Syzkaller reports "memory leak in cpu_map_update_elem" in 5.10 stable release. The problem has been fixed by the following patches which can be cleanly applied to the 5.10 branch.
Found by Linux Verification Center (linuxtesting.org) with Syzkaller.
From: Pu Lehui pulehui@huawei.com
[ Upstream commit 4369016497319a9635702da010d02af1ebb1849d ]
Syzkaller reported a memory leak as follows:
BUG: memory leak unreferenced object 0xff110001198ef748 (size 192): comm "syz-executor.3", pid 17672, jiffies 4298118891 (age 9.906s) hex dump (first 32 bytes): 00 00 00 00 4a 19 00 00 80 ad e3 e4 fe ff c0 00 ....J........... 00 b2 d3 0c 01 00 11 ff 28 f5 8e 19 01 00 11 ff ........(....... backtrace: [<ffffffffadd28087>] __cpu_map_entry_alloc+0xf7/0xb00 [<ffffffffadd28d8e>] cpu_map_update_elem+0x2fe/0x3d0 [<ffffffffadc6d0fd>] bpf_map_update_value.isra.0+0x2bd/0x520 [<ffffffffadc7349b>] map_update_elem+0x4cb/0x720 [<ffffffffadc7d983>] __se_sys_bpf+0x8c3/0xb90 [<ffffffffb029cc80>] do_syscall_64+0x30/0x40 [<ffffffffb0400099>] entry_SYSCALL_64_after_hwframe+0x61/0xc6
BUG: memory leak unreferenced object 0xff110001198ef528 (size 192): comm "syz-executor.3", pid 17672, jiffies 4298118891 (age 9.906s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [<ffffffffadd281f0>] __cpu_map_entry_alloc+0x260/0xb00 [<ffffffffadd28d8e>] cpu_map_update_elem+0x2fe/0x3d0 [<ffffffffadc6d0fd>] bpf_map_update_value.isra.0+0x2bd/0x520 [<ffffffffadc7349b>] map_update_elem+0x4cb/0x720 [<ffffffffadc7d983>] __se_sys_bpf+0x8c3/0xb90 [<ffffffffb029cc80>] do_syscall_64+0x30/0x40 [<ffffffffb0400099>] entry_SYSCALL_64_after_hwframe+0x61/0xc6
BUG: memory leak unreferenced object 0xff1100010fd93d68 (size 8): comm "syz-executor.3", pid 17672, jiffies 4298118891 (age 9.906s) hex dump (first 8 bytes): 00 00 00 00 00 00 00 00 ........ backtrace: [<ffffffffade5db3e>] kvmalloc_node+0x11e/0x170 [<ffffffffadd28280>] __cpu_map_entry_alloc+0x2f0/0xb00 [<ffffffffadd28d8e>] cpu_map_update_elem+0x2fe/0x3d0 [<ffffffffadc6d0fd>] bpf_map_update_value.isra.0+0x2bd/0x520 [<ffffffffadc7349b>] map_update_elem+0x4cb/0x720 [<ffffffffadc7d983>] __se_sys_bpf+0x8c3/0xb90 [<ffffffffb029cc80>] do_syscall_64+0x30/0x40 [<ffffffffb0400099>] entry_SYSCALL_64_after_hwframe+0x61/0xc6
In the cpu_map_update_elem flow, when kthread_stop is called before calling the threadfn of rcpu->kthread, since the KTHREAD_SHOULD_STOP bit of kthread has been set by kthread_stop, the threadfn of rcpu->kthread will never be executed, and rcpu->refcnt will never be 0, which will lead to the allocated rcpu, rcpu->queue and rcpu->queue->queue cannot be released.
Calling kthread_stop before executing kthread's threadfn will return -EINTR. We can complete the release of memory resources in this state.
Fixes: 6710e1126934 ("bpf: introduce new bpf cpu map type BPF_MAP_TYPE_CPUMAP") Signed-off-by: Pu Lehui pulehui@huawei.com Acked-by: Jesper Dangaard Brouer hawk@kernel.org Acked-by: Hou Tao houtao1@huawei.com Link: https://lore.kernel.org/r/20230711115848.2701559-1-pulehui@huaweicloud.com Signed-off-by: Alexei Starovoitov ast@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org Signed-off-by: Alexey Panov apanov@astralinux.ru --- kernel/bpf/cpumap.c | 40 ++++++++++++++++++++++++---------------- 1 file changed, 24 insertions(+), 16 deletions(-)
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c index c61a23b564aa..2097ed9bd583 100644 --- a/kernel/bpf/cpumap.c +++ b/kernel/bpf/cpumap.c @@ -139,22 +139,6 @@ static void get_cpu_map_entry(struct bpf_cpu_map_entry *rcpu) atomic_inc(&rcpu->refcnt); }
-/* called from workqueue, to workaround syscall using preempt_disable */ -static void cpu_map_kthread_stop(struct work_struct *work) -{ - struct bpf_cpu_map_entry *rcpu; - - rcpu = container_of(work, struct bpf_cpu_map_entry, kthread_stop_wq); - - /* Wait for flush in __cpu_map_entry_free(), via full RCU barrier, - * as it waits until all in-flight call_rcu() callbacks complete. - */ - rcu_barrier(); - - /* kthread_stop will wake_up_process and wait for it to complete */ - kthread_stop(rcpu->kthread); -} - static struct sk_buff *cpu_map_build_skb(struct xdp_frame *xdpf, struct sk_buff *skb) { @@ -225,6 +209,30 @@ static void put_cpu_map_entry(struct bpf_cpu_map_entry *rcpu) } }
+/* called from workqueue, to workaround syscall using preempt_disable */ +static void cpu_map_kthread_stop(struct work_struct *work) +{ + struct bpf_cpu_map_entry *rcpu; + int err; + + rcpu = container_of(work, struct bpf_cpu_map_entry, kthread_stop_wq); + + /* Wait for flush in __cpu_map_entry_free(), via full RCU barrier, + * as it waits until all in-flight call_rcu() callbacks complete. + */ + rcu_barrier(); + + /* kthread_stop will wake_up_process and wait for it to complete */ + err = kthread_stop(rcpu->kthread); + if (err) { + /* kthread_stop may be called before cpu_map_kthread_run + * is executed, so we need to release the memory related + * to rcpu. + */ + put_cpu_map_entry(rcpu); + } +} + static int cpu_map_bpf_prog_run_xdp(struct bpf_cpu_map_entry *rcpu, void **frames, int n, struct xdp_cpumap_stats *stats)
From: Hou Tao houtao1@huawei.com
commit 640a604585aa30f93e39b17d4d6ba69fcb1e66c9 upstream.
The following warning was reported when running stress-mode enabled xdp_redirect_cpu with some RT threads:
------------[ cut here ]------------ WARNING: CPU: 4 PID: 65 at kernel/bpf/cpumap.c:135 CPU: 4 PID: 65 Comm: kworker/4:1 Not tainted 6.5.0-rc2+ #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996) Workqueue: events cpu_map_kthread_stop RIP: 0010:put_cpu_map_entry+0xda/0x220 ...... Call Trace: <TASK> ? show_regs+0x65/0x70 ? __warn+0xa5/0x240 ...... ? put_cpu_map_entry+0xda/0x220 cpu_map_kthread_stop+0x41/0x60 process_one_work+0x6b0/0xb80 worker_thread+0x96/0x720 kthread+0x1a5/0x1f0 ret_from_fork+0x3a/0x70 ret_from_fork_asm+0x1b/0x30 </TASK>
The root cause is the same as commit 436901649731 ("bpf: cpumap: Fix memory leak in cpu_map_update_elem"). The kthread is stopped prematurely by kthread_stop() in cpu_map_kthread_stop(), and kthread() doesn't call cpu_map_kthread_run() at all but XDP program has already queued some frames or skbs into ptr_ring. So when __cpu_map_ring_cleanup() checks the ptr_ring, it will find it was not emptied and report a warning.
An alternative fix is to use __cpu_map_ring_cleanup() to drop these pending frames or skbs when kthread_stop() returns -EINTR, but it may confuse the user, because these frames or skbs have been handled correctly by XDP program. So instead of dropping these frames or skbs, just make sure the per-cpu kthread is running before __cpu_map_entry_alloc() returns.
After apply the fix, the error handle for kthread_stop() will be unnecessary because it will always return 0, so just remove it.
Fixes: 6710e1126934 ("bpf: introduce new bpf cpu map type BPF_MAP_TYPE_CPUMAP") Signed-off-by: Hou Tao houtao1@huawei.com Reviewed-by: Pu Lehui pulehui@huawei.com Acked-by: Jesper Dangaard Brouer hawk@kernel.org Link: https://lore.kernel.org/r/20230729095107.1722450-2-houtao@huaweicloud.com Signed-off-by: Martin KaFai Lau martin.lau@kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Alexey Panov apanov@astralinux.ru --- kernel/bpf/cpumap.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-)
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c index 2097ed9bd583..f54b7321a7bc 100644 --- a/kernel/bpf/cpumap.c +++ b/kernel/bpf/cpumap.c @@ -25,6 +25,7 @@ #include <linux/workqueue.h> #include <linux/kthread.h> #include <linux/capability.h> +#include <linux/completion.h> #include <trace/events/xdp.h>
#include <linux/netdevice.h> /* netif_receive_skb_core */ @@ -69,6 +70,7 @@ struct bpf_cpu_map_entry { struct rcu_head rcu;
struct work_struct kthread_stop_wq; + struct completion kthread_running; };
struct bpf_cpu_map { @@ -213,7 +215,6 @@ static void put_cpu_map_entry(struct bpf_cpu_map_entry *rcpu) static void cpu_map_kthread_stop(struct work_struct *work) { struct bpf_cpu_map_entry *rcpu; - int err;
rcpu = container_of(work, struct bpf_cpu_map_entry, kthread_stop_wq);
@@ -223,14 +224,7 @@ static void cpu_map_kthread_stop(struct work_struct *work) rcu_barrier();
/* kthread_stop will wake_up_process and wait for it to complete */ - err = kthread_stop(rcpu->kthread); - if (err) { - /* kthread_stop may be called before cpu_map_kthread_run - * is executed, so we need to release the memory related - * to rcpu. - */ - put_cpu_map_entry(rcpu); - } + kthread_stop(rcpu->kthread); }
static int cpu_map_bpf_prog_run_xdp(struct bpf_cpu_map_entry *rcpu, @@ -308,6 +302,7 @@ static int cpu_map_kthread_run(void *data) { struct bpf_cpu_map_entry *rcpu = data;
+ complete(&rcpu->kthread_running); set_current_state(TASK_INTERRUPTIBLE);
/* When kthread gives stop order, then rcpu have been disconnected @@ -462,6 +457,7 @@ __cpu_map_entry_alloc(struct bpf_cpumap_val *value, u32 cpu, int map_id) goto free_ptr_ring;
/* Setup kthread */ + init_completion(&rcpu->kthread_running); rcpu->kthread = kthread_create_on_node(cpu_map_kthread_run, rcpu, numa, "cpumap/%d/map:%d", cpu, map_id); if (IS_ERR(rcpu->kthread)) @@ -474,6 +470,12 @@ __cpu_map_entry_alloc(struct bpf_cpumap_val *value, u32 cpu, int map_id) kthread_bind(rcpu->kthread, cpu); wake_up_process(rcpu->kthread);
+ /* Make sure kthread has been running, so kthread_stop() will not + * stop the kthread prematurely and all pending frames or skbs + * will be handled by the kthread before kthread_stop() returns. + */ + wait_for_completion(&rcpu->kthread_running); + return rcpu;
free_prog:
linux-stable-mirror@lists.linaro.org
-
Alexey Panov