I have same problem with ix (X520-DA1) as I had with em/igb (https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=233022) 070.647811 main_thread [2464] 0 pps (0 pkts 0 bps in 1002979 usec) 0.00 avg_batch 99999 min_space 071.000000 sender_body [1647] pending tx tail 316 head 318 on ring 0 071.000020 sender_body [1647] pending tx tail 316 head 318 on ring 0 071.000031 sender_body [1647] pending tx tail 316 head 318 on ring 0 071.000042 sender_body [1647] pending tx tail 316 head 318 on ring 0 071.000052 sender_body [1647] pending tx tail 316 head 318 on ring 0 071.648690 main_thread [2464] 0 pps (0 pkts 0 bps in 1000879 usec) 0.00 avg_batch 99999 min_space 072.000002 sender_body [1647] pending tx tail 316 head 318 on ring 0 072.000020 sender_body [1647] pending tx tail 316 head 318 on ring 0 072.000032 sender_body [1647] pending tx tail 316 head 318 on ring 0 072.000042 sender_body [1647] pending tx tail 316 head 318 on ring 0 072.000052 sender_body [1647] pending tx tail 316 head 318 on ring 0 072.649762 main_thread [2464] 0 pps (0 pkts 0 bps in 1001072 usec) 0.00 avg_batch 99999 min_space 073.000002 sender_body [1647] pending tx tail 316 head 318 on ring 0 073.000014 sender_body [1647] pending tx tail 316 head 318 on ring 0 073.000024 sender_body [1647] pending tx tail 316 head 318 on ring 0 073.000034 sender_body [1647] pending tx tail 316 head 318 on ring 0 073.000044 sender_body [1647] pending tx tail 316 head 318 on ring 0 073.650790 main_thread [2464] 0 pps (0 pkts 0 bps in 1001028 usec) 0.00 avg_batch 99999 min_space
Yeah, looks like the same issue here too. Review here: https://reviews.freebsd.org/D18368 Please test, thanks!
(In reply to Stephen Hurd from comment #1) I could say, that it helps, as I've did this "fix" two hours ago myself :-)
A commit references this bug: Author: shurd Date: Wed Nov 28 17:37:18 UTC 2018 New revision: 341156 URL: https://svnweb.freebsd.org/changeset/base/341156 Log: Fix first-packet completion The first packet after the ring is initialized was never completed as isc_txd_credits_update() would not include it in the count of completed packets. This caused netmap to never complete a batch. See PR 233022 for more details. This is the same fix as the r340310 for e1000 PR: 233607 Reported by: lev Reviewed by: lev MFC after: 3 days Sponsored by: Limelight Networks Differential Revision: https://reviews.freebsd.org/D18368 Changes: head/sys/dev/ixgbe/ix_txrx.c head/sys/dev/ixl/ixl_txrx.c
A commit references this bug: Author: shurd Date: Mon Dec 3 15:18:35 UTC 2018 New revision: 341427 URL: https://svnweb.freebsd.org/changeset/base/341427 Log: MFC r341156: Fix first-packet completion The first packet after the ring is initialized was never completed as isc_txd_credits_update() would not include it in the count of completed packets. This caused netmap to never complete a batch. See PR 233022 for more details. This is the same fix as the r340310 for e1000 PR: 233607 Reported by: lev Reviewed by: lev Sponsored by: Limelight Networks Differential Revision: https://reviews.freebsd.org/D18368 Changes: _U stable/12/ stable/12/sys/dev/ixgbe/ix_txrx.c stable/12/sys/dev/ixl/ixl_txrx.c