On Wed, Nov 2, 2022 at 2:20 AM Wang Yufen <wangyufen@xxxxxxxxxx> wrote:
>
> kmemleak reports after running test_progs:
>
> unreferenced object 0xffff8881b1672dc0 (size 232):
> comm "test_progs", pid 394388, jiffies 4354712116 (age 841.975s)
> hex dump (first 32 bytes):
> e0 84 d7 a8 81 88 ff ff 80 2c 67 b1 81 88 ff ff .........,g.....
> 00 40 c5 9b 81 88 ff ff 00 00 00 00 00 00 00 00 .@..............
> backtrace:
> [<00000000c8f01748>] napi_skb_cache_get+0xd4/0x150
> [<0000000041c7fc09>] __napi_build_skb+0x15/0x50
> [<00000000431c7079>] __napi_alloc_skb+0x26e/0x540
> [<000000003ecfa30e>] napi_get_frags+0x59/0x140
> [<0000000099b2199e>] tun_get_user+0x183d/0x3bb0 [tun]
> [<000000008a5adef0>] tun_chr_write_iter+0xc0/0x1b1 [tun]
> [<0000000049993ff4>] do_iter_readv_writev+0x19f/0x320
> [<000000008f338ea2>] do_iter_write+0x135/0x630
> [<000000008a3377a4>] vfs_writev+0x12e/0x440
> [<00000000a6b5639a>] do_writev+0x104/0x280
> [<00000000ccf065d8>] do_syscall_64+0x3b/0x90
> [<00000000d776e329>] entry_SYSCALL_64_after_hwframe+0x63/0xcd
>
> The issue occurs in the following scenarios:
> tun_get_user()
> napi_gro_frags()
> napi_frags_finish()
> case GRO_NORMAL:
> gro_normal_one()
> list_add_tail(&skb->list, &napi->rx_list);
> <-- While napi->rx_count < READ_ONCE(gro_normal_batch),
> <-- gro_normal_list() is not called, napi->rx_list is not empty
> <-- not ask to complete the gro work, will cause memory leaks in
> <-- following tun_napi_del()
> ...
> tun_napi_del()
> netif_napi_del()
> __netif_napi_del()
> <-- &napi->rx_list is not empty, which caused memory leaks
>
> To fix, add napi_complete() after napi_gro_frags().
>
> Fixes: 90e33d459407 ("tun: enable napi_gro_frags() for TUN/TAP driver")
> Signed-off-by: Wang Yufen <wangyufen@xxxxxxxxxx>
> ---
> drivers/net/tun.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/net/tun.c b/drivers/net/tun.c
> index 27c6d23..07a0a61 100644
> --- a/drivers/net/tun.c
> +++ b/drivers/net/tun.c
> @@ -1976,6 +1976,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
>
> local_bh_disable();
> napi_gro_frags(&tfile->napi);
> + napi_complete(&tfile->napi);
> local_bh_enable();
> mutex_unlock(&tfile->napi_mutex);
> } else if (tfile->napi_enabled) {
> --
> 1.8.3.1
>
I think the intent of this code was to allow user space to send
multiple packets to GRO layer for test purposes.
(Check that GRO can aggregate multiple MSS, with respect of standard GRO rules)
This fix might break some tests, but they can be updated if needed
(using /sys/class/net/xxx/gro_flush_timeout)
Reviewed-by: Eric Dumazet <edumazet@xxxxxxxxxx>
