Hey guys, thanks for the responses!

I'm running Ubuntu 13.10 x64 and ICE version 3.5.1, I think, based on
pybombs list output:

ice    installed  inventory f00c59983cc904bca977133c0a9b3e8 wget://
http://www.zeroc.com/download/Ice/3.5/Ice-3.5.1.tar.gz

pybombs did successfully build ice from source.

Very Respectfully,

Dan CaJacob


On Thu, Jan 9, 2014 at 9:29 AM, Tom Rondeau <t...@trondeau.com> wrote:

> On Thu, Jan 9, 2014 at 7:15 AM, Martin Braun <martin.br...@ettus.com>
> wrote:
> > On Wed, Jan 08, 2014 at 10:18:20PM -0500, Dan CaJacob wrote:
> >> OK,
> >>
> >> Here's a specific error.  Seems like it's related to ICE, which compiled
> >> successfully as a dependency.
> >
> > Dan,
> >
> > can you please post your exact ICE version as well as other relevant
> > system info (distro, version etc.).
> >
> > MB
>
> I'll bet I can guess, the relevant parts at least. He's using Ice
> 3.4.2 and GCC >= 4.7. This is a nasty issue because of the default
> versions of many deb packages that don't actually work together.
> Either updating to Ice 3.5 or downgrading GCC to 4.6 will solve this
> issue. We were hoping for a patch against Ice 3.4.2, but I don't think
> that's going to happen.
>
> Now, PyBOMBS is set to look for Ice >= 3.5, though, so we still need
> more information. My guess is that you have Ice 3.4.2 installed on the
> system already and for some reason, GNU Radio is finding that and not
> 3.5, even though the script is supposed to look for 3.5 before
> anything else.
>
> In a slight look inside the process, I think it's time I do a blunt
> check for the GCC version if Ice 3.4 is found. I was holding off on
> this in hopes that a) it would be fixed in Ice and b) getting the
> compiler version isn't exactly simple. (Cmake does have a variable
> that holds the value of the compiler version, but it's only made
> available in newer versions of cmake, which we are not requiring yet,
> so we can't use it). But since it's only a problem if GCC is used, I
> can parse the version string from it and get the info I need. Much
> better to disable ControlPort in this case than to fail during
> compilation.
>
> Tom
>
> _______________________________________________
> Discuss-gnuradio mailing list
> Discuss-gnuradio@gnu.org
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>
_______________________________________________
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Reply via email to