[Bug 285837] rtwn with RTL8821AU and h/w checksum offload seems to treat 0 for IPv4+UDP as invalid
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 285837] rtwn with RTL8821AU and h/w checksum offload seems to treat 0 for IPv4+UDP as invalid"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 285837] rtwn with RTL8821AU and h/w checksum offload seems to treat 0 for IPv4+UDP as invalid"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 285837] rtwn with RTL8821AU and h/w checksum offload seems to treat 0 for IPv4+UDP as invalid"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Tue, 01 Apr 2025 20:27:25 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=285837 Bug ID: 285837 Summary: rtwn with RTL8821AU and h/w checksum offload seems to treat 0 for IPv4+UDP as invalid Product: Base System Version: 15.0-CURRENT Hardware: Any OS: Any Status: New Severity: Affects Only Me Priority: --- Component: wireless Assignee: wireless@FreeBSD.org Reporter: jrtc27@freebsd.org When talking to a DHCP server that leaves the UDP checksum as 0, I see r12a_check_frame_checksum report R12A_RXDW1_CKSUM and R12A_RXDW1_CKSUM_ERR both being set. When talking to a DHCP server that sets the UDP checksum to non-zero I see R12A_RXDW1_CKSUM being set but not R12A_RXDW1_CKSUM_ERR. Further investigation is surely needed, but it appears as if RTL8821AU does not correctly handle the optional nature of IPv4+UDP's checksum (and, as pure speculation, may also not handle all-ones correctly as the other related edge case?). Neither Linux nor OpenBSD appears to use the checksum offloading in their equivalent drivers, only letting the NIC do FCS checking. -- You are receiving this mail because: You are the assignee for the bug.