Re: [PATCH net-next v5 12/12] net: dsa: add driver for MaxLinear GSW1xx switch family

["Sverdlin, Alexander" <alexander.sverdlin@siemens.com>]

Hi Daniel,

On Fri, 2025-10-31 at 21:46 +0000, Daniel Golle wrote:
> On Thu, Oct 30, 2025 at 11:11:38PM +0000, Sverdlin, Alexander wrote:
> > [...]
> > For some reason with both v4 and v5 I can reliably reproduce the following
> > warning (ASSERT_RTNL()) at the very beginning of
> > drivers/net/dsa/local_termination.sh selftest:
> > 
> > RTNL: assertion failed at git/net/core/dev.c (9480)
> > WARNING: CPU: 1 PID: 529 at git/net/core/dev.c:9480 __dev_set_promiscuity+0x174/0x188
> > CPU: 1 UID: 996 PID: 529 Comm: systemd-resolve Tainted: G           O        6.18.0-rc2+gite9079300094d #1 PREEMPT 
> > pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> > pc : __dev_set_promiscuity+0x174/0x188
> > lr : __dev_set_promiscuity+0x174/0x188
> > Call trace:
> >   __dev_set_promiscuity+0x174/0x188 (P)
> >   __dev_set_rx_mode+0xa0/0xb0
> >   dev_mc_del+0x94/0xc0
> >   igmp6_group_dropped+0x124/0x410
> >   __ipv6_dev_mc_dec+0x108/0x168
> >   __ipv6_sock_mc_drop+0x64/0x188
> >   ipv6_sock_mc_drop+0x140/0x170
> >   do_ipv6_setsockopt+0x1408/0x1828
> >   ipv6_setsockopt+0x64/0xf8
> >   udpv6_setsockopt+0x28/0x58
> >   sock_common_setsockopt+0x24/0x38
> >   do_sock_setsockopt+0x78/0x158
> >   __sys_setsockopt+0x88/0x110
> >   __arm64_sys_setsockopt+0x30/0x48
> >   invoke_syscall+0x50/0x120
> >   el0_svc_common.constprop.0+0xc8/0xf0
> >   do_el0_svc+0x24/0x38
> >   el0_svc+0x50/0x2b0
> >   el0t_64_sync_handler+0xa0/0xe8
> >   el0t_64_sync+0x198/0x1a0
> > 
> > (testing with GSW145)
> > I'm not sure though, if it's related to the gsw1xx code, am65-cpsw-nuss driver
> > on my CPU port or if it's a fresh regression in net-next...
> > 
> > I can see the above splat if I apply the patchset onto bfe62db5422b1a5f25752bd0877a097d436d876d
> > but not with older patchset on top of e90576829ce47a46838685153494bc12cd1bc333.
> > 
> > I'll try to bisect the underlying net-next...
> 
> Did you try to rebase the patches necessary for the GSW145 on top of the
> last known-to-work net-next commit?
> 
> Also note that I've successfully tested local_termination.sh on top
> of ea7d0d60ebc9bddf3ad768557dfa1495bc032bf6, but using the modified
> RaspberryPi 4B provided by MaxLinear, so there obviously is a different
> Ethernet driver on the CPU port...

I'll need to follow up on this... It turned out to be not that reliable
reproducible, so my bisect led me to nowhere.

I don't think this is related to GSW145 code (even though it's being triggered
by GSW145 port going down), only core code seems to be involved here...
And I see quite some RTNL have been dropped in the net-next...

-- 
Alexander Sverdlin
Siemens AG
www.siemens.com