| Message ID | 20240622-linux-udpgso-v1-0-d2344157ab2a@cloudflare.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Lift UDP_SEGMENT restriction for egress via device w/o csum offload | expand |
Jakub Sitnicki wrote: > This is a follow-up to an earlier question [1] if we can make UDP GSO work with > any egress device, even those with no checksum offload capability. That's the > default setup for TUN/TAP. > > I leave it to the maintainers to decide if it qualifies as a fix. We plan to > backport it to our v6.6 either way, hence the submission to -net. > > [1] https://lore.kernel.org/netdev/87jzqsld6q.fsf@cloudflare.com/ Agreed with the change to allow UDP_SEGMENT to work regardless of device capabilities. In my opinion this is a new feature with sufficient risk of unintended side effects to be net-next material. Maybe worth recording in patch 1 the reason for the original check: that UDP_SEGMENT with software checksumming in the GSO stack may be a regression vs copy_and_checksum in the send syscall. > > Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com> > --- > Jakub Sitnicki (2): > udp: Allow GSO transmit from devices with no checksum offload > selftests/net: Add test coverage for UDP GSO software fallback > > net/ipv4/udp.c | 3 +-- > net/ipv4/udp_offload.c | 8 +++++++ > net/ipv6/udp.c | 3 +-- > tools/testing/selftests/net/udpgso.c | 15 +++++++++--- > tools/testing/selftests/net/udpgso.sh | 43 +++++++++++++++++++++++++++++++++++ > 5 files changed, 65 insertions(+), 7 deletions(-) >
On Sun, Jun 23, 2024 at 04:18 AM -04, Willem de Bruijn wrote: > Jakub Sitnicki wrote: >> This is a follow-up to an earlier question [1] if we can make UDP GSO work with >> any egress device, even those with no checksum offload capability. That's the >> default setup for TUN/TAP. >> >> I leave it to the maintainers to decide if it qualifies as a fix. We plan to >> backport it to our v6.6 either way, hence the submission to -net. >> >> [1] https://lore.kernel.org/netdev/87jzqsld6q.fsf@cloudflare.com/ > > Agreed with the change to allow UDP_SEGMENT to work regardless of > device capabilities. Thanks for the stamp. > In my opinion this is a new feature with sufficient risk of unintended > side effects to be net-next material. OK, I will resubmit for net-next. > Maybe worth recording in patch 1 the reason for the original check: > that UDP_SEGMENT with software checksumming in the GSO stack may be > a regression vs copy_and_checksum in the send syscall. Will expand the description.
This is a follow-up to an earlier question [1] if we can make UDP GSO work with any egress device, even those with no checksum offload capability. That's the default setup for TUN/TAP. I leave it to the maintainers to decide if it qualifies as a fix. We plan to backport it to our v6.6 either way, hence the submission to -net. [1] https://lore.kernel.org/netdev/87jzqsld6q.fsf@cloudflare.com/ Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com> --- Jakub Sitnicki (2): udp: Allow GSO transmit from devices with no checksum offload selftests/net: Add test coverage for UDP GSO software fallback net/ipv4/udp.c | 3 +-- net/ipv4/udp_offload.c | 8 +++++++ net/ipv6/udp.c | 3 +-- tools/testing/selftests/net/udpgso.c | 15 +++++++++--- tools/testing/selftests/net/udpgso.sh | 43 +++++++++++++++++++++++++++++++++++ 5 files changed, 65 insertions(+), 7 deletions(-)