Bug report

[Jonathan Melissant <jo...@gmail.com>]

During compilation of Celix for an ARM (a modified version of the arm926t)
processor, i've noticed a minor bug. I'm using the latest buildroot
(2012-02) with uclibc 9.33.

The celix framework uses the function fmaxl. Since this is not supported in
ucLibc 9.33 (which has been reported to uclibc as a bug), I noticed that
the function was used to compare 2 longs (instead of 2 doubles), and
therefore can be replaced by the following:

// framework->nextBundleId = fmaxl(framework->nextBundleId,  id + 1); //
origianl
// FIXME: 20120405 CHANGED TO REMOVE UNDEFINED fmaxl problem. Check later
if this "fix" is acceptable
            framework->nextBundleId = (framework->nextBundleId) > (id+1) ?
(framework->nextBundleId) : (id+1);

Re: Bug report

[Alexander Broekhuis <a....@gmail.com>]

Hi,

Thanks for your interest in running Celix on an ARM!

2012/4/10 Jonathan Melissant <jo...@gmail.com>

> During compilation of Celix for an ARM (a modified version of the arm926t)
> processor, i've noticed a minor bug. I'm using the latest buildroot
> (2012-02) with uclibc 9.33.
>

If you have this working, are you willing to post some sort of Howto? I
think it would be great to have this up on the Celix website!


>
> The celix framework uses the function fmaxl. Since this is not supported in
> ucLibc 9.33 (which has been reported to uclibc as a bug), I noticed that
> the function was used to compare 2 longs (instead of 2 doubles), and
> therefore can be replaced by the following:


> // framework->nextBundleId = fmaxl(framework->nextBundleId,  id + 1); //
> origianl
> // FIXME: 20120405 CHANGED TO REMOVE UNDEFINED fmaxl problem. Check later
> if this "fix" is acceptable
>            framework->nextBundleId = (framework->nextBundleId) > (id+1) ?
> (framework->nextBundleId) : (id+1);
>

Good catch, the fix looks ok to me. Could you file a bug at our Jira [1]?

[1]: https://issues.apache.org/jira/browse/CELIX



-- 
Met vriendelijke groet,

Alexander Broekhuis