[LetsCFD]

During the destruction of many classes, you can see something like this:

if (MG_PreSmooth!=NULL) delete [] MG_PreSmooth;

But the delete [] MG_PreSmooth also happens somewhere else in the class before the destruction. I guess the author may assume delete [] MG_PreSmooth will make the MG_PreSmooth become NULL. So, the destruction could skip it.

But the reality is that delete [] will never make a pointer NULL. So, the destruction will attempt to delete [] MG_PreSmooth again and it may destroy some other data.

The "new" and "delete" should be avoided as much as possible to favor the smart pointer management such as std::unique_ptr.

Right now, I have a lot of delete[] related crashes.

[hlk]

Originally Posted by LetsCFD

During the destruction of many classes, you can see something like this:

if (MG_PreSmooth!=NULL) delete [] MG_PreSmooth;

But the delete [] MG_PreSmooth also happens somewhere else in the class before the destruction. I guess the author may assume delete [] MG_PreSmooth will make the MG_PreSmooth become NULL. So, the destruction could skip it.

But the reality is that delete [] will never make a pointer NULL. So, the destruction will attempt to delete [] MG_PreSmooth again and it may destroy some other data.

The "new" and "delete" should be avoided as much as possible to favor the smart pointer management such as std::unique_ptr.

Right now, I have a lot of delete[] related crashes.

Please check out the most recent release - although we did not see memory related crashes in general, many issues similar to this have been fixed in the past couple of months. Please do let us know if you continue to see memory issues with the more recent code.