[PATCH RFC v5 00/10] tun: Introduce virtio-net hashing feature

From: Akihiko Odaki
Date: Tue Oct 08 2024 - 03:47:56 EST


virtio-net have two usage of hashes: one is RSS and another is hash
reporting. Conventionally the hash calculation was done by the VMM.
However, computing the hash after the queue was chosen defeats the
purpose of RSS.

Another approach is to use eBPF steering program. This approach has
another downside: it cannot report the calculated hash due to the
restrictive nature of eBPF.

Introduce the code to compute hashes to the kernel in order to overcome
thse challenges.

An alternative solution is to extend the eBPF steering program so that it
will be able to report to the userspace, but it is based on context
rewrites, which is in feature freeze. We can adopt kfuncs, but they will
not be UAPIs. We opt to ioctl to align with other relevant UAPIs (KVM
and vhost_net).

The patches for QEMU to use this new feature was submitted as RFC and
is available at:
https://patchew.org/QEMU/20240915-hash-v3-0-79cb08d28647@xxxxxxxxxx/

This work was presented at LPC 2024:
https://lpc.events/event/18/contributions/1963/

V1 -> V2:
Changed to introduce a new BPF program type.

Signed-off-by: Akihiko Odaki <akihiko.odaki@xxxxxxxxxx>
---
Changes in v5:
- Fixed a compilation error with CONFIG_TUN_VNET_CROSS_LE.
- Optimized the calculation of the hash value according to:
https://git.dpdk.org/dpdk/commit/?id=3fb1ea032bd6ff8317af5dac9af901f1f324cab4
- Added patch "tun: Unify vnet implementation".
- Dropped patch "tap: Pad virtio header with zero".
- Added patch "selftest: tun: Test vnet ioctls without device".
- Reworked selftests to skip for older kernels.
- Documented the case when the underlying device is deleted and packets
have queue_mapping set by TC.
- Reordered test harness arguments.
- Added code to handle fragmented packets.
- Link to v4: https://lore.kernel.org/r/20240924-rss-v4-0-84e932ec0e6c@xxxxxxxxxx

Changes in v4:
- Moved tun_vnet_hash_ext to if_tun.h.
- Renamed virtio_net_toeplitz() to virtio_net_toeplitz_calc().
- Replaced htons() with cpu_to_be16().
- Changed virtio_net_hash_rss() to return void.
- Reordered variable declarations in virtio_net_hash_rss().
- Removed virtio_net_hdr_v1_hash_from_skb().
- Updated messages of "tap: Pad virtio header with zero" and
"tun: Pad virtio header with zero".
- Fixed vnet_hash allocation size.
- Ensured to free vnet_hash when destructing tun_struct.
- Link to v3: https://lore.kernel.org/r/20240915-rss-v3-0-c630015db082@xxxxxxxxxx

Changes in v3:
- Reverted back to add ioctl.
- Split patch "tun: Introduce virtio-net hashing feature" into
"tun: Introduce virtio-net hash reporting feature" and
"tun: Introduce virtio-net RSS".
- Changed to reuse hash values computed for automq instead of performing
RSS hashing when hash reporting is requested but RSS is not.
- Extracted relevant data from struct tun_struct to keep it minimal.
- Added kernel-doc.
- Changed to allow calling TUNGETVNETHASHCAP before TUNSETIFF.
- Initialized num_buffers with 1.
- Added a test case for unclassified packets.
- Fixed error handling in tests.
- Changed tests to verify that the queue index will not overflow.
- Rebased.
- Link to v2: https://lore.kernel.org/r/20231015141644.260646-1-akihiko.odaki@xxxxxxxxxx

---
Akihiko Odaki (10):
virtio_net: Add functions for hashing
skbuff: Introduce SKB_EXT_TUN_VNET_HASH
net: flow_dissector: Export flow_keys_dissector_symmetric
tun: Unify vnet implementation
tun: Pad virtio header with zero
tun: Introduce virtio-net hash reporting feature
tun: Introduce virtio-net RSS
selftest: tun: Test vnet ioctls without device
selftest: tun: Add tests for virtio-net hashing
vhost/net: Support VIRTIO_NET_F_HASH_REPORT

Documentation/networking/tuntap.rst | 7 +
MAINTAINERS | 1 +
drivers/net/Kconfig | 1 +
drivers/net/tap.c | 218 ++++--------
drivers/net/tun.c | 293 ++++++----------
drivers/net/tun_vnet.h | 342 +++++++++++++++++++
drivers/vhost/net.c | 16 +-
include/linux/if_tap.h | 2 +
include/linux/skbuff.h | 3 +
include/linux/virtio_net.h | 188 +++++++++++
include/net/flow_dissector.h | 1 +
include/uapi/linux/if_tun.h | 75 +++++
net/core/flow_dissector.c | 3 +-
net/core/skbuff.c | 4 +
tools/testing/selftests/net/Makefile | 2 +-
tools/testing/selftests/net/tun.c | 630 ++++++++++++++++++++++++++++++++++-
16 files changed, 1430 insertions(+), 356 deletions(-)
---
base-commit: 752ebcbe87aceeb6334e846a466116197711a982
change-id: 20240403-rss-e737d89efa77

Best regards,
--
Akihiko Odaki <akihiko.odaki@xxxxxxxxxx>