| 1 | While poweroff with network active (iprowifi4965) i often got KDL.
|
|---|
| 2 | Here is the trace:
|
|---|
| 3 |
|
|---|
| 4 | ====
|
|---|
| 5 | wlan_control: 9234, 16
|
|---|
| 6 | ->iwn_rx_done begin
|
|---|
| 7 | ->Doing iwn5000_get_rssi
|
|---|
| 8 | iwn5000_get_rssi: agc 39 rssi 4885 5400 0 result -59
|
|---|
| 9 | ->Doing iwn_hw_stop
|
|---|
| 10 | ->Doing iwn_reset_rx_ring
|
|---|
| 11 | ->doing iwn_reset_tx_ring
|
|---|
| 12 | Last message repeated 19 times.
|
|---|
| 13 | ->iwn_rx_done: end
|
|---|
| 14 | iwn_notif_intr: cur=1; qid 4 idx 3 flags 0 type 162(BEACON_MISSED) len 20
|
|---|
| 15 | iwn_notif_intr: beacons missed 0/0
|
|---|
| 16 | iwn_notif_intr: cur=2; qid 4 idx 17 flags 0 type 192(RX_PHY) len 68
|
|---|
| 17 | iwn_rx_phy: received PHY stats
|
|---|
| 18 | iwn_notif_intr: cur=3; qid c idx 204 flags cc type 204(IWN_CMD_BT_COEX_PRIOTABLE) len 52428
|
|---|
| 19 | iwn_notif_intr: cur=4; qid 4 idx 19 flags 0 type 157(BEACON_STATS) len 528
|
|---|
| 20 | ->iwn_rx_statistics begin
|
|---|
| 21 | ->iwn_rx_statistics received during calib
|
|---|
| 22 | iwn_notif_intr: cur=5; qid 9 idx 129 flags 0 type 168(IWN_CMD_SET_SENSITIVITY) len 4
|
|---|
| 23 | iwn_notif_intr: cur=6; qid 4 idx 27 flags 0 type 192(RX_PHY) len 68
|
|---|
| 24 | iwn_rx_phy: received PHY stats
|
|---|
| 25 | iwn_notif_intr: cur=7; qid c idx 204 flags cc type 204(IWN_CMD_BT_COEX_PRIOTABLE) len 52428
|
|---|
| 26 | iwn_notif_intr: cur=8; qid 4 idx 29 flags 0 type 157(BEACON_STATS) len 528
|
|---|
| 27 | ->iwn_rx_statistics begin
|
|---|
| 28 | ->iwn_rx_statistics received during calib
|
|---|
| 29 | iwn_notif_intr: cur=9; qid 3 idx 7 flags 0 type 28(TX_DONE) len 44
|
|---|
| 30 | iwn5000_tx_done: qid 3 idx 7 RTS retries 0 ACK retries 0 nkill 0 rate 420a duration 746 status 201
|
|---|
| 31 | PANIC: no node
|
|---|
| 32 | Welcome to Kernel Debugging Land...
|
|---|
| 33 | Thread 759 "iwn intr handler" running on CPU 2
|
|---|
| 34 | stack trace for thread 759 "iwn intr handler"
|
|---|
| 35 | kernel stack: 0xffffffff91046000 to 0xffffffff9104b000
|
|---|
| 36 | frame caller <image>:function + offset
|
|---|
| 37 | 0 ffffffff9104abd8 (+ 16) ffffffff8014132d <kernel_x86_64> arch_debug_stack_trace + 0x13
|
|---|
| 38 | 1 ffffffff9104abe8 (+ 16) ffffffff800a7823 <kernel_x86_64> stack_trace_trampoline(void*) + 0x09
|
|---|
| 39 | 2 ffffffff9104abf8 (+ 24) ffffffff8013902c <kernel_x86_64> arch_debug_call_with_fault_handler + 0x16
|
|---|
| 40 | 3 ffffffff9104ac10 (+ 96) ffffffff800a820f <kernel_x86_64> debug_call_with_fault_handler + 0x8b
|
|---|
| 41 | 4 ffffffff9104ac70 (+ 96) ffffffff800a926b <kernel_x86_64> kernel_debugger_loop(char const*, char const*, __va_list_tag*, int) + 0x10a
|
|---|
| 42 | 5 ffffffff9104acd0 (+ 80) ffffffff800a96d4 <kernel_x86_64> kernel_debugger_internal(char const*, char const*, __va_list_tag*, int) + 0x1a2
|
|---|
| 43 | 6 ffffffff9104ad20 (+ 240) ffffffff800a98d3 <kernel_x86_64> panic + 0xba
|
|---|
| 44 | 7 ffffffff9104ae10 (+ 80) ffffffff80617096 </boot/system/non-packaged/add-ons/kernel/drivers/bin/iprowifi4965> iwn_tx_done + 0x1c6
|
|---|
| 45 | 8 ffffffff9104ae60 (+ 192) ffffffff806257f1 </boot/system/non-packaged/add-ons/kernel/drivers/bin/iprowifi4965> iwn_notif_intr + 0x3c1
|
|---|
| 46 | 9 ffffffff9104af20 (+ 112) ffffffff80627738 </boot/system/non-packaged/add-ons/kernel/drivers/bin/iprowifi4965> iwn_intr + 0x488
|
|---|
| 47 | 10 ffffffff9104af90 (+ 32) ffffffff8065edcd </boot/system/non-packaged/add-ons/kernel/drivers/bin/iprowifi4965> intr_handler(void*) + 0x1d
|
|---|
| 48 | 11 ffffffff9104afb0 (+ 32) ffffffff800858f6 <kernel_x86_64> common_thread_entry(void*) + 0x37
|
|---|
| 49 | 12 ffffffff9104afd0 (+1861963824) ffffffff9104afe0
|
|---|
| 50 | ====
|
|---|
| 51 |
|
|---|
| 52 |
|
|---|
| 53 | The [iwn_tx_done + 0x1c6] is KASSERT line in if_iwn.c (iwn_tx_done()):
|
|---|
| 54 |
|
|---|
| 55 | KASSERT(data->ni != NULL, ("no node"));
|
|---|
| 56 |
|
|---|
| 57 | So, the problem is in race beetwin iwn_hw_stop and iwn_intr (->iwn_notify_intr-> ... iwn_tx_done)
|
|---|
| 58 |
|
|---|
| 59 | The race is possible due the fact, that iwn_intr called from thread (freebsd compat glue) and iwn_hw_stop can be
|
|---|
| 60 | called while iwn_intr is in way and not grab IWN_LOCK yet.
|
|---|
| 61 |
|
|---|
| 62 | So, the problem is likely Haiku only (not bug in freebsd code).
|
|---|
| 63 |
|
|---|
| 64 | I fixed it with patch, i have attached. I just skip interrupts called while iface is not running.
|
|---|
| 65 |
|
|---|
| 66 | But i am not sure if it ok, and if another drivers affects this too. So, i patched two modules only for example.
|
|---|
| 67 |
|
|---|
| 68 | Please, check it.
|
|---|
