My wireless access point suddenly started logging tons of messages like these:
Jan 23 19:57:28 apu2 kernel: ath_hal_computetxtime: MCS rate? (index 12; hwrate 0x80)
Jan 23 19:58:28 apu2 syslogd: last message repeated 1 times
Jan 23 20:00:19 apu2 syslogd: last message repeated 22 times
Jan 23 20:01:19 apu2 kernel: ath_hal_computetxtime: MCS rate? (index 13; hwrate 0x81)
Jan 23 20:01:24 apu2 syslogd: last message repeated 1 times
Jan 23 20:01:29 apu2 kernel: ath_hal_computetxtime: MCS rate? (index 12; hwrate 0x80)
Jan 23 20:01:54 apu2 syslogd: last message repeated 5 times
Jan 23 20:02:59 apu2 syslogd: last message repeated 1 times

It's been going on for many hours and then suddenly stopped.
Not sure what happened but suspect that maybe a specific station appeared and then disappeared.
There is comment in the code indicating that this must be a driver bug:
    /* Warn if this function is called for 11n rates; it should not be! */

I captured a stack trace of a call:
CPU     ID                    FUNCTION:NAME
  0  63516             ath_hal_printf:entry
              kernel`ath_hal_computetxtime+0x46
              kernel`ath_tx_calc_duration+0x97
              kernel`ath_tx_xmit_normal+0x78
              kernel`ath_tx_swq+0x267
              kernel`ath_tx_start+0x210
              kernel`ath_transmit+0x1de
              kernel`ieee80211_parent_xmitpkt+0x36
              kernel`ieee80211_vap_pkt_send_dest+0x203
              kernel`ieee80211_start_pkt+0x249
              kernel`ieee80211_vap_transmit+0x7a
              kernel`bridge_enqueue+0xb4
              kernel`bridge_transmit+0x8c
              kernel`bridge_altq_transmit+0x25
              kernel`ether_output_frame+0x73
              kernel`ether_output+0x4a0
              kernel`ip_output_send+0x10
              kernel`ip_output+0x1029
              kernel`ip_forward+0x2c0
              kernel`ip_input+0x66b
              kernel`netisr_dispatch_src+0x9c

Not a huge issue but I didn't like that it spammed logs so much.
So, I'd like to get this fixed.
Does anyone have an idea where the problem might be?

Thanks!

--
Andriy Gapon

Reply via email to