Re: [USRP-users] set_tx_freq is not functioning properly

2019-12-30 Thread Rob Kossler via USRP-users 
Perhaps the digital values are all zero for benchmark rate?

On Wed, Dec 18, 2019 at 7:02 AM voonna santosh via USRP-users <
usrp-users@lists.ettus.com> wrote:

> One more thing to add:
> If I run benchmark_rate with out any modifications, then the output in
> scope if perfect.
>
> I have added the following code snippet to configure the tx_freq and there
> is nothing observed in scope.
>
> if (vm.count("tx_freq")){
>   for(size_t ch = 0; ch < tx_channel_nums.size(); ch++) {
> std::cout << boost::format("Setting TX Freq: %f MHz...") %
> (tx_freq/1e6) << std::endl;
> uhd::tune_request_t tune_request(tx_freq);
> if(vm.count("int-n")) tune_request.args =
> uhd::device_addr_t("mode_n=integer");
> usrp->set_tx_freq(tune_request, tx_channel_nums[ch]);
> std::cout << boost::format("Actual TX Freq: %f MHz...") %
> (usrp->get_tx_freq(tx_channel_nums[ch])/1e6) << std::endl << std::endl;
> }
> }
>
> Command used for testing: ./benchmark_rate --args="addr=192.168.40.2"
> --channels="0" --tx_rate 50e6 --tx_freq 20e6 --duration 50
>
>
> NOTE: I am using Agilent E4402B to analyze the spectrum (9KHz to 3 GHz)
>
> Thanks,
> Santosh
>
>
>
>
>
> On Wednesday, December 18, 2019, 11:20:10 AM GMT, voonna santosh <
> santu_voo...@yahoo.com> wrote:
>
>
> FYI:
> Does something like tx_waveforms[1] output a signal at the expected
> frequency?
>  - Yes, both tx_wavefrom and benchmark_rate works fine. I took benchamark
> rate and extended it to get tx_freq configured. This is not working.
>
>
> With respect to the 10MHz, this is the frequency of signals used to
> discipline the internal timebase to an external source. I wouldn't expect
> artifacts from this signal to be present at the TX port under normal
> circumstances. Are you using an external reference? What about the GPSDO?
>   - Since my base line is benchmark_rate, the default frequency is 10MHz
> which always works fine. But when I change the frequency, thats when I see
> the issue. I couldn't trace where it is getting set, but it does. Can you
> please let me know where in the code, it sets the frequency to 10MHz ?
>
> BR,
> Santosh
>
>
>
>
>
> On Tuesday, December 17, 2019, 11:45:29 PM GMT, Sam Reiter <
> sam.rei...@ettus.com> wrote:
>
>
> Does something like tx_waveforms[1] output a signal at the expected
> frequency?
>
> What are you using to measure the frequency output?
>
> With respect to the 10MHz, this is the frequency of signals used to
> discipline the internal timebase to an external source. I wouldn't expect
> artifacts from this signal to be present at the TX port under normal
> circumstances. Are you using an external reference? What about the GPSDO?
>
> [1]
> https://github.com/EttusResearch/uhd/blob/master/host/examples/tx_waveforms.cpp
>
> Sam Reiter
> Ettus Research
>
>
> On Mon, Dec 16, 2019 at 9:01 AM voonna santosh via USRP-users <
> usrp-users@lists.ettus.com> wrote:
>
> Hi There,
>I have bench marked X300 against my development PC. Then I took
> benchmark_rate code and modified a bit. Everything works fine except
> setting the center freq.
>When ever I set the center frequency, there is no error returned by the
> call. But when I transmit the data, it doesn't appear in scope. Whenever I
> remove the code to set the center frequency, things work fine and I cann
> see a spike at 10Mhz which is default frequency I guess. Any clues?
>
> std::cout << boost::format("Setting TX Freq: %f MHz...") % (freq/1e6) <<
> std::endl;
> uhd::tune_request_t tune_request(freq);
> //usrp->set_tx_freq(tune_request, 0); - This line also doesn't work
> usrp->set_tx_freq(freq, 0);
> std::cout << boost::format("Actual TX Freq: %f MHz...") %
> (usrp->get_tx_freq(0)/1e6) << std::endl << std::endl;
>
> Thanks and Regards
> ___
> USRP-users mailing list
> USRP-users@lists.ettus.com
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>
> ___
> USRP-users mailing list
> USRP-users@lists.ettus.com
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>
___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

[USRP-users] DPDK 18.11

2019-12-30 Thread Damon Qiu via USRP-users 
___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Re: [USRP-users] E320 Auto-Booting function not working?

2019-12-30 Thread Kyle A Logue via USRP-users 
As of today the E320 mender image still seems to not autoboot correctly after 
seeting EEPROM. I can confirm it sets the bit correctly.

$ md5sum usrp_e320_fs.mender
f8e18a0667012594309102425278  usrp_e320_fs.mender

Kyle Logue
Engineering Manager ⚝ Comm Software Implementation Dept
The Aerospace Corporation
___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Re: [USRP-users] DPDK 18.11

2019-12-30 Thread Ettus Research Support via USRP-users 
Damon,

Compatibility with DPDK 18.11 should be included in UHD 4.0, but as of
right now there aren't any publicly available branches compatible with DPDK
18.11.

-Sam

On Mon, Dec 30, 2019 at 10:31 AM Damon Qiu via USRP-users <
usrp-users@lists.ettus.com> wrote:

>
> Hi all,
> Is there anyone knows which uhd branch supports DPDK 18.11? My network
> card needs to use this DPDK version.
> I have tried master-next branch, but a warning appears: "Detected use_dpdk
> argument, but DPDK support not built in."
> Best regards,
> Damon
>
>
>
>
> ___
> USRP-users mailing list
> USRP-users@lists.ettus.com
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>
___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Re: [USRP-users] DPDK build with N310

2019-12-30 Thread Sam Reiter via USRP-users 
Akis,

With UHD 3.14.1 you need to run DPDK 17.11

http://files.ettus.com/manual_archive/v3.14.1.0/html/page_dpdk.html

Best,

Sam Reiter
Ettus Research


On Fri, Dec 20, 2019 at 9:53 AM Akis Kourtis 
wrote:

> Hello Sam,
>
>
>
> We are using the DPDK 19.11, and UHD 3.14.1.
>
> The hardware setup is N310(SFP1)<->(ethernet-to-thunderbolt)<->Laptop
> (with thunderbolt).
>
> The problem with the config file is, that we could not find the config
> file anywhere neither on the root or the user folder.
>
> We have created one, and added the DPDK one (from the site). But I
> seriously believe that it is not seen/used.
>
> As the error
>
> No CHDR interfaces found!
>
> Is related to the no IP DPDK interfaces (presumably).
>
>
>
> Thank you for your support!
>
>
>
> *Akis Kourtis*
>
> M.Sc, Ph.D
>
> Research Associate
>
> Media Networks Laboratory
>
> Institute of Information & Telecommunications
>
> National Centre for Scientific Research “DEMOKRITOS”
>
>
>
> akis.kourtis@ iit.demokritos.gr
>
> +306948386769
>
> [image: engl LOGOTYPO XORIS ETHNOSIMO ME DIEYTHINSI_3]
>
>
>
> *From:* Sam Reiter 
> *Sent:* Wednesday, December 18, 2019 9:22 PM
> *To:* Akis Kourtis 
> *Cc:* usrp-users 
> *Subject:* Re: [USRP-users] DPDK build with N310
>
>
>
> Hey Akis,
>
>
>
> What version of DPDK are you using? What version of UHD do you have on the
> host? I'm not sure that mode of failure is something I'd chalk up to the
> DPDK install.
>
>
>
> Could you give some detail on your config file, DPDK install version(s),
> and hardware setup as well?
>
>
>
> Best,
>
>
>
> Sam Reiter
>
> Ettus Research
>
>
>
>
>
> On Mon, Dec 9, 2019 at 6:21 PM Akis Kourtis via USRP-users <
> usrp-users@lists.ettus.com> wrote:
>
> Hello all,
>
>
>
> I am trying to build the oai-5g-gNB. I have managed to build the DPDK with
> uhd successfully, however when I run the probe command I receive the
> following error.
>
>
>
> EAL: Starting I/O threads!
>
> [INFO] [UHD] linux; GNU C++ version 7.4.0; Boost_106501;
> UHD_3.14.1.HEAD-0-g0347a6d8
>
> [INFO] [MPMD] Initializing 1 device(s) in parallel with args:
> mgmt_addr=10.30.0.218,type=n3xx,product=n310,serial=3177E48,claimed=False,addr=192.168.20.2,second_addr=192.168.20.2,use_dpdk=1
>
> [ERROR] [MPMD] No viable transport path found!
>
> [ERROR] [MPMD] Failure during block enumeration: RuntimeError: No viable
> transport path found!
>
> [INFO] [MPM.PeriphManager] init() called with device args
> `time_source=internal,clock_source=internal,second_addr=192.168.20.2,mgmt_addr=10.30.0.218,product=n310,use_dpdk=1'.
>
> [WARNING] [MPM.PeriphManager.UDP] Number of detected CHDR devices is
> inconsistent. Dropped from 1 to 0.
>
> [INFO] [MPM.PeriphManager.UDP] No CHDR interfaces found!
>
> Error: RuntimeError: Failed to run enumerate_rfnoc_blocks()
>
>
>
> My guess from reading back posts from the forum, is that the proper DPDK
> file is not read.
>
> I can see a /etc/conf/uhd.conf file, but no /root/.uhd/uhd.conf file.
>
> Am I in the right direction, and if so, how do I enable the user conf file?
>
>
>
> If not, is there a direction I should head towards?
>
>
>
> Thank you,
>
>
>
> Akis
>
>
>
> *Akis Kourtis*
>
> M.Sc, Ph.D
>
> Research Associate
>
> Media Networks Laboratory
>
> Institute of Information & Telecommunications
>
> National Centre for Scientific Research “DEMOKRITOS”
>
>
>
> akis.kourtis@ iit.demokritos.gr
>
> +306948386769
>
> [image: engl LOGOTYPO XORIS ETHNOSIMO ME DIEYTHINSI_3]
>
>
>
> ___
> USRP-users mailing list
> USRP-users@lists.ettus.com
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>
>
___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Re: [USRP-users] set_tx_freq is not functioning properly

2019-12-30 Thread voonna santosh via USRP-users 
Yes .. all are zeros

Sent from Yahoo Mail on Android 
 
  On Mon, 30 Dec 2019 at 15:45, Rob Kossler wrote:   Perhaps 
the digital values are all zero for benchmark rate?
On Wed, Dec 18, 2019 at 7:02 AM voonna santosh via USRP-users 
 wrote:

 One more thing to add:If I run benchmark_rate with out any modifications, then 
the output in scope if perfect. 

I have added the following code snippet to configure the tx_freq and there is 
nothing observed in scope.
if (vm.count("tx_freq")){
  for(size_t ch = 0; ch < tx_channel_nums.size(); ch++) {
    std::cout << boost::format("Setting TX Freq: %f MHz...") % 
(tx_freq/1e6) << std::endl;
    uhd::tune_request_t tune_request(tx_freq);
    if(vm.count("int-n")) tune_request.args = 
uhd::device_addr_t("mode_n=integer");
    usrp->set_tx_freq(tune_request, tx_channel_nums[ch]);
    std::cout << boost::format("Actual TX Freq: %f MHz...") % 
(usrp->get_tx_freq(tx_channel_nums[ch])/1e6) << std::endl << std::endl;
    }
}
Command used for testing: ./benchmark_rate --args="addr=192.168.40.2" 
--channels="0" --tx_rate 50e6 --tx_freq 20e6 --duration 50


NOTE: I am using Agilent E4402B to analyze the spectrum (9KHz to 3 GHz)

Thanks,Santosh





On Wednesday, December 18, 2019, 11:20:10 AM GMT, voonna santosh 
 wrote:  
 
 FYI:Does something like tx_waveforms[1] output a signal at the expected 
frequency? - Yes, both tx_wavefrom and benchmark_rate works fine. I took 
benchamark rate and extended it to get tx_freq configured. This is not working.

With respect to the 10MHz, this is the frequency of signals used to discipline 
the internal timebase to an external source. I wouldn't expect artifacts from 
this signal to be present at the TX port under normal circumstances. Are you 
using an external reference? What about the GPSDO? 
  - Since my base line is benchmark_rate, the default frequency is 10MHz which 
always works fine. But when I change the frequency, thats when I see the issue. 
I couldn't trace where it is getting set, but it does. Can you please let me 
know where in the code, it sets the frequency to 10MHz ? 

BR,Santosh





   On Tuesday, December 17, 2019, 11:45:29 PM GMT, Sam Reiter 
 wrote:  
 
 Does something like tx_waveforms[1] output a signal at the expected frequency?
What are you using to measure the frequency output?
With respect to the 10MHz, this is the frequency of signals used to discipline 
the internal timebase to an external source. I wouldn't expect artifacts from 
this signal to be present at the TX port under normal circumstances. Are you 
using an external reference? What about the GPSDO? 
[1] 
https://github.com/EttusResearch/uhd/blob/master/host/examples/tx_waveforms.cpp

Sam Reiter Ettus Research

On Mon, Dec 16, 2019 at 9:01 AM voonna santosh via USRP-users 
 wrote:

Hi There,   I have bench marked X300 against my development PC. Then I took 
benchmark_rate code and modified a bit. Everything works fine except setting 
the center freq.   When ever I set the center frequency, there is no error 
returned by the call. But when I transmit the data, it doesn't appear in scope. 
Whenever I remove the code to set the center frequency, things work fine and I 
cann see a spike at 10Mhz which is default frequency I guess. Any clues?
std::cout << boost::format("Setting TX Freq: %f MHz...") % (freq/1e6) << 
std::endl;
uhd::tune_request_t tune_request(freq);
//usrp->set_tx_freq(tune_request, 0); - This line also doesn't work
usrp->set_tx_freq(freq, 0);
std::cout << boost::format("Actual TX Freq: %f MHz...") % 
(usrp->get_tx_freq(0)/1e6) << std::endl << std::endl;

Thanks and Regards
___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

  
___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Re: [USRP-users] set_tx_freq is not functioning properly

2019-12-30 Thread Rob Kossler via USRP-users 
If the digital values are all zeros, then there will not be much RF
output.  Sam previously requested that you try "tx_waveforms" and you
replied that it works fine.  Does this mean that you are able to set the
frequency to any desired value and then observe the expected result on your
spectrum analyzer?

Rob

On Mon, Dec 30, 2019 at 4:44 PM voonna santosh 
wrote:

> Yes .. all are zeros
>
> Sent from Yahoo Mail on Android
> 
>
> On Mon, 30 Dec 2019 at 15:45, Rob Kossler
>  wrote:
> Perhaps the digital values are all zero for benchmark rate?
>
> On Wed, Dec 18, 2019 at 7:02 AM voonna santosh via USRP-users <
> usrp-users@lists.ettus.com> wrote:
>
> One more thing to add:
> If I run benchmark_rate with out any modifications, then the output in
> scope if perfect.
>
> I have added the following code snippet to configure the tx_freq and there
> is nothing observed in scope.
>
> if (vm.count("tx_freq")){
>   for(size_t ch = 0; ch < tx_channel_nums.size(); ch++) {
> std::cout << boost::format("Setting TX Freq: %f MHz...") %
> (tx_freq/1e6) << std::endl;
> uhd::tune_request_t tune_request(tx_freq);
> if(vm.count("int-n")) tune_request.args =
> uhd::device_addr_t("mode_n=integer");
> usrp->set_tx_freq(tune_request, tx_channel_nums[ch]);
> std::cout << boost::format("Actual TX Freq: %f MHz...") %
> (usrp->get_tx_freq(tx_channel_nums[ch])/1e6) << std::endl << std::endl;
> }
> }
>
> Command used for testing: ./benchmark_rate --args="addr=192.168.40.2"
> --channels="0" --tx_rate 50e6 --tx_freq 20e6 --duration 50
>
>
> NOTE: I am using Agilent E4402B to analyze the spectrum (9KHz to 3 GHz)
>
> Thanks,
> Santosh
>
>
>
>
>
> On Wednesday, December 18, 2019, 11:20:10 AM GMT, voonna santosh <
> santu_voo...@yahoo.com> wrote:
>
>
> FYI:
> Does something like tx_waveforms[1] output a signal at the expected
> frequency?
>  - Yes, both tx_wavefrom and benchmark_rate works fine. I took benchamark
> rate and extended it to get tx_freq configured. This is not working.
>
>
> With respect to the 10MHz, this is the frequency of signals used to
> discipline the internal timebase to an external source. I wouldn't expect
> artifacts from this signal to be present at the TX port under normal
> circumstances. Are you using an external reference? What about the GPSDO?
>   - Since my base line is benchmark_rate, the default frequency is 10MHz
> which always works fine. But when I change the frequency, thats when I see
> the issue. I couldn't trace where it is getting set, but it does. Can you
> please let me know where in the code, it sets the frequency to 10MHz ?
>
> BR,
> Santosh
>
>
>
>
>
> On Tuesday, December 17, 2019, 11:45:29 PM GMT, Sam Reiter <
> sam.rei...@ettus.com> wrote:
>
>
> Does something like tx_waveforms[1] output a signal at the expected
> frequency?
>
> What are you using to measure the frequency output?
>
> With respect to the 10MHz, this is the frequency of signals used to
> discipline the internal timebase to an external source. I wouldn't expect
> artifacts from this signal to be present at the TX port under normal
> circumstances. Are you using an external reference? What about the GPSDO?
>
> [1]
> https://github.com/EttusResearch/uhd/blob/master/host/examples/tx_waveforms.cpp
>
> Sam Reiter
> Ettus Research
>
>
> On Mon, Dec 16, 2019 at 9:01 AM voonna santosh via USRP-users <
> usrp-users@lists.ettus.com> wrote:
>
> Hi There,
>I have bench marked X300 against my development PC. Then I took
> benchmark_rate code and modified a bit. Everything works fine except
> setting the center freq.
>When ever I set the center frequency, there is no error returned by the
> call. But when I transmit the data, it doesn't appear in scope. Whenever I
> remove the code to set the center frequency, things work fine and I cann
> see a spike at 10Mhz which is default frequency I guess. Any clues?
>
> std::cout << boost::format("Setting TX Freq: %f MHz...") % (freq/1e6) <<
> std::endl;
> uhd::tune_request_t tune_request(freq);
> //usrp->set_tx_freq(tune_request, 0); - This line also doesn't work
> usrp->set_tx_freq(freq, 0);
> std::cout << boost::format("Actual TX Freq: %f MHz...") %
> (usrp->get_tx_freq(0)/1e6) << std::endl << std::endl;
>
> Thanks and Regards
> ___
> USRP-users mailing list
> USRP-users@lists.ettus.com
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>
> ___
> USRP-users mailing list
> USRP-users@lists.ettus.com
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>
>
___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Re: [USRP-users] DPDK 18.11

2019-12-30 Thread qiu.guowang...@gmail.com via USRP-users 
___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com