Incorrect netdev->dev_addr on 5.11.2.1 branch
jameshilliard opened this issue · 6 comments
While testing on the 5.11.2.1 branch with kernel 6.0.1 I'm seeing this error in dmesg, although the driver does appear to work regardless.
[ 329.010970] ------------[ cut here ]------------
[ 329.011527] netdevice: wlp0s5u1: Incorrect netdev->dev_addr
[ 329.012189] WARNING: CPU: 2 PID: 488 at net/core/dev_addr_lists.c:519 dev_addr_check.cold+0x43/0x74
[ 329.013300] Modules linked in: 8192eu(O) xt_mark mac80211 libarc4 cfg80211 efivarfs [last unloaded: 8192eu(O)]
[ 329.014532] CPU: 2 PID: 488 Comm: NetworkManager Tainted: G O 6.0.1 #1
[ 329.015465] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015
[ 329.016404] RIP: 0010:dev_addr_check.cold+0x43/0x74
[ 329.017024] Code: 01 e8 79 25 fa ff 0f 0b 48 c7 c5 a2 2d 89 94 80 3b 00 75 27 48 c7 c6 ad 2d 89 94 48 89 ea 48 c7 c7 10 a8 8c 94 e8 55 25 fa ff <0f> 0b e9 82 54 d9 ff 48 8b 2c d5 40 19 55 94 eb d4 be 25 00 00 00
[ 329.019224] RSP: 0018:ffff9e69c03db7b8 EFLAGS: 00010282
[ 329.019886] RAX: 0000000000000000 RBX: ffff8c6750f3f000 RCX: 0000000000000000
[ 329.020764] RDX: 0000000000000001 RSI: ffffffff947e62f1 RDI: 00000000ffffffff
[ 329.021600] RBP: ffffffff947e62f1 R08: 00000000ffffdfff R09: 00000000ffffdfff
[ 329.022450] R10: ffffffff94a59f60 R11: ffffffff94a59f60 R12: ffffffffc05ec400
[ 329.023327] R13: ffff8c6750f3f250 R14: 0000000000000000 R15: 0000000000000001
[ 329.024212] FS: 00007f86d8a943c0(0000) GS:ffff8c68b7d00000(0000) knlGS:0000000000000000
[ 329.025194] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 329.026000] CR2: 00007f2ea1408bd0 CR3: 0000000105b30000 CR4: 00000000003506e0
[ 329.026860] Call Trace:
[ 329.027174] <TASK>
[ 329.027441] __dev_open+0x3a/0x1a0
[ 329.027868] __dev_change_flags+0x1a1/0x210
[ 329.028357] dev_change_flags+0x1d/0x60
[ 329.028840] do_setlink+0x7fd/0x12b0
[ 329.029285] ? __nla_validate_parse+0x5f/0xc40
[ 329.029853] __rtnl_newlink+0x618/0x9b0
[ 329.030326] rtnl_newlink+0x3e/0x60
[ 329.030747] rtnetlink_rcv_msg+0x149/0x3b0
[ 329.031235] ? rtnl_calcit.isra.0+0x130/0x130
[ 329.031784] netlink_rcv_skb+0x4c/0xf0
[ 329.032259] netlink_unicast+0x23d/0x390
[ 329.032764] netlink_sendmsg+0x23b/0x4b0
[ 329.033242] sock_sendmsg+0x5a/0x60
[ 329.033678] ____sys_sendmsg+0x227/0x260
[ 329.034168] ? copy_msghdr_from_user+0x78/0xb0
[ 329.034749] ___sys_sendmsg+0x95/0xe0
[ 329.035224] __sys_sendmsg+0x71/0xc0
[ 329.035685] do_syscall_64+0x3a/0x90
[ 329.036152] entry_SYSCALL_64_after_hwframe+0x63/0xcd
[ 329.036809] RIP: 0033:0x7f86db53eb32
[ 329.037255] Code: 28 89 54 24 1c 48 89 74 24 10 89 7c 24 08 e8 f9 9e f9 ff 8b 54 24 1c 48 8b 74 24 10 41 89 c0 8b 7c 24 08 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 76 10 48 8b 15 b7 d2 0a 00 f7 d8 64 89 02 48 83
[ 329.039445] RSP: 002b:00007fff332741a0 EFLAGS: 00000293 ORIG_RAX: 000000000000002e
[ 329.040368] RAX: ffffffffffffffda RBX: 0000564ea7a2ad00 RCX: 00007f86db53eb32
[ 329.041256] RDX: 0000000000000000 RSI: 00007fff332741f0 RDI: 000000000000000c
[ 329.042129] RBP: 00007fff332741f0 R08: 0000000000000000 R09: 0000000000000000
[ 329.043008] R10: 0000000000000010 R11: 0000000000000293 R12: 0000000000000000
[ 329.043875] R13: 00007fff33274398 R14: 00007fff3327438c R15: 0000000000000000
[ 329.044914] </TASK>
[ 329.045327] ---[ end trace 0000000000000000 ]---Are you on Ubuntu 22.10? I've tested the driver on 22.04 with 6.0.1 and there were no (x86) issues. This seems to be a bug in 22.10 unrelated to the driver.
This looks similar: https://bugs.launchpad.net/ubuntu/+source/broadcom-sta/+bug/1992269
Are you on Ubuntu 22.10?
Nope, I'm using buildroot.
I've tested the driver on 22.04 with 6.0.1 and there were are no (x86) issues.
I'm running a stock 6.0.1 kernel.
This seems to be a bug in 22.10 unrelated to the driver.
It def gets triggered by modprobing the 8192eu driver.
This looks similar: https://bugs.launchpad.net/ubuntu/+source/broadcom-sta/+bug/1992269
It does indeed, however since that's reported against a different wifi driver(one for broadcom hardware) which I'm not using at all here I suspect that something got changed in newer kernel versions which may have triggered an issue in both drivers.
Do you have the same trace with an older branch?
Do you have the same trace with an older branch?
Haven't tried but they probably have the same bug, should be fixed by #59 which probably should be backported to older branches.
Can you rename to "Improve support for kernel 5.15(+)", this makes it easier to find later on.
Also, did you verify that this works?
Can you rename to "Improve support for kernel 5.15(+)", this makes it easier to find later on.
Updated PR to that.
Also, did you verify that this works?
Yeah, from what I can tell it fully fixes the issue.