[Spread-users] crash bug report

[Tim Peters]

[Mikhail Terekhov]
> Tim, after rereading the relevant part of the C standard you
> mentioned, I completely agree with you.

Then I completely agree with you too <wink>.


>> Did gcc, or did it not, also produce segfaults under the patched
>> code?  I didn't see any justification for gcc producing segfaults in
>> the patched version.

> No, gcc did not produce segfaults after my patch has been applied.
> At least for me.

Ah!  OK then, I misread part of the msg to which I first replied.  I agree
the patched code shouldn't segfault, so if it actually doesn't segfault,
that settles that.  Jonathan may still wish to keep his declaration as
int32, though, and trick gcc in some other way.  Since Spread wants to do
something C wants to discourage (moving the guts of an aligned type to a
possibly unaligned address), I'm afraid he's in for some obscurity one way
or another.  But that's why C supports comments too <wink>.