Re: [PATCH] di-set, ino-map: new modules, from coreutils

[Paul Eggert]

On 02/07/2011 04:45 PM, Bruno Haible wrote:
> In lib/di-set.c, function di_ent_hash, there is the assumption that
> dev_t is an integral type. But POSIX
> <http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/sys_types.h.html>
> does not guarantee this. It could also be a floating-point type.

Let's leave that alone, or at best just put in a comment saying
we assume it's integral.  Nobody uses floating-point for dev_t,
and if anybody did, there would be a compile-time error anyway,
so no real harm done.  In that sense dev_t is the same as time_t,
which we also assume is an integer despite POSIX's saying it might
be floating.