AnsweredAssumed Answered

libiio segmentation fault on write

Question asked by Daniel_Camara on Dec 5, 2014
Latest reply on Jan 9, 2015 by pcercuei

Hi everyone,

 

  I am trying to use a FMCOMMS4 card with and a zedboard and trying to run the examples that come with the analog device gnu-radio distribution but I am having a segmentation fault on "iio_channel_attr_write_raw () from /usr/lib/libiio.so.0", as I manage to get passing gdb over the core dump (gdb python core).  The core is also strange, it complains about possible corrupted stack (see below).

 

  I have seen this discussion (ZedBoard & fmcomms2 : running iiostream example) the problem is close, but I manage to run the ad9361-iiostream without a problem,  I try updating the image as advised, but no use. The interesting thing, the different is that the problem occurs only when I try to call the examples of gnu-radio. For example, if I try to call one of the examples on gnuradio/iio-examples (the top_block.py gennerated by cyclic-sine.grc) I get a core.

> gdb python core

(gdb) bt

[Thread debugging using libthread_db enabled]

Using host libthread_db library "/lib/arm-linux-gnueabihf/libthread_db.so.1".

Core was generated by `python ./top_block.py'.

Program terminated with signal 11, Segmentation fault.

#0  0xb42bc412 in iio_channel_attr_write_raw () from /usr/lib/libiio.so.0

(gdb) bt

#0  0xb42bc412 in iio_channel_attr_write_raw () from /usr/lib/libiio.so.0

#1  0xb42bca70 in iio_channel_attr_write_double () from /usr/lib/libiio.so.0

#2  0xb42d3f28 in gr::iio::fmcomms2_source_impl::set_params(unsigned long long, unsigned long, unsigned long, bool, bool, bool, char const*, double, char const*, double, char const*) () from /usr/lib/libgnuradio-iio-3.7.4git.so.0.0.0

#3  0xb42e0780 in vtable for gr::iio::fmcomms2_source_impl ()

   from /usr/lib/libgnuradio-iio-3.7.4git.so.0.0.0

#4  0xb42e0780 in vtable for gr::iio::fmcomms2_source_impl ()

   from /usr/lib/libgnuradio-iio-3.7.4git.so.0.0.0

Backtrace stopped: previous frame identical to this frame (corrupt stack?)

 

   Any ideas?!?!?

 

            Best regards...

 

               Daniel

Outcomes