mbox series

[net-next,00/18] nexthop: Add support for nexthop objects offload

Message ID 20201104133040.1125369-1-idosch@idosch.org (mailing list archive)
Headers show
Series nexthop: Add support for nexthop objects offload | expand

Message

Ido Schimmel Nov. 4, 2020, 1:30 p.m. UTC
From: Ido Schimmel <idosch@nvidia.com>

This patch set adds support for nexthop objects offload with a dummy
implementation over netdevsim. mlxsw support will be added later.

The general idea is very similar to route offload in that notifications
are sent whenever nexthop objects are changed. A listener can veto the
change and the error will be communicated to user space with extack.

To keep listeners as simple as possible, they not only receive
notifications for the nexthop object that is changed, but also for all
the other objects affected by this change. For example, when a single
nexthop is replaced, a replace notification is sent for the single
nexthop, but also for all the nexthop groups this nexthop is member in.
This relieves listeners from the need to track such dependencies.

To simplify things further for listeners, the notification info does not
contain the raw nexthop data structures (e.g., 'struct nexthop'), but
less complex data structures into which the raw data structures are
parsed into.

Tested with a new selftest over netdevsim and with fib_nexthops.sh:

Tests passed: 164
Tests failed:   0

Patch set overview:

Patches #1-#4 introduce the aforementioned data structures and convert
existing listeners (i.e., the VXLAN driver) to use them.

Patches #5-#6 add a new RTNH_F_TRAP flag and the ability to set it and
RTNH_F_OFFLOAD on nexthops. This flag is used by netdevsim for testing
purposes and will also be used by mlxsw. These flags are consistent with
the existing RTM_F_OFFLOAD and RTM_F_TRAP flags.

Patches #7-#14 gradually add the new nexthop notifications.

Patches #15-#18 add a dummy implementation for nexthop offload over
netdevsim and a selftest to exercise both good and bad flows.

Changes since RFC [1]:

Patch #1: s/is_encap/has_encap/
Patch #3: Add a blank line in __nh_notifier_single_info_init()
Patch #5: Reword commit message
Patch #6: s/nexthop_hw_flags_set/nexthop_set_hw_flags/
Patch #7: Reword commit message
Patch #11: Allocate extack on the stack

Follow-up patch sets:

selftests: forwarding: Add nexthop objects tests
mlxsw: Preparations for nexthop objects support - part 1/2
mlxsw: Preparations for nexthop objects support - part 2/2
mlxsw: Add support for nexthop objects
mlxsw: Add support for blackhole nexthops
mlxsw: Update adjacency index more efficiently

[1] https://lore.kernel.org/netdev/20200908091037.2709823-1-idosch@idosch.org/

Ido Schimmel (18):
  nexthop: Add nexthop notification data structures
  nexthop: Pass extack to nexthop notifier
  nexthop: Prepare new notification info
  nexthop: vxlan: Convert to new notification info
  rtnetlink: Add RTNH_F_TRAP flag
  nexthop: Allow setting "offload" and "trap" indications on nexthops
  nexthop: Emit a notification when a nexthop is added
  nexthop: Emit a notification when a nexthop group is replaced
  nexthop: Emit a notification when a single nexthop is replaced
  nexthop: Emit a notification when a nexthop group is modified
  nexthop: Emit a notification when a nexthop group is reduced
  nexthop: Pass extack to register_nexthop_notifier()
  nexthop: Replay nexthops when registering a notifier
  nexthop: Remove in-kernel route notifications when nexthop changes
  netdevsim: Add devlink resource for nexthops
  netdevsim: Add dummy implementation for nexthop offload
  netdevsim: Allow programming routes with nexthop objects
  selftests: netdevsim: Add test for nexthop offload API

 .../networking/devlink/netdevsim.rst          |   3 +-
 drivers/net/netdevsim/dev.c                   |   6 +
 drivers/net/netdevsim/fib.c                   | 265 ++++++++++-
 drivers/net/netdevsim/netdevsim.h             |   1 +
 drivers/net/vxlan.c                           |  12 +-
 include/net/nexthop.h                         |  42 +-
 include/uapi/linux/rtnetlink.h                |   6 +-
 net/ipv4/fib_semantics.c                      |   2 +
 net/ipv4/fib_trie.c                           |   9 -
 net/ipv4/nexthop.c                            | 255 +++++++++-
 net/ipv6/route.c                              |   5 -
 .../drivers/net/netdevsim/nexthop.sh          | 436 ++++++++++++++++++
 12 files changed, 995 insertions(+), 47 deletions(-)
 create mode 100755 tools/testing/selftests/drivers/net/netdevsim/nexthop.sh

Comments

Jakub Kicinski Nov. 6, 2020, 7:31 p.m. UTC | #1
On Wed,  4 Nov 2020 15:30:22 +0200 Ido Schimmel wrote:
> From: Ido Schimmel <idosch@nvidia.com>
> 
> This patch set adds support for nexthop objects offload with a dummy
> implementation over netdevsim. mlxsw support will be added later.
> 
> The general idea is very similar to route offload in that notifications
> are sent whenever nexthop objects are changed. A listener can veto the
> change and the error will be communicated to user space with extack.
> 
> To keep listeners as simple as possible, they not only receive
> notifications for the nexthop object that is changed, but also for all
> the other objects affected by this change. For example, when a single
> nexthop is replaced, a replace notification is sent for the single
> nexthop, but also for all the nexthop groups this nexthop is member in.
> This relieves listeners from the need to track such dependencies.
> 
> To simplify things further for listeners, the notification info does not
> contain the raw nexthop data structures (e.g., 'struct nexthop'), but
> less complex data structures into which the raw data structures are
> parsed into.

Applied, thank you!

BTW no need to follow up on my else-after-return comment, 
just something to keep in mind.
Ido Schimmel Nov. 9, 2020, 3:41 p.m. UTC | #2
On Fri, Nov 06, 2020 at 11:31:59AM -0800, Jakub Kicinski wrote:
> On Wed,  4 Nov 2020 15:30:22 +0200 Ido Schimmel wrote:
> > From: Ido Schimmel <idosch@nvidia.com>
> > 
> > This patch set adds support for nexthop objects offload with a dummy
> > implementation over netdevsim. mlxsw support will be added later.
> > 
> > The general idea is very similar to route offload in that notifications
> > are sent whenever nexthop objects are changed. A listener can veto the
> > change and the error will be communicated to user space with extack.
> > 
> > To keep listeners as simple as possible, they not only receive
> > notifications for the nexthop object that is changed, but also for all
> > the other objects affected by this change. For example, when a single
> > nexthop is replaced, a replace notification is sent for the single
> > nexthop, but also for all the nexthop groups this nexthop is member in.
> > This relieves listeners from the need to track such dependencies.
> > 
> > To simplify things further for listeners, the notification info does not
> > contain the raw nexthop data structures (e.g., 'struct nexthop'), but
> > less complex data structures into which the raw data structures are
> > parsed into.
> 
> Applied, thank you!

Great, thank you. And thanks David for the awesome work on the nexthop
infrastructure.

> 
> BTW no need to follow up on my else-after-return comment, 
> just something to keep in mind.

Ack
David Ahern Nov. 10, 2020, 3:48 a.m. UTC | #3
On 11/9/20 8:41 AM, Ido Schimmel wrote:
> Great, thank you. And thanks David for the awesome work on the nexthop
> infrastructure.

thanks. Looking forward to the next round of patches for h/w offload.