Re: NSK(OSS) compilation problem

[Paul Eggert]

Matthew Woehlke <address@hidden> writes:

> Notice that it looks like the result of 'x>>y' under certain
> circumstances is '0x100000000 | (x & 0xFFFFFFFF)', regardless of the
> value of 'y'... which is of course Just Plain Wrong.

That's what we're looking for.  Ideally, I'd like a sample program
that fails to _compile_ due to the problem.  But I guess we'd settle
for one that fails to _run_.

>    1288 27a1ad6e467a7e63>>? == 1467a7e63 ?

OK, what happens if you compile and run the following program, with
and without -O?  What is the exit status of the program?  If it works,
then we don't yet have a complete handle on the problem.

#define BIG 0x27a1ad6e467a7e63LL
#define HALF_BIG 0x13d0d6b7233d3f31LL

#if BIG >> 1 != HALF_BIG
error;
#endif

int array[BIG >> 1 == HALF_BIG ? 1 : -1];

long long int big = BIG;

int main (void) { return big >> 1 != HALF_BIG; }