RE: 32-bit profiling counts?

[Jannick]

On Sat, 6 Jun 2020 14:47:27 -0400, Andrew J. Schorr wrote:
> On Sat, Jun 06, 2020 at 01:23:27PM -0500, Peter Lindgren wrote:
> > Given that “long” is a slippery quantity - sometimes 32-bit, sometimes 64-
> bit - depending on the platform, perhaps these counts should be defined as
> “long long”, or int64_t, or even uint64_t, which would be 64-bits everywhere.
> 
> Yup, that's the question: could we change that "long" to int64_t or uint64_t
> and would that break other parts of the code and what would it do to
> platform portability?  I can make either of those changes on my system and
> still pass "make check", but I don't know how it would impact other
> platforms.

For portability uintmax_t might be used, where its existence is ensured by 
AC_TYPE_UINTMAX_T.

>From the autoconf manual: Macro AC_TYPE_UINTMAX_T  - If stdint.h or inttypes.h 
>defines the type uintmax_t, define HAVE_UINTMAX_T. Otherwise, define uintmax_t 
>to the widest unsigned integer type.  

Best,
J.