Hello, I'm trying to use the libRocket library for my projects but I stumbled upon a bug I can't figure out why it is happening. The problem is that it seems like a gcc problem. I extracted the code, and tried to simplify it so that others are able to easily test by themselves. The output of this code for me is the following: $ ./main Okay, the data_block->data_ptr has the address: 0xd37014 The address of this new StringBase is: 0xd37014 The string_id address is: 0xd3701400000000 Now, after string_id=0 the data_block->data_ptr has the address: 0 What is bugging me is that string_id (void*) is a member of StringBase and has nothing to do with data_block->data_ptr (void*), a member of Variant, and still, it starts out with the same address data_ptr is holding. Then I set string_id to 0, and by doing this, data_block->data_ptr is set to 0 too. It is much easier to understand this by running the code with gdb, my session goes something like this (the important part): // In Variant NewDataBlock method data_block = new DataBlock(); data_block->data_ptr = data; //data is a char[16], a buffer new (data_block->data_ptr) String(); // String is a StringBase<char> // In StringBase<char> constructor string_id = 0 // This changes the value of data_block->data_ptr (!??) If, for instance, I would set data_block->data_ptr = data, this would change the value of string_id to data too, inside the instance of class String (!?) (string_id is even protected) Anyway, the code is very easy to run and test, and the bug is very interesting. You can clone it from github: git clone git://github.com/hiltonm/cpptest.git Then run ./make.sh to build. It will create a libvariant.so and an executable 'main'. Nothing needs to be installed. Please, test it and let me know the output. I tested this with gcc 4.5 (and 4.3): $ gcc -v Using built-in specs. COLLECT_GCC=gcc COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-unknown-linux-gnu/4.5.1/lto-wrapper Target: x86_64-unknown-linux-gnu Configured with: ../configure --prefix=/usr --enable-languages=c,c++,fortran,objc,obj-c++,ada --enable-shared --enable-threads=posix --enable-__cxa_atexit --enable-clocale=gnu --enable-gnu-unique-object --enable-lto --enable-plugin --disable-multilib --disable-libstdcxx-pch --with-system-zlib --with-ppl --with-cloog --libdir=/usr/lib --libexecdir=/usr/lib --mandir=/usr/share/man --infodir=/usr/share/info Thread model: posix gcc version 4.5.1 (GCC) $ uname -a Linux hilton.laptop 2.6.35-ARCH #1 SMP PREEMPT Wed Sep 29 08:45:18 CEST 2010 x86_64 AMD Athlon(tm) 64 X2 Dual-Core Processor TK-57 AuthenticAMD GNU/Linux Any help would be really appreciated. Also, note that I removed lots of code and left only the necessary to test the bug, but there is some leftovers. Cheers, Hilton
