Bloat » Cerowrt

Can't just fix #360 without looking at what else might have gone wrong..

I am presently in an incredibly radio-noisy environment, I recently updated the kernel on the laptop driving this test to 3.3.2, and the laptop has the known buggy iwl driver in it. And this is crypted...

Bit rate bounced around between 60-120Mbit on this test, and the cpu on the router ate less than 15%.

wlan0     IEEE 802.11abgn  ESSID:"testbox5"  
          Mode:Managed  Frequency:5.785 GHz  Access Point: E0:46:9A:34:30:D4   
          Bit Rate=120 Mb/s   Tx-Power=15 dBm   
          Retry  long limit:7   RTS thr:off   Fragment thr:off
          Power Management:off
          Link Quality=67/70  Signal level=-43 dBm  
          Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
          Tx excessive retries:1104096  Invalid misc:29226   Missed beacon:0

wlan0     IEEE 802.11abgn  ESSID:"testbox5"  
          Mode:Managed  Frequency:5.785 GHz  Access Point: E0:46:9A:34:30:D4   
          Bit Rate=90 Mb/s   Tx-Power=15 dBm   
          Retry  long limit:7   RTS thr:off   Fragment thr:off
          Power Management:off
          Link Quality=66/70  Signal level=-44 dBm  
          Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
          Tx excessive retries:1321241  Invalid misc:30513   Missed beacon:0

The excessive retries is excessive.

I am here testing the VI queue, but the results on the BE queue were similar.

m@cruithne:~/git/CerowrtCredits$ netperf -Y AF42,AF42 -l 60 -H 172.29.9.1
MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 172.29.9.1 () port 0 AF_INET
Recv   Send    Send                          
Socket Socket  Message  Elapsed              
Size   Size    Size     Time     Throughput  
bytes  bytes   bytes    secs.    10^6bits/sec  

 87380  65536  65536    60.03      46.40   

I'm very happy to see no unaligned instruction traps anymore and cpu usage so good.

It's also cool to be able to finally exercise all the queues, passing
-Y CS1,CS1 for background, or -Y AF42,AF42 for VI, or EF,EF for VO...

root@OpenWrt:/sys/kernel/debug/ieee80211/phy1/ath9k# cat xmit 
Num-Tx-Queues: 10  tx-queues-setup: 0x10f poll-work-seen: 1411
                            BE         BK        VI        VO

MPDUs Queued:             2361         43       805       568
MPDUs Completed:          2361         43       805       568
MPDUs XRetried:              0          0         0         0
Aggregates:              48881        569     11034         0
AMPDUs Queued HW:        96951        431      8016         0
AMPDUs Queued SW:       422582      10787    237179         0
AMPDUs Completed:       519028      11140    244406         0
AMPDUs Retried:          19303         13     27131         0
AMPDUs XRetried:           504          0       789         0
FIFO Underrun:               0          0         0         0
TXOP Exceeded:               0          0         0         0
TXTIMER Expiry:              0          0         0         0
DESC CFG Error:              0          0         0         0
DATA Underrun:               0          0         0         0
DELIM Underrun:              0          0         0         0
TX-Pkts-All:            521893      11183    246000       568
TX-Bytes-All:         51900022    1073513  23702065     79062
hw-put-tx-buf:               1          1         1         2
hw-tx-start:            192364       1315     48853       568
hw-tx-proc-desc:        192363       1313     48853       568
txq-memory-address:   828120e0   8281215c  82812064  82811fe8
axq-qnum:                    2          3         1         0
axq-depth:                   1          2         0         0
axq-ampdu_depth:             1          2         0         0
axq-stopped                  0          0         0         0
tx-in-progress               0          0         0         0
pending-frames               1         78         0         0
txq_headidx:                 0          0         0         0
txq_tailidx:                 0          0         0         0
axq_q empty:                   0          0         0         0
axq_acq empty:                 1          0         1         1
txq_fifo[0] empty:             1          1         1         1
txq_fifo[1] empty:             1          1         1         1
txq_fifo[2] empty:             1          1         1         1
txq_fifo[3] empty:             1          1         1         1
txq_fifo[4] empty:             1          1         1         1
txq_fifo[5] empty:             1          1         1         1
txq_fifo[6] empty:             1          1         1         1
txq_fifo[7] empty:             1          1         1         1

I'm aware of two patches for the ath9k that I don't have (one involving framing problems that seems a likely candidate), and the iwl sucks rocks in general.