Miro Kropáček

I'd also explicitly mention that the effect of `STRUCTURE_SIZE_BOUNDARY=16` (one `char` in a `struct` being two bytes big) cannot be mitigated by `__attribute__((packed))`, the size will be two bytes as...

Now I know where the confusion came from: I mention in https://github.com/dxx-rebirth/dxx-rebirth/issues/695#issuecomment-1400986448 that `__attribute__((packed)) ` indeed did not help. However the reason why it didn't help was that I was...

Can you clarify what they mean by "making bit-fields work" ?

That would be most welcome! :)

That's some hardcore makefileing there. :) I hope @mfro0 will find some time to make a proper patch.

There you go: #122. ;)

Yes, this would be really useful. Additional idea - make it a part of PRG flags, so lazy programmers wont start creating buggy applications when under MP; existing (which are...

Ah ok, you meant it as an xaaes.cnf entry. Yep, that would work as well (although it would be limited, obviously, to XaAES only while PRG flags would work with...

That PRG flags was meant mostly for advanced users, i.e. we'd provide a tool to edit the new flag and if they encounter a strange memory violation, we'd tell them...

> Only if you change N.Aes / Geneva to also support that flag ;) Hmm, actually you are right. My initial idea was to support this purely on kernel level...