Hello,
We are developing a tool to perform static analysis on the bpf
subsystem to detect locking violations. Our tool reported the
raw_spin_lock_irqsave() in bpf_common_lru_push_free(). This function
is used by htab_lru_map_update_elem() and htab_lru_map_delete_elem(),
both of which can be called from NMI. A deadlock can happen if a bpf
program holding the lock is interrupted by the same program in NMI.
The report was generated for kernel version 6.6-rc4, however, we
believe this should still exist in the latest kernel.
We tried to validate the report on v6.10 by running a PoC. Below is
the lockdep splat. The PoC is attached at the end.
While executing the PoC lockdep makes two other reports. The first one
is for the raw_spin_lock_irqsave() usage in bpf_common_lru_pop_free()
for similar reasons to the above issue. The second report is on the
spin_lock used in htab_lock_bucket() which we believe to be a false
positive.
Thanks,
Priya
[ 3796.222898] ================================
[ 3796.222904] WARNING: inconsistent lock state
[ 3796.222906] 6.10.0-rc7+ #14 Not tainted
[ 3796.222909] --------------------------------
[ 3796.222914] inconsistent {INITIAL USE} -> {IN-NMI} usage.
[ 3796.222917] lru_perf/1458 [HC1[1]:SC0[0]:HE0:SE1] takes:
[ 3796.222923] ffffe8fffee26ab0 (&loc_l->lock){....}-{2:2}, at:
bpf_lru_push_free+0x241/0x510
[ 3796.222997] {INITIAL USE} state was registered at:
[ 3796.223000] lock_acquire+0x1cf/0x5a0
[ 3796.223024] _raw_spin_lock_irqsave+0x57/0xa0
[ 3796.223058] bpf_lru_pop_free+0x5e5/0x1e00
[ 3796.223063] prealloc_lru_pop+0x24/0xd0
[ 3796.223073] htab_lru_map_update_elem+0x18d/0x810
[ 3796.223078] bpf_prog_47d4157ca618f90f_lru_tp+0x61/0xa1
[ 3796.223087] trace_call_bpf+0x235/0x820
[ 3796.223097] perf_trace_run_bpf_submit+0x8a/0x230
[ 3796.223122] perf_trace_sched_switch+0x4c6/0x760
[ 3796.223142] __traceiter_sched_switch+0x77/0xe0
[ 3796.223155] __schedule+0x2109/0x6090
[ 3796.223171] schedule+0xf6/0x410
[ 3796.223175] irqentry_exit_to_user_mode+0xdb/0x210
[ 3796.223190] irqentry_exit+0x6f/0xa0
[ 3796.223195] sysvec_apic_timer_interrupt+0x5b/0xc0
[ 3796.223199] asm_sysvec_apic_timer_interrupt+0x1f/0x30
[ 3796.223212] irq event stamp: 20004
[ 3796.223214] hardirqs last enabled at (20003): [<ffffffff8410631f>]
irqentry_exit+0x6f/0xa0
[ 3796.223220] hardirqs last disabled at (20004): [<ffffffff84101a76>]
exc_nmi+0x106/0x2a0
[ 3796.223226] softirqs last enabled at (19984): [<ffffffff817c8c0c>]
bpf_perf_link_attach+0x33c/0x540
[ 3796.223235] softirqs last disabled at (19982): [<ffffffff817c8bcf>]
bpf_perf_link_attach+0x2ff/0x540
[ 3796.223239]
other info that might help us debug this:
[ 3796.223244] Possible unsafe locking scenario:
[ 3796.223246] CPU0
[ 3796.223248] ----
[ 3796.223249] lock(&loc_l->lock);
[ 3796.223253] <Interrupt>
[ 3796.223254] lock(&loc_l->lock);
[ 3796.223257]
*** DEADLOCK ***
[ 3796.223259] no locks held by lru_perf/1458.
[ 3796.223261]
stack backtrace:
[ 3796.223268] CPU: 12 PID: 1458 Comm: lru_perf Not tainted 6.10.0-rc7+ #14
[ 3796.223273] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996),
BIOS 1.13.0-1ubuntu1.1 04/01/2014
[ 3796.223277] Call Trace:
[ 3796.223283] <TASK>
[ 3796.223287] dump_stack_lvl+0x9f/0xf0
[ 3796.223313] dump_stack+0x14/0x20
[ 3796.223318] print_usage_bug.part.0+0x3ff/0x680
[ 3796.223326] lock_acquire+0x4e1/0x5a0
[ 3796.223331] ? __pfx_lock_acquire+0x10/0x10
[ 3796.223336] ? bpf_lru_push_free+0x241/0x510
[ 3796.223343] ? __pfx_do_raw_spin_trylock+0x10/0x10
[ 3796.223353] _raw_spin_lock_irqsave+0x57/0xa0
[ 3796.223358] ? bpf_lru_push_free+0x241/0x510
[ 3796.223364] bpf_lru_push_free+0x241/0x510
[ 3796.223368] ? htab_unlock_bucket+0xcd/0x140
[ 3796.223375] htab_lru_map_delete_elem+0x2f9/0x480
[ 3796.223382] ? __pfx_htab_lru_map_delete_elem+0x10/0x10
[ 3796.223392] bpf_prog_73160903ac17fb89_lru_perf+0x9c/0xa1
[ 3796.223397] __perf_event_overflow+0x239/0xb30
[ 3796.223401] ? x86_perf_event_set_period+0x289/0x5c0
[ 3796.223428] ? __pfx___perf_event_overflow+0x10/0x10
[ 3796.223439] perf_event_overflow+0x1d/0x30
[ 3796.223444] handle_pmi_common+0x540/0xa10
[ 3796.223459] ? __lock_acquire+0x17f3/0x6630
[ 3796.223467] ? __pfx_handle_pmi_common+0x10/0x10
[ 3796.223475] ? __pfx___lock_acquire+0x10/0x10
[ 3796.223483] ? kvm_sched_clock_read+0x15/0x30
[ 3796.223488] ? sched_clock_noinstr+0xd/0x20
[ 3796.223494] ? __this_cpu_preempt_check+0x17/0x20
[ 3796.223503] ? __pfx_lock_release+0x10/0x10
[ 3796.223508] ? debug_smp_processor_id+0x1b/0x30
[ 3796.223514] ? intel_bts_interrupt+0xf0/0x590
[ 3796.223519] ? __pfx_intel_bts_interrupt+0x10/0x10
[ 3796.223526] ? debug_smp_processor_id+0x1b/0x30
[ 3796.223533] intel_pmu_handle_irq+0x23f/0xb10
[ 3796.223543] perf_event_nmi_handler+0x42/0x70
[ 3796.223549] nmi_handle+0x165/0x490
[ 3796.223571] default_do_nmi+0x71/0x190
[ 3796.223577] exc_nmi+0x1cd/0x2a0
[ 3796.223582] asm_exc_nmi+0xbc/0x105
[ 3796.223587] RIP: 0033:0x55d6a3668601
[ 3796.223606] Code: ff 00 75 25 48 8b 05 5e 9a 04 00 48 89 c1 ba 1a
00 00 00 be 01 00 00 00 48 8d 05 1d 1c 03 00 48 89 c7 e8 e2 f6 ff ff
eb 10 90 <0f> b6 05 41 9a 04 00 83 f0 01 84 c0 75 f2 90 48 83 bd 08 fe
ff ff
[ 3796.223611] RSP: 002b:00007fff330b1790 EFLAGS: 00000202
[ 3796.223623] RAX: 0000000000000001 RBX: 000055d6da67e038 RCX: 0000000000000000
[ 3796.223626] RDX: 000000055d6da67d RSI: 000055d6da67c010 RDI: 0000000000000007
[ 3796.223630] RBP: 00007fff330b19c0 R08: 000055d6da67df60 R09: 000055d6da67c2e0
[ 3796.223633] R10: 0000000000000000 R11: 835347dc077a8333 R12: 00007fff330b1ad8
[ 3796.223636] R13: 000055d6a3668084 R14: 000055d6a36af9d8 R15: 00007f9020938040
[ 3796.223646] </TASK>
The lockdep warning can be triggered using the following bpf and
userspace programs:
==========================================================
#include "vmlinux.h"
#include <linux/version.h>
#include <bpf/bpf_helpers.h>
#include <bpf/bpf_tracing.h>
#include <bpf/bpf_core_read.h>
struct {
__uint(type, BPF_MAP_TYPE_LRU_HASH);
__type(key, int);
__type(value,int);
__uint(max_entries, 255);
} pb SEC(".maps");
SEC("perf_event")
int lru_perf(void *ctx)
{
int key = 2;
int init_val = 1;
long *value;
int i;
bpf_printk("lru_perf");
bpf_map_update_elem(&pb, &key, &init_val, BPF_ANY);
value = bpf_map_lookup_elem(&pb, &key);
int ret = bpf_map_delete_elem(&pb, &key);
return 0;
}
SEC("tp/sched/sched_switch")
int lru_tp(void *ctx)
{
int key = 2;
int init_val = 1;
long *value;
int i;
bpf_printk("lru_tp");
bpf_map_update_elem(&pb, &key, &init_val, BPF_ANY);
value = bpf_map_lookup_elem(&pb, &key);
int ret = bpf_map_delete_elem(&pb, &key);
return 0;
}
char _license[] SEC("license") = "GPL";
============================================================
#include <unistd.h>
#include <sys/syscall.h>
#include <linux/perf_event.h>
#include <bpf/libbpf.h>
#include <bpf/bpf.h>
#include <sys/resource.h>
#include <signal.h>
#include "lru_perf.skel.h"
static volatile bool exiting = false;
static void sig_handler(int sig)
{
exiting = true;
return;
}
extern int parse_cpu_mask_file(const char *fcpu, bool **mask, int *mask_sz);
static long perf_event_open(struct perf_event_attr *hw_event, pid_t
pid, int cpu, int group_fd,
unsigned long flags)
{
int ret;
ret = syscall(__NR_perf_event_open, hw_event, pid, cpu,
group_fd, flags);
return ret;
}
void bump_memlock_rlimit(void)
{
struct rlimit rlim_new = {
.rlim_cur = RLIM_INFINITY,
.rlim_max = RLIM_INFINITY,
};
if (setrlimit(RLIMIT_MEMLOCK, &rlim_new)) {
fprintf(stderr, "Failed to increase RLIMIT_MEMLOCK limit!\n");
exit(1);
}
return;
}
int main(int argc, char *const argv[])
{
const char *online_cpus_file = "/sys/devices/system/cpu/online";
int cpu;
struct lru_perf_bpf *skel = NULL;
struct perf_event_attr attr;
struct bpf_link **links = NULL;
int num_cpus, num_online_cpus;
int *pefds = NULL, pefd;
int i, err = 0;
bool *online_mask = NULL;
struct bpf_program *prog;
struct bpf_object *obj;
struct bpf_map *map;
char filename[256];
bump_memlock_rlimit();
signal(SIGINT, sig_handler);
signal(SIGTERM, sig_handler);
err = parse_cpu_mask_file(online_cpus_file, &online_mask,
&num_online_cpus);
if (err) {
fprintf(stderr, "Fail to get online CPU numbers: %d\n", err);
goto cleanup;
}
num_cpus = libbpf_num_possible_cpus();
if (num_cpus <= 0) {
fprintf(stderr, "Fail to get the number of processors\n");
err = -1;
goto cleanup;
}
snprintf(filename, sizeof(filename), ".output/lru_perf.bpf.o");
obj = bpf_object__open_file(filename, NULL);
if (libbpf_get_error(obj)) {
fprintf(stderr, "ERROR: opening BPF object file failed\n");
goto cleanup;
}
map = bpf_object__find_map_by_name(obj, "pb");
if (libbpf_get_error(map)) {
fprintf(stderr, "ERROR: finding a map in obj file failed\n");
goto cleanup;
}
if (bpf_object__load(obj)) {
fprintf(stderr, "ERROR: loading BPF object file failed\n");
goto cleanup;
}
pefds = malloc(num_cpus * sizeof(int));
for (i = 0; i < num_cpus; i++) {
pefds[i] = -1;
}
links = calloc(num_cpus, sizeof(struct bpf_link *));
memset(&attr, 0, sizeof(attr));
attr.type = PERF_TYPE_HARDWARE;
attr.config = PERF_COUNT_HW_CPU_CYCLES;
attr.sample_freq = 10;
attr.inherit = 1;
attr.freq = 1;
for (cpu = 0; cpu < 2; cpu++) {
//skip offline/not present CPUs
if (cpu >= num_online_cpus || !online_mask[cpu])
continue;
// Set up performance monitoring on a CPU/Core
pefd = perf_event_open(&attr, 0, -1, -1, 0);
if (pefd < 0) {
fprintf(stderr, "Fail to set up performance
monitor on a CPU/Core\n");
err = -1;
goto cleanup;
}
pefds[cpu] = pefd;
prog = bpf_object__find_program_by_name(obj, "lru_perf");
if (!prog) {
fprintf(stderr, "ERROR: finding a prog in obj
file failed\n");
goto cleanup;
}
links[cpu] = bpf_program__attach_perf_event(prog, pefds[cpu]);
if (!links[cpu]) {
err = -1;
fprintf(stderr, "ERROR: bpf_program__attach failed\n");
goto cleanup;
}
}
struct bpf_program *prog_tp =
bpf_object__find_program_by_name(obj, "lru_tp");
if (!prog_tp) {
fprintf(stderr, "ERROR: finding tp prog in obj
file failed\n");
goto cleanup;
}
struct bpf_link *link_tp = bpf_program__attach(prog_tp);
if (!link_tp) {
fprintf(stderr, "ERROR: failed to attach tp");
goto cleanup;
}
while(!exiting){
}
cleanup:
if (links) {
for (cpu = 0; cpu < num_cpus; cpu++)
bpf_link__destroy(links[cpu]);
free(links);
}
if (pefds) {
for (i = 0; i < num_cpus; i++) {
if (pefds[i] >= 0)
close(pefds[i]);
}
free(pefds);
}
lru_perf_bpf__destroy(skel);
free(online_mask);
return -err;
}
