Juan Lang a écrit :
I just ran across an evil little bug in the WINSPOOL_GetPrinter_2 function. It looks like this type of bug could be hiding in other API functions too. It causes a segmentation fault because of an unaligned access on Solaris (sparc).
Yikes. That's a bad one. The trouble is MS loves this sort of return value. Even if the dll itself doesn't dereference an unaligned pointer, the caller might depending on how things are layed out. The trouble is i386 allows unaligned memory access, so we don't see this sort of thing.
If you happen to know how MS handles alignment on platforms that require it, we might be able to fix it. Otherwise it'll have to be case-by-case I'm afraid.
MSVC supports __unaligned as pointer attribute. So declaring *(__unaligned long*)ptr will force the compiler (on CPUs/OSes which requires it) to generate the relevant code for this. It doesn't seem however that gcc supports such keyword.
A+