1.30.10.0356 is the version I hade the trouble with. This was tested on a clean installation of windows 7.

I have also had some problems with single stepping and breakpoints in 1.30 that I did not have in 1.26. I have noticed that the source code and machine code relationships are sometimes incorrect and that lines that should have generated code cannot be used for source code breakpoints because the system apparently does not have a correct link to the associated machine code.

Hi,

There have been some major changes in the C compiler optimization with 1.30.10.0356. For his reason, breakpoints and even single stepping sometimes looks weird when the optimizations are active.

See http://forum.raisonance.com/viewtopic.php?id=2694 for details about this.

In case you see strange behavior in Ride7 while debugging a project with optimizations disabled, then it is a problem.
In such a case, please send us a zipped copy of your project and detailed instructions that may help us reproduce the problem you are facing.
As of now, the only known problem with 1.30.10.0356 is about the local variables that may have an invalid watch/evaluate.

Best Regards,

Hi,

fastmapper, what you describe looks very much like a problem introduced with the new version of gcc in the latest version of the RKit-ARM, which uses some elf constructs that the previous versions did not use and that Ride was not interpreting correctly. We found and corrected this a short time after the latest release, but we didn't release it because we thought it was happenning only very rarely... But maybe we got that wrong. ;)

Here is a test patch that solves it:
ftp://www.raisonance.com/temp/arm/test_patch_110207_ARM_1_30_HLL_ana.zip

It is not fully validated yet, so please tell us here the results of your tests, positive or negative. If the patch does not solve the problem, please send us (support@raisonance.com) a test project that shows your specific problem.

sima, your problem might be the same one or another one. Your description is too vague to be sure.
Please give a try at the patch above and if the problem persists, send us a test project that shows your problem.

Best Regards,

Vincent

Vincent

Thank you very much for providing a patch. I have tried it and it definitely addresses the issues I was seeing. I'll let you know if I see any similar problems while running with the patch.

... and thanks for the great customer service.

Vincent

I can also confirm that this patch is REQUIRED in order to perform any debugging. Without this patch, I was unable to set breakpoints in the editor on lines which should have breakpoints possible, breakpoints which were set did not get triggered, and in files with code longer than about 1000 lines, breakpoint setting capabilities disappear!

I believe that this patch (as an update) is necessary, otherwise debugging and breakpoints are unavailable. I don't believe this to be a rare occurrence, as it's happened to me as well - demonstrable in a number of source files.

Regards,
Robert.