This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH v2 1/2] Y2038: make __mktime_internal compatible with __time64_t


On 3/29/19 7:24 AM, Lukasz Majewski wrote:
> With __timespec64 exported (installed in usr/include/*) - there would
> be an explicit type to show that we are Y2038 safe (when e.g.
> debugging).

That's not much of an argument for exporting __timespec64 to user code.
When debugging a module, a developer can easily determine whether
whether time_t is 32- or 64-bit by printing sizeof (time_t) in GDB, so
the visibility of __timespec64 wouldn't give them any information that
they don't already have.

> I do have a setup where only for glibc compilation I do disable
> -D_TIME_BITS64, but for _all_ other source code it is enabled.

Sorry, I'm not following what you mean by "disable -DTIME_BITS64". Do
you mean that you compile applications with -D_TIME_BITS=64 and compile
glibc without it?

Does -D_TIME_BITS=64 affect building the GNU C library itself, and if
so, why? (Of course -D_TIME_BITS=64 will affect glibc's auxiliary
programs like zdump, but these are exceptional cases; I'm worried about
the library proper.) I thought _TIME_BITS was intended for use in user
programs, not for use in glibc itself.

The answers to the above questions should be documented in
manual/maint.texi; could you add that to your list of things to do?


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]