discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] gnuradio radio test repeatability


From: Cinaed Simson
Subject: Re: [Discuss-gnuradio] gnuradio radio test repeatability
Date: Thu, 8 Jun 2017 17:34:05 -0700
User-agent: Mozilla/5.0 (X11; Linux i686; rv:52.0) Gecko/20100101 Thunderbird/52.1.1

On 06/08/2017 01:47 AM, address@hidden wrote:
> I downloaded gnuradio off of github. Hopefully this is sufficient
> identification of the version:
> 
> "Latest commit 37d373a  14 hours ago @jmcorgan jmcorgan Merge branch
> 'maint'"
> 

It's the bleeding edge of 3.7.12.

> I noticed ZMQ failed, which is odd since it worked before. I reran the
> test and ZMQ worked. This makes me wonder if some variable isn't being
> set.

But you're not running the test in gr-zeromq - you're running the test
in gr-fec.

gr-zeromq has it own tests which don't depend on gsl.

The fact it can't run the sub test qa_zeromq_sub is an artifact of not
being able to run *ANY* of the tests in gr-fec with gsl-2.x.

In short, nothing has changed.

> 
> Here are the results when ZMQ worked:

False.

It just didn't print a Failed message - presumably a race condition -
which is not the same as Passed.

If 1 one tests passed , it would indicate the number tests which Passed
and the number of tests which Failed.

It's failing all the tests - it can't complete a single test.

> 
> Total Test time (real) = 210.83 sec
> 
> The following tests FAILED:
>          86 - qa_polar_encoder_systematic (Failed)
>          89 - qa_polar_encoder (Failed)
>          91 - qa_polar_decoder_sc (Failed)
>          92 - qa_polar_decoder_sc_list (Failed)
>          95 - qa_polar_decoder_sc_systematic (Failed)
>         162 - qa_packet_format (Failed)
> Errors while running CTest
> Makefile:72: recipe for target 'test' failed
> ------------------------------>
> Here we have ZMQ failing:
> 
> Total Test time (real) = 206.96 sec
> 
> The following tests FAILED:
>          86 - qa_polar_encoder_systematic (Failed)
>          89 - qa_polar_encoder (Failed)
>          91 - qa_polar_decoder_sc (Failed)
>          92 - qa_polar_decoder_sc_list (Failed)
>          95 - qa_polar_decoder_sc_systematic (Failed)
>         162 - qa_packet_format (Failed)
>         215 - qa_zeromq_sub (Failed)
>         216 - qa_zeromq_pub (Failed)
> Errors while running CTest
> Makefile:72: recipe for target 'test' failed
> make: *** [test] Error 8
> ----------------------------
> 
> I thought I'd try setting the affinity to one CPU with results:
> 
> taskset 0x1 make test
> Total Test time (real) = 206.35 sec
> 
> The following tests FAILED:
>          86 - qa_polar_encoder_systematic (Failed)
>          89 - qa_polar_encoder (Failed)
>          91 - qa_polar_decoder_sc (Failed)
>          92 - qa_polar_decoder_sc_list (Failed)
>          95 - qa_polar_decoder_sc_systematic (Failed)
>         215 - qa_zeromq_sub (Failed)
>         216 - qa_zeromq_pub (Failed)
> Errors while running CTest
> Makefile:72: recipe for target 'test' failed
> make: *** [test] Error 8
> ------------------
> 
> Well I see test 162 worked this time. Of course, is this repeatable?
> Next test:

You're

> 
> taskset 0x1 make test
> Total Test time (real) = 206.65 sec
> 
> The following tests FAILED:
>          86 - qa_polar_encoder_systematic (Failed)
>          89 - qa_polar_encoder (Failed)
>          91 - qa_polar_decoder_sc (Failed)
>          92 - qa_polar_decoder_sc_list (Failed)
>          95 - qa_polar_decoder_sc_systematic (Failed)
>         159 - qa_header_payload_demux (Failed)
>         162 - qa_packet_format (Failed)
>         215 - qa_zeromq_sub (Failed)
>         216 - qa_zeromq_pub (Failed)
> Errors while running CTest
> Makefile:72: recipe for target 'test' failed
> make: *** [test] Error 8
> -----------------
> 
> Not repeatable, plus an additional test (159) failed.
>

But it is repeatable - it fails everything every time.



> I don't think affinity is an issue. I ran the gnu
> system monitor and swap was never used. The PC has 8Gbytes and the
> maxium memory used never exceeded 1.9G. 
> 
> I'm going to do a memtest on the PC, but I kind of doubt there is a ram
> issue.
> 
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
> 




reply via email to

[Prev in Thread] Current Thread [Next in Thread]