|
From: | chai E |
Subject: | Re: [Discuss-gnuradio] Segmentation fault |
Date: | Wed, 25 Nov 2015 20:22:27 +0800 |
On Tue, Nov 24, 2015 at 10:47 AM, West, Nathan <address@hidden> wrote:We're short on information to actually determine what's wrong here. It looks like you're still using pdb, but I see different output/problem areas from your screenshots. The first one looks like it crashed in tb.start, and the second one looks like it crashed just creating the channel model block. Can you confirm what line is actually crashing, and report your gnuradio + volk version?If it's at runtime and you're using an older ettus image (rel3? rel2? I haven't kept up with versions and dates) you might be getting bit by this https://github.com/gnuradio/volk/issues/25 which can be fixed by either adding some lines to volk_config or by upgrading to that latest release.On Tue, Nov 24, 2015 at 9:40 AM, chai E <address@hidden> wrote:hi marcusi use the gdb as you said,the i debug the python file againrun the fie step by stepi found that the there is some problem with ‘channel model’,just like
you can see thatself.channels_channel_model_0 = channels.channel_model(function is init the channel model blockand the noise_voltage =0.0001 ,frequency_offset=0 and so on is the parameter of 'channel model'after the parameter is inited,the program should call theSubroutine to pass this parameters,but you can see that the program aborted,then i use command p to view the value of self.channels_channel_model_0,but i got the outputthere is no attribute of this ,in other wordsi think that there is some problem with the channel model in the pre-installed gnuradio.hope you can confirm this problem--EkkoMarcus is usually quite to point out this link. Surprised he hasn't so far :pThese are some techniques and ideas for debugging applications to better engage with us on the real problems:Tom
[Prev in Thread] | Current Thread | [Next in Thread] |