#513: netgui alignment mismatching on 32bit/64bit machines

netgui data structure now is identical for all platforms. Manually aligned
the fields, so compilers do not add (CPU architecture-specific) padding.

src/Network/net_gui.hxx

@@ -20,7 +20,7 @@
 // and 8 bytes for double.
 
 
-const uint32_t FG_NET_GUI_VERSION = 7;
+const uint32_t FG_NET_GUI_VERSION = 8;
 
 
 // Define a structure containing the top level flight dynamics model
@@ -36,11 +36,21 @@ public:
         FG_MAX_TANKS = 4
     };
 
-    uint32_t version;		// increment when data values change
+// Note: align fields properly and manually to avoid incompatibilities
+// between 32bit and 64bit CPUs. Make sure that each field is already
+// placed on an offset which is a multiple of the size of its data
+// type, i.e. uint32/float need to have on offset of 4, doubles need
+// an offset of 8. This guarantees that compilers will _not_ add
+// CPU-specific padding bytes. Whenever in doubt about padding rules,
+// check "data structure alignment" in Wikipedia/Google :).
 
-    // Positions
+    uint32_t version;           // increment when data values change
+    uint32_t padding1;          // 4 padding bytes, so the next (64bit) var is aligned to 8
+
+    // Positions (note: offset for these doubles is already aligned to 8 - to avoid architecture specific alignments)
     double longitude;           // geodetic (radians)
     double latitude;            // geodetic (radians)
+
     float altitude;             // above sea level (meters)
     float agl;                  // above ground level (meters)
     float phi;                  // roll (radians)