-----Original Message-----
From: Simon Horman <horms@kernel.org>
Sent: Tuesday, July 15, 2025 9:06 AM
To: Haiyang Zhang <redacted>
Cc: linux-hyperv@vger.kernel.org; netdev@vger.kernel.org; Haiyang Zhang
[off-list ref]; KY Srinivasan [off-list ref];
wei.liu@kernel.org; Dexuan Cui [off-list ref];
andrew+netdev@lunn.ch; edumazet@google.com; kuba@kernel.org;
pabeni@redhat.com; davem@davemloft.net; linux-kernel@vger.kernel.org;
stable@vger.kernel.org; cavery@redhat.com
Subject: [EXTERNAL] Re: [PATCH net,v2] hv_netvsc: Switch VF namespace in
netvsc_open instead

On Mon, Jul 14, 2025 at 09:41:37AM -0700, Haiyang Zhang wrote:

quoted

From: Haiyang Zhang <haiyangz@microsoft.com>

The existing code move the VF NIC to new namespace when NETDEV_REGISTER

is

quoted

received on netvsc NIC. During deletion of the namespace,
default_device_exit_batch() >> default_device_exit_net() is called. When
netvsc NIC is moved back and registered to the default namespace, it
automatically brings VF NIC back to the default namespace. This will

cause

quoted

the default_device_exit_net() >> for_each_netdev_safe loop unable to

detect

quoted

the list end, and hit NULL ptr:

[  231.449420] mana 7870:00:00.0 enP30832s1: Moved VF to namespace with:

eth0

quoted

[  231.449656] BUG: kernel NULL pointer dereference, address:

0000000000000010

quoted

[  231.450246] #PF: supervisor read access in kernel mode
[  231.450579] #PF: error_code(0x0000) - not-present page
[  231.450916] PGD 17b8a8067 P4D 0
[  231.451163] Oops: Oops: 0000 [#1] SMP NOPTI
[  231.451450] CPU: 82 UID: 0 PID: 1394 Comm: kworker/u768:1 Not tainted

6.16.0-rc4+ #3 VOLUNTARY

quoted

[  231.452042] Hardware name: Microsoft Corporation Virtual

Machine/Virtual Machine, BIOS Hyper-V UEFI Release v4.1 11/21/2024

quoted

[  231.452692] Workqueue: netns cleanup_net
[  231.452947] RIP: 0010:default_device_exit_batch+0x16c/0x3f0
[  231.453326] Code: c0 0c f5 b3 e8 d5 db fe ff 48 85 c0 74 15 48 c7 c2

f8 fd ca b2 be 10 00 00 00 48 8d 7d c0 e8 7b 77 25 00 49 8b 86 28 01 00 00
<48> 8b 50 10 4c 8b 2a 4c 8d 62 f0 49 83 ed 10 4c 39 e0 0f 84 d6 00

quoted

[  231.454294] RSP: 0018:ff75fc7c9bf9fd00 EFLAGS: 00010246
[  231.454610] RAX: 0000000000000000 RBX: 0000000000000002 RCX:

61c8864680b583eb

quoted

[  231.455094] RDX: ff1fa9f71462d800 RSI: ff75fc7c9bf9fd38 RDI:

0000000030766564

quoted

[  231.455686] RBP: ff75fc7c9bf9fd78 R08: 0000000000000000 R09:

0000000000000000

quoted

[  231.456126] R10: 0000000000000001 R11: 0000000000000004 R12:

ff1fa9f70088e340

quoted

[  231.456621] R13: ff1fa9f70088e340 R14: ffffffffb3f50c20 R15:

ff1fa9f7103e6340

quoted

[  231.457161] FS:  0000000000000000(0000) GS:ff1faa6783a08000(0000)

knlGS:0000000000000000

quoted

[  231.457707] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  231.458031] CR2: 0000000000000010 CR3: 0000000179ab2006 CR4:

0000000000b73ef0

quoted

[  231.458434] Call Trace:
[  231.458600]  <TASK>
[  231.458777]  ops_undo_list+0x100/0x220
[  231.459015]  cleanup_net+0x1b8/0x300
[  231.459285]  process_one_work+0x184/0x340

To fix it, move the VF namespace switching code from the NETDEV_REGISTER
event handler to netvsc_open().

Cc: stable@vger.kernel.org
Cc: cavery@redhat.com
Fixes: 4c262801ea60 ("hv_netvsc: Fix VF namespace also in synthetic NIC

NETDEV_REGISTER event")

quoted

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>

With this change do we go back to the situation that existed prior
to the cited patch? Quoting the cited commit:

    The existing code moves VF to the same namespace as the synthetic NIC
    during netvsc_register_vf(). But, if the synthetic device is moved to
a
    new namespace after the VF registration, the VF won't be moved
together.

Or perhaps not because if synthetic device is moved then, in practice, it
will subsequently be reopened? (Because it is closed as part of the move
to a different netns?)

There are two cases:
1) the synthetic device is moved to a new namespace before the VF device is 
offered from PCI:
During netvsc_register_vf() >> dev_change_net_namespace() will put VF to 
the same namespace.

2) the synthetic device is moved to a new namespace after the VF device is 
offered from PCI:
The commit 4c262801ea60 does the move in netvsc_event_set_vf_ns >> dev_change_net_namespace().
But it will cause Null ptr error during namespace deletion >> default_device_exit_net().

This patch keeps the code path (1) unchanged, and fix the code path (2).
And yes, __dev_change_net_namespace() >> netif_close(dev), so in the new 
namespace the NIC always needs to be re-opened before using.

Thanks,
- Haiyang