| Message ID | 20220217181902.808742-1-eric.dumazet@gmail.com (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 75134f16e7dd0007aa474b281935c5f42e79f2c8 |
| Delegated to: | BPF |
| Headers | show |
| Series | [bpf] bpf: add schedule points in batch ops | expand |
On 02/17, Eric Dumazet wrote: > From: Eric Dumazet <edumazet@google.com> > syzbot reported various soft lockups caused by bpf batch operations. > INFO: task kworker/1:1:27 blocked for more than 140 seconds. > INFO: task hung in rcu_barrier > Nothing prevents batch ops to process huge amount of data, > we need to add schedule points in them. > Note that maybe_wait_bpf_programs(map) calls from > generic_map_delete_batch() can be factorized by moving > the call after the loop. > This will be done later in -next tree once we get this fix merged, > unless there is strong opinion doing this optimization sooner. > Fixes: aa2e93b8e58e ("bpf: Add generic support for update and delete > batch ops") > Fixes: cb4d03ab499d ("bpf: Add generic support for lookup batch op") > Signed-off-by: Eric Dumazet <edumazet@google.com> > Cc: Brian Vazquez <brianvv@google.com> > Cc: Stanislav Fomichev <sdf@google.com> Looks good, thank you! Reviewed-by: Stanislav Fomichev <sdf@google.com> > Reported-by: syzbot <syzkaller@googlegroups.com> > --- > kernel/bpf/syscall.c | 3 +++ > 1 file changed, 3 insertions(+) > diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c > index > fa4505f9b6119bcb219ab9733847a98da65d1b21..ca70fe6fba387937dfb54f10826f19ac55a8a8e7 > 100644 > --- a/kernel/bpf/syscall.c > +++ b/kernel/bpf/syscall.c > @@ -1355,6 +1355,7 @@ int generic_map_delete_batch(struct bpf_map *map, > maybe_wait_bpf_programs(map); > if (err) > break; > + cond_resched(); > } > if (copy_to_user(&uattr->batch.count, &cp, sizeof(cp))) > err = -EFAULT; > @@ -1412,6 +1413,7 @@ int generic_map_update_batch(struct bpf_map *map, > if (err) > break; > + cond_resched(); > } > if (copy_to_user(&uattr->batch.count, &cp, sizeof(cp))) > @@ -1509,6 +1511,7 @@ int generic_map_lookup_batch(struct bpf_map *map, > swap(prev_key, key); > retry = MAP_LOOKUP_RETRIES; > cp++; > + cond_resched(); > } > if (err == -EFAULT) > -- > 2.35.1.265.g69c8d7142f-goog
Acked-by: Brian Vazquez <brianvv@google.com> On Thu, Feb 17, 2022 at 10:19 AM Eric Dumazet <eric.dumazet@gmail.com> wrote: > > From: Eric Dumazet <edumazet@google.com> > > syzbot reported various soft lockups caused by bpf batch operations. > > INFO: task kworker/1:1:27 blocked for more than 140 seconds. > INFO: task hung in rcu_barrier > > Nothing prevents batch ops to process huge amount of data, > we need to add schedule points in them. > > Note that maybe_wait_bpf_programs(map) calls from > generic_map_delete_batch() can be factorized by moving > the call after the loop. > > This will be done later in -next tree once we get this fix merged, > unless there is strong opinion doing this optimization sooner. > > Fixes: aa2e93b8e58e ("bpf: Add generic support for update and delete batch ops") > Fixes: cb4d03ab499d ("bpf: Add generic support for lookup batch op") > Signed-off-by: Eric Dumazet <edumazet@google.com> > Cc: Brian Vazquez <brianvv@google.com> > Cc: Stanislav Fomichev <sdf@google.com> > Reported-by: syzbot <syzkaller@googlegroups.com> > --- > kernel/bpf/syscall.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c > index fa4505f9b6119bcb219ab9733847a98da65d1b21..ca70fe6fba387937dfb54f10826f19ac55a8a8e7 100644 > --- a/kernel/bpf/syscall.c > +++ b/kernel/bpf/syscall.c > @@ -1355,6 +1355,7 @@ int generic_map_delete_batch(struct bpf_map *map, > maybe_wait_bpf_programs(map); > if (err) > break; > + cond_resched(); > } > if (copy_to_user(&uattr->batch.count, &cp, sizeof(cp))) > err = -EFAULT; > @@ -1412,6 +1413,7 @@ int generic_map_update_batch(struct bpf_map *map, > > if (err) > break; > + cond_resched(); > } > > if (copy_to_user(&uattr->batch.count, &cp, sizeof(cp))) > @@ -1509,6 +1511,7 @@ int generic_map_lookup_batch(struct bpf_map *map, > swap(prev_key, key); > retry = MAP_LOOKUP_RETRIES; > cp++; > + cond_resched(); > } > > if (err == -EFAULT) > -- > 2.35.1.265.g69c8d7142f-goog >
Hello: This patch was applied to bpf/bpf.git (master) by Alexei Starovoitov <ast@kernel.org>: On Thu, 17 Feb 2022 10:19:02 -0800 you wrote: > From: Eric Dumazet <edumazet@google.com> > > syzbot reported various soft lockups caused by bpf batch operations. > > INFO: task kworker/1:1:27 blocked for more than 140 seconds. > INFO: task hung in rcu_barrier > > [...] Here is the summary with links: - [bpf] bpf: add schedule points in batch ops https://git.kernel.org/bpf/bpf/c/75134f16e7dd You are awesome, thank you!
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index fa4505f9b6119bcb219ab9733847a98da65d1b21..ca70fe6fba387937dfb54f10826f19ac55a8a8e7 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -1355,6 +1355,7 @@ int generic_map_delete_batch(struct bpf_map *map, maybe_wait_bpf_programs(map); if (err) break; + cond_resched(); } if (copy_to_user(&uattr->batch.count, &cp, sizeof(cp))) err = -EFAULT; @@ -1412,6 +1413,7 @@ int generic_map_update_batch(struct bpf_map *map, if (err) break; + cond_resched(); } if (copy_to_user(&uattr->batch.count, &cp, sizeof(cp))) @@ -1509,6 +1511,7 @@ int generic_map_lookup_batch(struct bpf_map *map, swap(prev_key, key); retry = MAP_LOOKUP_RETRIES; cp++; + cond_resched(); } if (err == -EFAULT)