Re: [PATCH bpf-next] bpf: add missing map_delete_elem method to bloom filter map

[Yonghong Song <yhs@fb.com>]

On 10/31/21 10:13 AM, Eric Dumazet wrote:
> From: Eric Dumazet <edumazet@google.com>
> 
> Without it, kernel crashes in map_delete_elem(), as reported
> by syzbot.
> 
> BUG: kernel NULL pointer dereference, address: 0000000000000000
> PGD 72c97067 P4D 72c97067 PUD 1e20c067 PMD 0
> Oops: 0010 [#1] PREEMPT SMP KASAN
> CPU: 0 PID: 6518 Comm: syz-executor196 Not tainted 5.15.0-rc3-syzkaller #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> RIP: 0010:0x0
> Code: Unable to access opcode bytes at RIP 0xffffffffffffffd6.
> RSP: 0018:ffffc90002bafcb8 EFLAGS: 00010246
> RAX: dffffc0000000000 RBX: 1ffff92000575f9f RCX: 0000000000000000
> RDX: 1ffffffff1327aba RSI: 0000000000000000 RDI: ffff888025a30c00
> RBP: ffffc90002baff08 R08: 0000000000000000 R09: 0000000000000001
> R10: ffffffff818525d8 R11: 0000000000000000 R12: ffffffff8993d560
> R13: ffff888025a30c00 R14: ffff888024bc0000 R15: 0000000000000000
> FS:  0000555557491300(0000) GS:ffff8880b9c00000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: ffffffffffffffd6 CR3: 0000000070189000 CR4: 00000000003506f0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
>   map_delete_elem kernel/bpf/syscall.c:1220 [inline]
>   __sys_bpf+0x34f1/0x5ee0 kernel/bpf/syscall.c:4606
>   __do_sys_bpf kernel/bpf/syscall.c:4719 [inline]
>   __se_sys_bpf kernel/bpf/syscall.c:4717 [inline]
>   __x64_sys_bpf+0x75/0xb0 kernel/bpf/syscall.c:4717
>   do_syscall_x64 arch/x86/entry/common.c:50 [inline]
> 
> Fixes: 9330986c0300 ("bpf: Add bloom filter map implementation")
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Cc: Joanne Koong <joannekoong@fb.com>
> Cc: Alexei Starovoitov <ast@kernel.org>
> Cc: Andrii Nakryiko <andrii@kernel.org>
> Reported-by: syzbot <syzkaller@googlegroups.com>

LGTM with a suggestion below.

Acked-by: Yonghong Song <yhs@fb.com>

> ---
>   kernel/bpf/bloom_filter.c | 6 ++++++
>   1 file changed, 6 insertions(+)
> 
> diff --git a/kernel/bpf/bloom_filter.c b/kernel/bpf/bloom_filter.c
> index 7c50232b7571f3f038dd45b5c0bd7289125e6d43..31a7af15a83d74af1d88d04cc8d71aa7d403b4ef 100644
> --- a/kernel/bpf/bloom_filter.c
> +++ b/kernel/bpf/bloom_filter.c
> @@ -77,6 +77,11 @@ static int pop_elem(struct bpf_map *map, void *value)
>   	return -EOPNOTSUPP;
>   }
>   
> +static int delete_elem(struct bpf_map *map, void *value)
> +{
> +	return -EOPNOTSUPP;
> +}
> +
>   static struct bpf_map *map_alloc(union bpf_attr *attr)
>   {
>   	u32 bitset_bytes, bitset_mask, nr_hash_funcs, nr_bits;
> @@ -189,6 +194,7 @@ const struct bpf_map_ops bloom_filter_map_ops = {
>   	.map_pop_elem = pop_elem,
>   	.map_lookup_elem = lookup_elem,
>   	.map_update_elem = update_elem,
> +	.map_delete_elem = delete_elem,

There is a pending patch
https://lore.kernel.org/bpf/20211029224909.1721024-2-joannekoong@fb.com/T/#u
to rename say lookup_elem to bloom_map_lookup_elem.
I think we should change
this delete_elem to bloom_map_delete_elem as well.

>   	.map_check_btf = check_btf,
>   	.map_btf_name = "bpf_bloom_filter",
>   	.map_btf_id = &bpf_bloom_btf_id,
>