[Intel-wired-lan] What are management packets (rx_smbus, MGTPRC)?
Paul Menzel
pmenzel at molgen.mpg.de
Thu Oct 8 15:57:49 UTC 2020
Dear Linux folks,
We are trying to debug a network issue, and, while looking at the
network device statistics, noticed a high value of `rx_smbus`.
$ sudo ethtool -S net00
[…]
tx_smbus: 0
rx_smbus: 557816
dropped_smbus: 9
[…]
This seems unrelated to our problem, but we noticed, that around 20 % of
our desktops have a positive value. From the source:
drivers/net/ethernet/intel/e1000/e1000_main.c:
adapter->stats.mgprc += er32(MGTPRC);
drivers/net/ethernet/intel/e1000e/ethtool.c:
E1000_STAT("rx_smbus", stats.mgprc
drivers/net/ethernet/intel/e1000e/regs.h:#define E1000_MGTPRC
0x040B4 /* Management Packets Rx Count - R/clr *
Out of curiosity, what are these management packets?
Kind regards,
Paul
$ sudo ethtool -S net00
NIC statistics:
rx_packets: 993001516
tx_packets: 1331956350
rx_bytes: 938128015910
tx_bytes: 1842492467154
rx_broadcast: 84929255
tx_broadcast: 237026
rx_multicast: 3869371
tx_multicast: 47602
rx_errors: 250
tx_errors: 0
tx_dropped: 0
multicast: 3869371
collisions: 0
rx_length_errors: 0
rx_over_errors: 0
rx_crc_errors: 187
rx_frame_errors: 0
rx_no_buffer_count: 0
rx_missed_errors: 0
tx_aborted_errors: 0
tx_carrier_errors: 0
tx_fifo_errors: 0
tx_heartbeat_errors: 0
tx_window_errors: 0
tx_abort_late_coll: 0
tx_deferred_ok: 0
tx_single_coll_ok: 0
tx_multi_coll_ok: 0
tx_timeout_count: 0
tx_restart_queue: 0
rx_long_length_errors: 0
rx_short_length_errors: 0
rx_align_errors: 0
tx_tcp_seg_good: 31498083
tx_tcp_seg_failed: 0
rx_flow_control_xon: 0
rx_flow_control_xoff: 0
tx_flow_control_xon: 0
tx_flow_control_xoff: 0
rx_csum_offload_good: 927463008
rx_csum_offload_errors: 34
rx_header_split: 0
alloc_rx_buff_failed: 0
tx_smbus: 0
rx_smbus: 557816
dropped_smbus: 9
rx_dma_failed: 0
tx_dma_failed: 0
rx_hwtstamp_cleared: 0
uncorr_ecc_errors: 0
corr_ecc_errors: 0
tx_hwtstamp_timeouts: 0
tx_hwtstamp_skipped: 0
More information about the Intel-wired-lan
mailing list