https://bugs.kde.org/show_bug.cgi?id=388787
--- Comment #11 from Mark Wielaard <m...@klomp.org> --- I hadn't realized we were so short on bits. But I agree that reducing the szB is questionable, at least on 32 bit. On the other hand we might be keeping a stack of ExeContext pointers per allocation already. So maybe just making both the allockind and szB a full word isn't that dramatic an idea. But lets do that as a separate issue. Just one comment about the alignment being upgraded and being a power of 2 in ALLOC_or_NULL_ALIGNED and ALLOC_or_BOMB_ALIGNED. If I am reading the standard correctly then ff the value of an alignment argument is not a valid alignment value, then the behavior is undefined. So it might make sense to don't adjust/fixup the alignment in these cases, but let the mc_malloc_wrappers handle this, just like they handle bad sizes with record_fishy_value_error. -- You are receiving this mail because: You are watching all bug changes.
