520 net/core/dev.c:6522
 __netif_receive_skb_list net/core/dev.c:6589 [inline]
 netif_receive_skb_list_internal+0x471/0x5e0 net/core/dev.c:6680
 gro_normal_list include/net/gro.h:539 [inline]
 gro_flush_normal include/net/gro.h:547 [inline]
 napi_complete_done+0x19c/0x410 net/core/dev.c:7048
 virtqueue_napi_complete drivers/net/virtio_net.c:736 [inline]
 virtnet_poll+0x1a9a/0x1f60 drivers/net/virtio_net.c:3048
 __napi_poll+0x73/0x350 net/core/dev.c:7997
 napi_poll net/core/dev.c:8060 [inline]
 net_rx_action+0x33f/0x850 net/core/dev.c:8217
 handle_softirqs+0x109/0x310 kernel/softirq.c:669
 __do_softirq kernel/softirq.c:703 [inline]
 invoke_softirq kernel/softirq.c:542 [inline]
 __irq_exit_rcu+0x40/0xf0 kernel/softirq.c:786
 common_interrupt+0x83/0x90 arch/x86/kernel/irq.c:273
 asm_common_interrupt+0x26/0x40 arch/x86/include/asm/idtentry.h:665
 __preempt_count_add arch/x86/include/asm/preempt.h:98 [inline]
 delay_tsc+0x4b/0xa0 arch/x86/lib/delay.c:77
 udelay include/asm-generic/delay.h:64 [inline]
 delay_access kernel/kcsan/core.c:332 [inline]
 kcsan_setup_watchpoint+0x255/0x3f0 kernel/kcsan/core.c:622
 __migrate_disable include/linux/sched.h:2496 [inline]
 migrate_disable include/linux/sched.h:2512 [inline]
 bpf_prog_run_array include/linux/bpf.h:2796 [inline]
 trace_call_bpf+0xee/0x420 kernel/trace/bpf_trace.c:147
 perf_trace_run_bpf_submit+0x4e/0x110 kernel/events/core.c:11549
 do_perf_trace_kmalloc include/trace/events/kmem.h:57 [inline]
 perf_trace_kmalloc+0xec/0x130 include/trace/events/kmem.h:57
 __do_trace_kmalloc include/trace/events/kmem.h:57 [inline]
 trace_kmalloc include/trace/events/kmem.h:57 [inline]
 __do_kmalloc_node mm/slub.c:5532 [inline]
 __kmalloc_node_noprof+0x45d/0x4d0 mm/slub.c:5543
 _kmalloc_node_noprof include/linux/slab.h:1200 [inline]
 __vmalloc_area_node mm/vmalloc.c:3930 [inline]
 __vmalloc_node_range_noprof+0x458/0x10b0 mm/vmalloc.c:4152
 __vmalloc_node_noprof mm/vmalloc.c:4213 [inline]
 vzalloc_noprof+0x82/0xb0 mm/vmalloc.c:4291
 alloc_counters net/ipv4/netfilter/arp_tables.c:672 [inline]
 copy_entries_to_user net/ipv4/netfilter/arp_tables.c:693 [inline]
 get_entries net/ipv4/netfilter/arp_tables.c:788 [inline]
 do_arpt_get_ctl+0x64a/0xaf0 net/ipv4/netfilter/arp_tables.c:1014
 nf_getsockopt+0x190/0x1b0 net/netfilter/nf_sockopt.c:116
 ip_getsockopt+0x137/0x180 net/ipv4/ip_sockglue.c:1785
 tcp_getsockopt+0xad/0xe0 net/ipv4/tcp.c:4885
 sock_common_getsockopt+0x60/0x70 net/core/sock.c:4096
 do_sock_getsockopt+0x2ce/0x350 net/socket.c:2505
 __sys_getsockopt net/socket.c:2536 [inline]
 __do_sys_getsockopt net/socket.c:2543 [inline]
 __se_sys_getsockopt net/socket.c:2540 [inline]
 __x64_sys_getsockopt+0x11d/0x1a0 net/socket.c:2540
 x64_sys_call+0x1032/0x2550 arch/x86/include/generated/asm/syscalls_64.h:56
 do_syscall_x64 arch/x86/entry/syscall_64.c:61 [inline]
 do_syscall_64+0x11e/0x3a0 arch/x86/entry/syscall_64.c:89
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

write to 0xffff88810244ebf8 of 2 bytes by task 3710 on cpu 0:
 __migrate_disable include/linux/sched.h:2496 [inline]
 migrate_disable include/linux/sched.h:2512 [inline]
 bpf_prog_run_array include/linux/bpf.h:2796 [inline]
 trace_call_bpf+0xee/0x420 kernel/trace/bpf_trace.c:147
 perf_trace_run_bpf_submit+0x4e/0x110 kernel/events/core.c:11549
 do_perf_trace_kmalloc include/trace/events/kmem.h:57 [inline]
 perf_trace_kmalloc+0xec/0x130 include/trace/events/kmem.h:57
 __do_trace_kmalloc include/trace/events/kmem.h:57 [inline]
 trace_kmalloc include/trace/events/kmem.h:57 [inline]
 __do_kmalloc_node mm/slub.c:5532 [inline]
 __kmalloc_node_noprof+0x45d/0x4d0 mm/slub.c:5543
 _kmalloc_node_noprof include/linux/slab.h:1200 [inline]
 __vmalloc_area_node mm/vmalloc.c:3930 [inline]
 __vmalloc_node_range_noprof+0x458/0x10b0 mm/vmalloc.c:4152
 __vmalloc_node_noprof mm/vmalloc.c:4213 [inline]
 vzalloc_noprof+0x82/0xb0 mm/vmalloc.c:4291
 alloc_counters net/ipv4/netfilter/arp_tables.c:672 [inline]
 copy_entries_to_user net/ipv4/netfilter/arp_tables.c:693 [inline]
 get_entries net/ipv4/netfilter/arp_tables.c:788 [inline]
 do_arpt_get_ctl+0x64a/0xaf0 net/ipv4/netfilter/arp_tables.c:1014
 nf_getsockopt+0x190/0x1b0 net/netfilter/nf_sockopt.c:116
 ip_getsockopt+0x137/0x180 net/ipv4/ip_sockglue.c:1785
 tcp_getsockopt+0xad/0xe0 net/ipv4/tcp.c:4885
 sock_common_getsockopt+0x60/0x70 net/core/sock.c:4096
 do_sock_getsockopt+0x2ce/0x350 net/socket.c:2505
 __sys_getsockopt net/socket.c:2536 [inline]
 __do_sys_getsockopt net/socket.c:2543 [inline]
 __se_sys_getsockopt net/socket.c:2540 [inline]
 __x64_sys_getsockopt+0x11d/0x1a0 net/socket.c:2540
 x64_sys_call+0x1032/0x2550 arch/x86/include/generated/asm/syscalls_64.h:56
 do_syscall_x64 arch/x86/entry/syscall_64.c:61 [inline]
 do_syscall_64+0x11e/0x3a0 arch/x86/entry/syscall_64.c:89
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Reported by Kernel Concurrency Sanitizer on:
CPU: 0 UID: 0 PID: 3710 Comm: syz-executor Not tainted syzkaller #0 PREEMPT(lazy) 
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 11/06/2026
==================================================================

Crashes (58):
Time Kernel Commit Syzkaller Config Log Report Syz repro C repro VM info Assets (help?) Manager Title
2026/11/17 20:01 upstream 666539eecdf4 122f94d5 .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 18:30 upstream 666539eecdf4 122f94d5 .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 16:25 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 16:20 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 16:16 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 16:16 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 16:13 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 16:13 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 16:12 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 16:12 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 16:10 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 15:11 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 14:41 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 12:38 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 12:19 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 12:12 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 12:10 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 12:04 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 11:42 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 11:23 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 11:16 upstream 666539eecdf4 43af6cac .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 08:13 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 08:10 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 07:42 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 06:50 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 06:49 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 06:41 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 06:41 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 06:40 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 06:19 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 06:18 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 04:37 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 04:33 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 04:13 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 04:13 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 04:12 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 03:32 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 03:15 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 03:07 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 03:05 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 02:46 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 02:45 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
2026/11/17 02:43 upstream 666539eecdf4 a3e64cfc .config console log report info [disk image] [vmlinux] [kernel image] ci2-upstream-kcsan-gce KCSAN: data-race in bpf_prog_run_clear_cb / trace_call_bpf
* Struck through repros no longer work on HEAD.