Quantcast

Floating point exception (core dumped)

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Floating point exception (core dumped)

Ruecan
Hello list,

1/ I have been stuck into debugging process of some code,
Am getting a: 'Floating point exception (core dumped)' which pop up randomly:

In other words, I have tried to find out was that related really to an arithmetic error of division by zero, and I fixed all possibilities of getting the error because of that, however now every possibility of getting division by zero is eliminated I still get the error and worse every time am running the code I get the exception raised from a different place.
Reading more on that exception on GR forum, I did not understood how could connecting a packet modulator with a USRP sink, raise such exceptions,

(the context is as follows: a "mod_pkts_sptr" connected to "usrp_sink"
                                    Inside the section of "mod_pkts_sptr" there is a thread generating packets and sending them continuously)

Please, if there is any hints or explanations I will be very pleased,
2/ Apart from that is there any accurate way to ask gdb to locate exactly which instruction is erroneous or which section of the code is causing trouble.  

3/ Is that a correct way to connect foo block to a currently described block:

opaque_self gr_hier_block2_self();
connect(foo, 0, self(), 0);  




Regards,
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Floating point exception (core dumped)

Ruecan
Hi again,

I just realized how to pinpoint the erroneous section from gdb,
by doing  'gdb --core=mycoredump ./NeoTx'
              >gdb run

It was the "gr_diff_encoder_bb::work" which cause the exception,
I am not initializing the variable of the modulus (the parameter to gr_make_diff_encoder_bb).
Loading...