I think it's going wrong in getActionFromSize, which easy as it
looks[1], I don't get the hang of.
Why do we need a specific QAction based on the FFT length we're setting?
And why does an FFT size of 4 lead to negative indices in a QList?
Anyway, the actual bug was on my side, as I was constantly
Hi!
Alfredo: Jep, did that. GDB was how I figured out things went wrong in
FreqDisplayForm::setFFTSize() in the first place :).
Jörg: Interesting; but the error is quite reproducibly at the same
place; I'll try to set a breakpoint at the kernel call and see if that
happens again. But that will req
Hey,
I know another GSoC Student has problems with the a GT GUI sink as well.
It would randomly SEGFAULT in a volk multiply kernel. I think it is the
volk_32_fc_32f_multiply_32fc kernel used in
waterfall_sink_X_impl::fft(). This kernel takes the FFT size as an input
argument, so it could be r
On Tue, Jun 3, 2014 at 4:35 PM, Marcus Müller wrote:
> I hope someone more familiar with the qtgui infrastructure might chime in.
>
Hey Marcus,
Did you try debugging the SEGFAULT using gdb? This is a good reference to
try:
http://www.unknownroad.com/rtfm/gdbtut/gdbsegfault.html
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
So I couldn't resist, and the line that would actually call the
set_fft_size callback function correctly would've been
set_fft_size(min(int($fftsize),4))
However, QT Gui spectacularly SEGFAULTs on me when doing that.
That actually happens in
line 12
