mbox series

[net,v2,0/4] bonding: properly restore flags when bond changes ether type

Message ID 20230314111426.1254998-1-razor@blackwall.org (mailing list archive)
Headers show
Series bonding: properly restore flags when bond changes ether type | expand

Message

Nikolay Aleksandrov March 14, 2023, 11:14 a.m. UTC 
Hi,
A bug was reported by syzbot[1] that causes a warning and a myriad of
other potential issues if a bond, that is also a slave, fails to enslave a
non-eth device. While fixing that bug I found that we have the same
issues when such enslave passes and after that the bond changes back to
ARPHRD_ETHER (again due to ether_setup). This set fixes all issues by
extracting the ether_setup() sequence in a helper which does the right
thing about bond flags when it needs to change back to ARPHRD_ETHER. It
also adds selftests for these cases.

Patch 01 adds the new bond_ether_setup helper that is used in the
following patches to fix the bond dev flag issues. Patch 02 fixes the
issues when a bond device changes its ether type due to successful
enslave. Patch 03 fixes the issues when it changes its ether type due to
an unsuccessful enslave. Note we need two patches because the bugs were
introduced by different commits. Patch 04 adds the new selftests.

v2: new set, all patches are new due to new approach of fixing these bugs

Thanks,
 Nik

[1] https://syzkaller.appspot.com/bug?id=391c7b1f6522182899efba27d891f1743e8eb3ef

Nikolay Aleksandrov (4):
  bonding: add bond_ether_setup helper
  bonding: restore IFF_MASTER/SLAVE flags on bond enslave ether type
    change
  bonding: restore bond's IFF_SLAVE flag if a non-eth dev enslave fails
  selftests: bonding: add tests for ether type changes

 drivers/net/bonding/bond_main.c               | 22 +++--
 .../selftests/drivers/net/bonding/Makefile    |  3 +-
 .../net/bonding/bond-eth-type-change.sh       | 85 +++++++++++++++++++
 3 files changed, 102 insertions(+), 8 deletions(-)
 create mode 100755 tools/testing/selftests/drivers/net/bonding/bond-eth-type-change.sh

Comments

Jonathan Toppins March 14, 2023, 3:15 p.m. UTC | #1 
On 3/14/23 07:14, Nikolay Aleksandrov wrote:
> Hi,
> A bug was reported by syzbot[1] that causes a warning and a myriad of
> other potential issues if a bond, that is also a slave, fails to enslave a
> non-eth device. While fixing that bug I found that we have the same
> issues when such enslave passes and after that the bond changes back to
> ARPHRD_ETHER (again due to ether_setup). This set fixes all issues by
> extracting the ether_setup() sequence in a helper which does the right
> thing about bond flags when it needs to change back to ARPHRD_ETHER. It
> also adds selftests for these cases.
> 
> Patch 01 adds the new bond_ether_setup helper that is used in the
> following patches to fix the bond dev flag issues. Patch 02 fixes the
> issues when a bond device changes its ether type due to successful
> enslave. Patch 03 fixes the issues when it changes its ether type due to
> an unsuccessful enslave. Note we need two patches because the bugs were
> introduced by different commits. Patch 04 adds the new selftests.
> 
> v2: new set, all patches are new due to new approach of fixing these bugs
> 
> Thanks,
>   Nik
> 
> [1] https://syzkaller.appspot.com/bug?id=391c7b1f6522182899efba27d891f1743e8eb3ef
> 
> Nikolay Aleksandrov (4):
>    bonding: add bond_ether_setup helper
>    bonding: restore IFF_MASTER/SLAVE flags on bond enslave ether type
>      change
>    bonding: restore bond's IFF_SLAVE flag if a non-eth dev enslave fails
>    selftests: bonding: add tests for ether type changes
> 
>   drivers/net/bonding/bond_main.c               | 22 +++--
>   .../selftests/drivers/net/bonding/Makefile    |  3 +-
>   .../net/bonding/bond-eth-type-change.sh       | 85 +++++++++++++++++++
>   3 files changed, 102 insertions(+), 8 deletions(-)
>   create mode 100755 tools/testing/selftests/drivers/net/bonding/bond-eth-type-change.sh
> 

For the series.
Acked-by: Jonathan Toppins <jtoppins@redhat.com>
Jay Vosburgh March 14, 2023, 4:20 p.m. UTC | #2 
Nikolay Aleksandrov <razor@blackwall.org> wrote:

>Hi,
>A bug was reported by syzbot[1] that causes a warning and a myriad of
>other potential issues if a bond, that is also a slave, fails to enslave a
>non-eth device. While fixing that bug I found that we have the same
>issues when such enslave passes and after that the bond changes back to
>ARPHRD_ETHER (again due to ether_setup). This set fixes all issues by
>extracting the ether_setup() sequence in a helper which does the right
>thing about bond flags when it needs to change back to ARPHRD_ETHER. It
>also adds selftests for these cases.
>
>Patch 01 adds the new bond_ether_setup helper that is used in the
>following patches to fix the bond dev flag issues. Patch 02 fixes the
>issues when a bond device changes its ether type due to successful
>enslave. Patch 03 fixes the issues when it changes its ether type due to
>an unsuccessful enslave. Note we need two patches because the bugs were
>introduced by different commits. Patch 04 adds the new selftests.
>
>v2: new set, all patches are new due to new approach of fixing these bugs

	For the series:

Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>


>Thanks,
> Nik
>
>[1] https://syzkaller.appspot.com/bug?id=391c7b1f6522182899efba27d891f1743e8eb3ef
>
>Nikolay Aleksandrov (4):
>  bonding: add bond_ether_setup helper
>  bonding: restore IFF_MASTER/SLAVE flags on bond enslave ether type
>    change
>  bonding: restore bond's IFF_SLAVE flag if a non-eth dev enslave fails
>  selftests: bonding: add tests for ether type changes
>
> drivers/net/bonding/bond_main.c               | 22 +++--
> .../selftests/drivers/net/bonding/Makefile    |  3 +-
> .../net/bonding/bond-eth-type-change.sh       | 85 +++++++++++++++++++
> 3 files changed, 102 insertions(+), 8 deletions(-)
> create mode 100755 tools/testing/selftests/drivers/net/bonding/bond-eth-type-change.sh
>
>-- 
>2.39.2
>