| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <2063d59.8f9f4a3a.1892a3ec50f@tnonline.net>
Date: Thu, 6 Jul 2023 10:08:24 +0200 (GMT+02:00)
From: Forza <forza@...nline.net>
To: Uladzislau Rezki <urezki@...il.com>,
Bagas Sanjaya <bagasdotme@...il.com>
Cc: Uladzislau Rezki <urezki@...il.com>,
Bagas Sanjaya <bagasdotme@...il.com>,
Linux btrfs <linux-btrfs@...r.kernel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
Linux Regressions <regressions@...ts.linux.dev>,
Linux Stable <stable@...r.kernel.org>,
Chris Mason <clm@...com>, Josef Bacik <josef@...icpanda.com>,
David Sterba <dsterba@...e.com>, a1bert@...as.cz
Subject: Re: Fwd: vmalloc error: btrfs-delalloc btrfs_work_helper [btrfs] in
kernel 6.3.x
---- From: Forza <forza@...nline.net> -- Sent: 2023-07-03 - 01:28 ----
>
>
> ---- From: Uladzislau Rezki <urezki@...il.com> -- Sent: 2023-05-26 - 14:24 ----
>
>> On Wed, May 24, 2023 at 11:13:57AM +0200, David Sterba wrote:
>>> This looks like a different set of problems, though all of them seem to
>>> start on the compression write path in btrfs.
>>>
>>> On Wed, May 24, 2023 at 07:57:19AM +0200, Forza wrote:
>>> > [ 8.641506] 8021q: adding VLAN 0 to HW filter on device enp4s0
>>> > [ 13.841691] wireguard: WireGuard 1.0.0 loaded. See www.wireguard.com for information.
>>> > [ 13.841705] wireguard: Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@...c4.com>. All Rights Reserved.
>>> > [13917.280527] ------------[ cut here ]------------
>>> > [13917.280753] default_enter_idle leaked IRQ state
>>> > [13917.281004] WARNING: CPU: 3 PID: 0 at drivers/cpuidle/cpuidle.c:269 cpuidle_enter_state+0x3bb/0x430
>>>
>>>
> ... Snip
>>>
>>> And again, so something is going wrong
>>>
>> Indeed.
>>
>> I suggest you run your kernel with CONFIG_KASAN=y to see if there are
>> any use-after-free or out-of-bounds bugs.
>>
>> --
>> Uladzislau Rezki
>
>
> Pardon the delay... I have enabled KASAN and UBSAN on this kernel. It produced a lot of output and plenty of warnings for misalignment.
>
> The full dmesg is at https://paste.tnonline.net/files/aBoUMuTd5KBC_dmesg.ubsan.txt (approx 1.7MiB)
>
> The full kernel .conf is. at https://paste.tnonline.net/files/z1mX8TWFgZQ3_kernel.conf-kasan-ubsan.txt
>
> A small exctract around what I think is the default_enter_idle leaked IRQ event. Is this helpful?
>
> ================================================================================
> Jul 03 00:33:57 git kernel: UBSAN: misaligned-access in net/ipv4/tcp_ipv4.c:1848:13
> Jul 03 00:33:57 git kernel: member access within misaligned address 000000007604d82f for type 'const struct tcphdr'
> Jul 03 00:33:57 git kernel: which requires 4 byte alignment
> Jul 03 00:33:57 git kernel: CPU: 2 PID: 29 Comm: ksoftirqd/2 Not tainted 6.3.10-ksan-ubsan #8
> Jul 03 00:33:57 git kernel: Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.0-20220807_005459-localhost 04/01/2014
> Jul 03 00:33:57 git kernel: Call Trace:
> Jul 03 00:33:57 git kernel: <TASK>
> Jul 03 00:33:57 git kernel: dump_stack_lvl+0x86/0xd0
> Jul 03 00:33:57 git kernel: ubsan_type_mismatch_common+0xdf/0x240
> Jul 03 00:33:57 git kernel: __ubsan_handle_type_mismatch_v1+0x44/0x60
> Jul 03 00:33:57 git kernel: tcp_add_backlog+0x1fac/0x3ab0
> Jul 03 00:33:57 git kernel: ? sk_filter_trim_cap+0xcc/0xb60
> Jul 03 00:33:57 git kernel: ? __pfx_tcp_add_backlog+0x10/0x10
> Jul 03 00:33:57 git kernel: ? __pfx__raw_spin_lock+0x10/0x10
> Jul 03 00:33:57 git kernel: tcp_v4_rcv+0x3583/0x4c40
> Jul 03 00:33:57 git kernel: ? __pfx_tcp_v4_rcv+0x10/0x10
> Jul 03 00:33:57 git kernel: ? __pfx__raw_spin_lock_irqsave+0x10/0x10
> Jul 03 00:33:57 git kernel: ip_protocol_deliver_rcu+0x6c/0x480
> Jul 03 00:33:57 git kernel: ip_local_deliver_finish+0x2ae/0x4d0
> Jul 03 00:33:57 git kernel: ? __pfx_ip_local_deliver+0x10/0x10
> Jul 03 00:33:57 git kernel: ? __pfx_ip_local_deliver+0x10/0x10
> Jul 03 00:33:57 git kernel: ip_local_deliver+0x1ba/0x380
> Jul 03 00:33:57 git kernel: ? __pfx_ip_local_deliver+0x10/0x10
> Jul 03 00:33:57 git kernel: ? ipv4_dst_check+0x104/0x250
> Jul 03 00:33:57 git kernel: ? __ubsan_handle_type_mismatch_v1+0x44/0x60
> Jul 03 00:33:57 git kernel: ip_sublist_rcv_finish+0x172/0x380
> Jul 03 00:33:57 git kernel: ------------[ cut here ]------------
> Jul 03 00:33:57 git kernel: ip_sublist_rcv+0x3cd/0x900
> Jul 03 00:33:57 git kernel: default_enter_idle leaked IRQ state
> Jul 03 00:33:57 git kernel: ? __pfx_ip_sublist_rcv+0x10/0x10
> Jul 03 00:33:57 git kernel: ? __ubsan_handle_type_mismatch_v1+0x44/0x60
> Jul 03 00:33:57 git kernel: ? ip_rcv_core+0x972/0x1b20
> Jul 03 00:33:57 git kernel: ip_list_rcv+0x318/0x750
> Jul 03 00:33:57 git kernel: ? __pfx_ip_list_rcv+0x10/0x10
> Jul 03 00:33:57 git kernel: ? __pfx_ip_list_rcv+0x10/0x10
> Jul 03 00:33:57 git kernel: __netif_receive_skb_list_core+0x5ad/0x1170
> Jul 03 00:33:57 git kernel: ? tcp_gro_receive+0x1f45/0x2990
> Jul 03 00:33:57 git kernel: ? __pfx___netif_receive_skb_list_core+0x10/0x10
> Jul 03 00:33:57 git kernel: ? kvm_clock_read+0x16/0x40
> Jul 03 00:33:57 git kernel: ? ktime_get_with_offset+0xd0/0x1f0
> Jul 03 00:33:57 git kernel: netif_receive_skb_list_internal+0x76f/0x1530
> Jul 03 00:33:57 git kernel: ? __pfx_netif_receive_skb_list_internal+0x10/0x10
> Jul 03 00:33:57 git kernel: ? dev_gro_receive+0x67f/0x4900
> Jul 03 00:33:57 git kernel: ? free_unref_page+0x2fd/0x680
> Jul 03 00:33:57 git kernel: ? put_page+0x69/0x2b0
> Jul 03 00:33:57 git kernel: ? __pfx_eth_type_trans+0x10/0x10
> Jul 03 00:33:57 git kernel: napi_gro_receive+0x77b/0xdc0
> Jul 03 00:33:57 git kernel: receive_buf+0x1001/0xac40
> Jul 03 00:33:57 git kernel: ? _raw_spin_lock_irqsave+0xaa/0x180
> Jul 03 00:33:57 git kernel: ? __pfx__raw_spin_lock_irqsave+0x10/0x10
> Jul 03 00:33:57 git kernel: ? __pfx_receive_buf+0x10/0x10
> Jul 03 00:33:57 git kernel: ? _raw_spin_unlock_irqrestore+0x40/0x80
> Jul 03 00:33:57 git kernel: ? trace_hardirqs_on+0x2d/0xd0
> Jul 03 00:33:57 git kernel: ? detach_buf_split+0x27e/0xa70
> Jul 03 00:33:57 git kernel: ? virtqueue_get_buf_ctx_split+0x3c3/0x1400
> Jul 03 00:33:57 git kernel: ? virtqueue_enable_cb_delayed+0x5d0/0x1180
> Jul 03 00:33:57 git kernel: virtnet_poll+0x7c7/0x2030
> Jul 03 00:33:57 git kernel: ? __pfx_virtnet_poll+0x10/0x10
> Jul 03 00:33:57 git kernel: ? __pfx__raw_spin_lock+0x10/0x10
> Jul 03 00:33:57 git kernel: ? __run_timers+0x43d/0xf70
> Jul 03 00:33:57 git kernel: __napi_poll.constprop.0+0xd4/0x840
> Jul 03 00:33:57 git kernel: net_rx_action+0x7a0/0x26e0
> Jul 03 00:33:57 git kernel: ? __pfx_net_rx_action+0x10/0x10
> Jul 03 00:33:57 git kernel: __do_softirq+0x277/0x95d
> Jul 03 00:33:57 git kernel: ? __pfx___do_softirq+0x10/0x10
> Jul 03 00:33:57 git kernel: ? __pfx_run_ksoftirqd+0x10/0x10
> Jul 03 00:33:57 git kernel: ? __pfx_run_ksoftirqd+0x10/0x10
> Jul 03 00:33:57 git kernel: run_ksoftirqd+0x2c/0x40
> Jul 03 00:33:57 git kernel: smpboot_thread_fn+0x380/0xbc0
> Jul 03 00:33:57 git kernel: ? __kthread_parkme+0xdc/0x280
> Jul 03 00:33:57 git kernel: ? schedule+0x158/0x360
> Jul 03 00:33:57 git kernel: ? __pfx_smpboot_thread_fn+0x10/0x10
> Jul 03 00:33:57 git kernel: kthread+0x259/0x3d0
> Jul 03 00:33:57 git kernel: ? __pfx_kthread+0x10/0x10
> Jul 03 00:33:57 git kernel: ret_from_fork+0x2c/0x50
> Jul 03 00:33:57 git kernel: </TASK>
> Jul 03 00:33:57 git kernel: ================================================================================
>
A small update.
I have been able test 6.2.16, all 6.3.x and 6.4.1 and they all show the same issue.
I am now trying 6.1.37 since two days and have not been able to reproduce this issue on any of my virtual qemu/kvm machines. Perhaps this information is helpful in finding the root cause?
~Forza
Powered by blists - more mailing lists