Skip to contentSkip to conversation

Issue with ATH10K driver with 802.11 a/g mode datarates

DISCUSSION Sridhar Kondabathini 3 participants 7 messages Sep 19, 2016 – Sep 20, 2016 Qualcomm Atheros ATH10K wireless driver

lore
Share & tools
lore
#0 lore raw

Hi,

When we are testing 802.11 a/g mode throughput measurements on T1024RDB ( we used
kernel-3.12; ath10k drivers backported from 4.4;hostapd 2.5 for 11AC support), then we are getting 26-27 Mbps , which is not meeting the required TEC specification (which is ~28 Mbps). How can we increase the throughput in these modes.

Can you please help on this.

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely for
the use of the individual or entity to whom they are addressed.  If you are not the intended recipient you are notified that disclosing, copying, distributing or taking any
action in reliance on the contents of this information is strictly prohibited. The sender does not a

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

#1 lore raw

Out of curiosity, does ath9k perform at the required rates in a similar setup?

Sending tcp, udp, or some other protocol? How are you generating the traffic?

Thanks,
Ben

Quoted 38 lines · Sridhar Kondabathini

Hi,

When we are testing 802.11 a/g mode throughput measurements on T1024RDB ( we used
kernel-3.12; ath10k drivers backported from 4.4;hostapd 2.5 for 11AC support), then we are getting 26-27 Mbps , which is not meeting the required TEC specification (which is ~28 Mbps). How can we increase the throughput in these modes.

Can you please help on this.

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely for
the use of the individual or entity to whom they are addressed. If you are not the intended recipient you are notified that disclosing, copying, distributing or taking any
action in reliance on the contents of this information is strictly prohibited. The sender does not a

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

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
#2 lore raw

... and what's "TEC specification" ?

-
a

Quoted 65 lines · Ben Greear

Out of curiosity, does ath9k perform at the required rates in a similar setup?

Sending tcp, udp, or some other protocol? How are you generating the traffic?

Thanks,
Ben

Quoted 41 lines · Sridhar Kondabathini

Hi,

When we are testing 802.11 a/g mode throughput measurements on
T1024RDB ( we used
kernel-3.12; ath10k drivers backported from 4.4;hostapd 2.5 for 11AC support), then we
are getting 26-27 Mbps , which is not meeting the required TEC specification (which is
~28 Mbps). How can we increase the throughput in these modes.

Can you please help on this.

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely
for
the use of the individual or entity to whom they are addressed. If you are not the
intended recipient you are notified that disclosing, copying, distributing or taking
any
action in reliance on the contents of this information is strictly prohibited. The
sender does not a

Quoted 5 lines

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

--
Ben Greear <[email protected]>
Candela Technologies Inc http://www.candelatech.com

Quoted 4 lines

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

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

#3 lore raw

Hi,
I didnt try with ath9k drivers on the same setup.

we are using WiFi capacity test tool on Lanforge to test the performance, this will
generate the traffic. we are selecting the UDP protocol (packet - 1472 Bytes).

As per TEC spec. required data rates are given below.

Mode Asso. rate Throughput

802.11a 54 Mbps 28 Mbps

802.11g 54 Mbps 28 Mbps

But we are getting only 26-27Mbps, how to improve this throughput ?

[TEC(Telecommunication Engineering Centre, Govt. India) is a technical body,
prepares specification of common Telecom standards.]

Regards,
Sridhar

On Mon, 19 Sep 2016 10:22:26 -0700, Adrian Chadd wrote

Quoted 76 lines

.... and what's "TEC specification" ?

-
a

Quoted 65 lines · Ben Greear

Out of curiosity, does ath9k perform at the required rates in a similar setup?

Sending tcp, udp, or some other protocol? How are you generating the traffic?

Thanks,
Ben

Quoted 41 lines · Sridhar Kondabathini

Hi,

When we are testing 802.11 a/g mode throughput measurements on
T1024RDB ( we used
kernel-3.12; ath10k drivers backported from 4.4;hostapd 2.5 for 11AC support), then we
are getting 26-27 Mbps , which is not meeting the required TEC specification (which is
~28 Mbps). How can we increase the throughput in these modes.

Can you please help on this.

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely
for
the use of the individual or entity to whom they are addressed. If you are not the
intended recipient you are notified that disclosing, copying, distributing or taking
any
action in reliance on the contents of this information is strictly prohibited. The
sender does not a

Quoted 5 lines

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

--
Ben Greear <[email protected]>
Candela Technologies Inc http://www.candelatech.com

Quoted 4 lines

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

Quoted 4 lines

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

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely for
the use of the individual or entity to whom they are addressed.  If you are not the intended recipient you are notified that disclosing, copying, distributing or taking any
action in reliance on the contents of this information is strictly prohibited. The sender does not a

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

#4 lore raw

LANforge is going to be running a modified driver and firmware, though likely it is no worse at performance in this case than stock code.

What is the device under test?

And, are you in an isolation chamber, using SMA cables, attenuation, etc?

There are lots of things that can impact wifi throughput so you have to provide lots more details in order for anyone to understand why the throughput might be less than expected.

And if it is specific to LANforge, please send email to [email protected]

Thanks,
Ben

Quoted 127 lines · Sridhar Konabathini

Hi,
I didnt try with ath9k drivers on the same setup.

we are using WiFi capacity test tool on Lanforge to test the performance, this will
generate the traffic. we are selecting the UDP protocol (packet - 1472 Bytes).

As per TEC spec. required data rates are given below.

Mode Asso. rate Throughput

802.11a 54 Mbps 28 Mbps

802.11g 54 Mbps 28 Mbps

But we are getting only 26-27Mbps, how to improve this throughput ?

[TEC(Telecommunication Engineering Centre, Govt. India) is a technical body,
prepares specification of common Telecom standards.]

Regards,
Sridhar

On Mon, 19 Sep 2016 10:22:26 -0700, Adrian Chadd wrote

Quoted 76 lines

.... and what's "TEC specification" ?

-
a

Quoted 65 lines · Ben Greear

Out of curiosity, does ath9k perform at the required rates in a similar setup?

Sending tcp, udp, or some other protocol? How are you generating the traffic?

Thanks,
Ben

Quoted 41 lines · Sridhar Kondabathini

Hi,

When we are testing 802.11 a/g mode throughput measurements on
T1024RDB ( we used
kernel-3.12; ath10k drivers backported from 4.4;hostapd 2.5 for 11AC support), then we
are getting 26-27 Mbps , which is not meeting the required TEC specification (which is
~28 Mbps). How can we increase the throughput in these modes.

Can you please help on this.

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely
for
the use of the individual or entity to whom they are addressed. If you are not the
intended recipient you are notified that disclosing, copying, distributing or taking
any
action in reliance on the contents of this information is strictly prohibited. The
sender does not a

Quoted 5 lines

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

--
Ben Greear <[email protected]>
Candela Technologies Inc http://www.candelatech.com

Quoted 4 lines

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

Quoted 4 lines

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

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely for
the use of the individual or entity to whom they are addressed. If you are not the intended recipient you are notified that disclosing, copying, distributing or taking any
action in reliance on the contents of this information is strictly prohibited. The sender does not a

Signature · 7 lines
Ben Greear <[email protected]>
Candela Technologies Inc  http://www.candelatech.com

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

Hi,
Yes, we are doing the testing in an isolation chamber (no attenuators used), antennas
are connected to RF card via SMA cable. Tx power is 5dBm. we are not suspecting Lanforge as it gives required throughput in different WiFi setup.
here Device Under test is T2014 based card.

Regards,
Sridhar

On Mon, 19 Sep 2016 21:33:03 -0700, Ben Greear wrote

Quoted 11 lines

LANforge is going to be running a modified driver and firmware, though likely it is no worse at performance in this case than stock code.

What is the device under test?

And, are you in an isolation chamber, using SMA cables, attenuation, etc?

There are lots of things that can impact wifi throughput so you have to provide lots more details in order for anyone to understand why the throughput might be less than expected.

Quoted 10 lines

And if it is specific to LANforge, please send email to [email protected]

Thanks,
Ben

Quoted 4 lines · Sridhar Konabathini

Hi,
I didnt try with ath9k drivers on the same setup.

we are using WiFi capacity test tool on Lanforge to test the performance, this

will

Quoted 113 lines
Quoted 113 lines

generate the traffic. we are selecting the UDP protocol (packet - 1472 Bytes).

As per TEC spec. required data rates are given below.

Mode Asso. rate Throughput

802.11a 54 Mbps 28 Mbps

802.11g 54 Mbps 28 Mbps

But we are getting only 26-27Mbps, how to improve this throughput ?

[TEC(Telecommunication Engineering Centre, Govt. India) is a technical body,
prepares specification of common Telecom standards.]

Regards,
Sridhar

On Mon, 19 Sep 2016 10:22:26 -0700, Adrian Chadd wrote

Quoted 76 lines

.... and what's "TEC specification" ?

-
a

Quoted 65 lines · Ben Greear

Out of curiosity, does ath9k perform at the required rates in a similar setup?

Sending tcp, udp, or some other protocol? How are you generating the traffic?

Thanks,
Ben

Quoted 41 lines · Sridhar Kondabathini

Hi,

When we are testing 802.11 a/g mode throughput measurements on
T1024RDB ( we used
kernel-3.12; ath10k drivers backported from 4.4;hostapd 2.5 for 11AC support), then we
are getting 26-27 Mbps , which is not meeting the required TEC specification (which is
~28 Mbps). How can we increase the throughput in these modes.

Can you please help on this.

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely
for
the use of the individual or entity to whom they are addressed. If you are not the
intended recipient you are notified that disclosing, copying, distributing or taking
any
action in reliance on the contents of this information is strictly prohibited. The
sender does not a

Quoted 5 lines

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

--
Ben Greear <[email protected]>
Candela Technologies Inc http://www.candelatech.com

Quoted 4 lines

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

Quoted 4 lines

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

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Quoted 19 lines
Quoted 10 lines

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely for
the use of the individual or entity to whom they are addressed. If you are not the intended recipient you are notified that disclosing, copying, distributing or taking any
action in reliance on the contents of this information is strictly prohibited. The sender does not a

Signature · 7 lines
Ben Greear <[email protected]>
Candela Technologies Inc  http://www.candelatech.com

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

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely for
the use of the individual or entity to whom they are addressed.  If you are not the intended recipient you are notified that disclosing, copying, distributing or taking any
action in reliance on the contents of this information is strictly prohibited. The sender does not a

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

#6 lore raw

You normally need to add 30-60 db of attenuation to get best throughput when directly cabling with SMA cables. Please try adding more attenuation.

What firmware are you using on the T1024RDB? Maybe also try using a different firmware version to see if that helps.

Thanks,
Ben

Quoted 187 lines · Sridhar Konabathini

Hi,
Yes, we are doing the testing in an isolation chamber (no attenuators used), antennas
are connected to RF card via SMA cable. Tx power is 5dBm. we are not suspecting Lanforge as it gives required throughput in different WiFi setup.
here Device Under test is T2014 based card.

Regards,
Sridhar

On Mon, 19 Sep 2016 21:33:03 -0700, Ben Greear wrote

Quoted 11 lines

LANforge is going to be running a modified driver and firmware, though likely it is no worse at performance in this case than stock code.

What is the device under test?

And, are you in an isolation chamber, using SMA cables, attenuation, etc?

There are lots of things that can impact wifi throughput so you have to provide lots more details in order for anyone to understand why the throughput might be less than expected.

Quoted 10 lines

And if it is specific to LANforge, please send email to [email protected]

Thanks,
Ben

Quoted 4 lines · Sridhar Konabathini

Hi,
I didnt try with ath9k drivers on the same setup.

we are using WiFi capacity test tool on Lanforge to test the performance, this

will

Quoted 124 lines
Quoted 113 lines

generate the traffic. we are selecting the UDP protocol (packet - 1472 Bytes).

As per TEC spec. required data rates are given below.

Mode Asso. rate Throughput

802.11a 54 Mbps 28 Mbps

802.11g 54 Mbps 28 Mbps

But we are getting only 26-27Mbps, how to improve this throughput ?

[TEC(Telecommunication Engineering Centre, Govt. India) is a technical body,
prepares specification of common Telecom standards.]

Regards,
Sridhar

On Mon, 19 Sep 2016 10:22:26 -0700, Adrian Chadd wrote

Quoted 76 lines

.... and what's "TEC specification" ?

-
a

Quoted 65 lines · Ben Greear

Out of curiosity, does ath9k perform at the required rates in a similar setup?

Sending tcp, udp, or some other protocol? How are you generating the traffic?

Thanks,
Ben

Quoted 41 lines · Sridhar Kondabathini

Hi,

When we are testing 802.11 a/g mode throughput measurements on
T1024RDB ( we used
kernel-3.12; ath10k drivers backported from 4.4;hostapd 2.5 for 11AC support), then we
are getting 26-27 Mbps , which is not meeting the required TEC specification (which is
~28 Mbps). How can we increase the throughput in these modes.

Can you please help on this.

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely
for
the use of the individual or entity to whom they are addressed. If you are not the
intended recipient you are notified that disclosing, copying, distributing or taking
any
action in reliance on the contents of this information is strictly prohibited. The
sender does not a

Quoted 5 lines

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

--
Ben Greear <[email protected]>
Candela Technologies Inc http://www.candelatech.com

Quoted 4 lines

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

Quoted 4 lines

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

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Quoted 10 lines

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely for
the use of the individual or entity to whom they are addressed. If you are not the intended recipient you are notified that disclosing, copying, distributing or taking any
action in reliance on the contents of this information is strictly prohibited. The sender does not a

Quoted 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

Regards,

Sridhar Kondabathini,
Research engineer,
PSG lab,
Center for Development Of Telematics
9480484821.

Disclaimer:
----------
This email and any files transmitted with it are confidential and intended solely for
the use of the individual or entity to whom they are addressed. If you are not the intended recipient you are notified that disclosing, copying, distributing or taking any
action in reliance on the contents of this information is strictly prohibited. The sender does not a

Signature · 7 lines
Ben Greear <[email protected]>
Candela Technologies Inc  http://www.candelatech.com

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