Re: [BUG] kernel BUG in team driver: buffer overflow in team_add_slave()

[Jiayuan Chen <jiayuan.chen@linux.dev>]

On 6/8/26 6:00 PM, Yeswanth Krishna wrote:
> Hi Team ,
>
> I encountered a kernel crash while running selftests on kernel 7.1.0-rc6
> on a POWER10 system. The crash occurs when adding a slave device to a
> team interface, triggered by FORTIFY_SOURCE detecting a buffer overflow.
>
> **System Information:**
> - Kernel: 7.1.0-rc6-160099.42-default+ (commit d548c6f4301b)
> - Architecture: powerpc64le (ppc64le)
> - Hardware: IBM POWER10 (9043-MRX), pSeries
> - Config: CONFIG_FORTIFY_SOURCE=y
>
> **Crash Location:**
>
> [ 3492.897824][T77143] kernel BUG at lib/string_helpers.c:1044!
> [ 3492.898057][T77143] NIP [c000000000ac1120] __fortify_panic+0x18/0x28
> [ 3492.898096][T77143] [c00000000efdb350] [c00800000b857a18] 
> team_add_slave+0xc60/0xcc0 [team]
>
>
> **Call Trace:**
>
> __fortify_panic+0x18/0x28
> team_add_slave+0xc60/0xcc0 [team]
> do_set_master+0x19c/0x240
> do_setlink.isra.0+0x388/0x1450
> rtnl_newlink+0xac8/0x1030
> rtnetlink_rcv_msg+0x450/0x530
> netlink_rcv_skb+0x74/0x1b0
> rtnetlink_rcv+0x24/0x40
> netlink_unicast+0x2e0/0x430
> netlink_sendmsg+0x210/0x580
> ____sys_sendmsg+0x30c/0x470
> ___sys_sendmsg+0x94/0xf0
> __sys_sendmsg+0x84/0x100
> system_call_exception+0x154/0x2b0
>
> **Reproducer:**
> The crash is 100% reproducible via selftests:
> ```bash
> cd tools/testing/selftests
> make -C drivers/net/team run_tests


I tried this under x86, but the warning was not triggered with 
CONFIG_FORTIFY_SOURCE.


Hope somebody who has a PPC system can test it.