Skip to contentSkip to conversation

TX-Power limits for ath10k based card on Linux 3.18.24 (long term support)?

DISCUSSION Joerg Pommnitz 2 participants 2 messages Nov 17, 2015 1 message not in the archive Qualcomm Atheros ATH10K wireless driver

lore
Share & tools
lore
#1 lore raw
In reply to a message not in the archive

Ben,
thanks for taking the time to answer. I followed your suggestion and repeated my tests with your beta-15 firmware.
The results are interesting. At first I thought nothing had changed. The txpower was about 10dB,

far short of the expected 20dB. Playing around with "iw dev ap set txpower fixed ... " did not result in significiant changes.

However, when I add an attenuator between the powermeter and the antenna, the measured txpower markedly increased. With 10dB attenuation I get 12dB, with 20dB attenuation the txpower increases to more than 15dB.

Question: Is this the expected behaviour? Can I force a fixed txpower instead of this dynamic adaptation?

-- Regards Joerg

Quoted 60 lines

Ben Greear <[email protected]> schrieb am 17:04 Dienstag, 17.November 2015:

Quoted 1 line

You might try CT firmware...I fixed some issues with tx-power for single chain

rates.

http://www.candelatech.com/ath10k.php

I suggest version 'beta-15', it is about to be officile release-15.

I'm curious to know the results if you do test with it.

Thanks,
Ben

Quoted 2 lines · Joerg Pommnitz

Hello all,
I'm trying to use a COMPEX WLE600VX (see

http://www.compex.com.sg:809/Datasheets/WLE600VX_Dsv1.0.2-141013-I.pdf) as an Access Point. According to the data sheet, the card should be able to transmit with about 20 dB (100mW). However, doing some measurements with a card on 2.4GHz channel 3 in 802.11g 54MBit coding I could only observe a TX power of about 7..9dB. During the measurement the card was under full load with iperf transmitting at 20..25 mbps. This is only a fraction of the stated TX power, so something strange must be going on. The measurements were done with a Marconi 6460 powermeter.

Quoted 18 lines

My hostapd.conf is very simple:
interface=ap
driver=nl80211
ssid2=<omitted>
channel=3
hw_mode=g
auth_algs=1
ignore_broadcast_ssid=0
eap_server=0
wpa=3
wpa_passphrase=<omitted>
wpa_key_mgmt=WPA-PSK WPA-PSK-SHA256
wpa_pairwise=TKIP CCMP
ctrl_interface=/var/run/hostapd
macaddr_acl=0

Quoted 1 line

The regdb (compiled directly into the kernel) shouldn't be a problem

either:

Quoted 14 lines

global
country 00: DFS-UNSET
(2402 - 2472 @ 40), (N/A, 100), (N/A)
(2457 - 2482 @ 40), (N/A, 100), (N/A)
(2474 - 2494 @ 20), (N/A, 20), (N/A), NO-OFDM, PASSIVE-SCAN (5170 - 5250 @ 80), (N/A, 20), (N/A), AUTO-BW, PASSIVE-SCAN (5250 - 5330 @ 80), (N/A, 20), (0 ms), DFS, AUTO-BW, PASSIVE-SCAN (5490 - 5730 @ 160), (N/A, 20), (0 ms), DFS, PASSIVE-SCAN (5735 - 5835 @ 80), (N/A, 20), (N/A), PASSIVE-SCAN (57240 - 63720 @ 2160), (N/A, 0), (N/A)

Does anybody have an idea, why the TX power seems to be so very low?

Quoted 12 lines
Quoted 7 lines

-- Regards Joerg

_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k

Signature · 2 lines
Ben Greear <[email protected]>
Candela Technologies Inc  http://www.candelatech.com

_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k

#2 lore raw
Quoted 10 lines · Joerg Pommnitz

Ben,
thanks for taking the time to answer. I followed your suggestion and repeated my tests with your beta-15 firmware.
The results are interesting. At first I thought nothing had changed. The txpower was about 10dB,

far short of the expected 20dB. Playing around with "iw dev ap set txpower fixed ... " did not result in significiant changes.

However, when I add an attenuator between the powermeter and the antenna, the measured txpower markedly increased. With 10dB attenuation I get 12dB, with 20dB attenuation the txpower increases to more than 15dB.

Question: Is this the expected behaviour? Can I force a fixed txpower instead of this dynamic adaptation?

I'm not aware of any logic in the firmware that would adjust power automatically based on attenuation on the antenna connectors.

Any chance the issue is how you are measuring things?

Maybe try the same test with an ath9k NIC and see how it performs?

Do you see similar results with other firmware when you add attenuation?

Are you connecting just to channel 0 on the NIC, or are you connecting to all three antenna ports? With 11g, you should only see signal on the first channel I think.

Thanks,
Ben

Signature · 8 lines
Ben Greear <[email protected]>
Candela Technologies Inc  http://www.candelatech.com


_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k