Re: [PATCH v4 1/2] y2038: linux: Provide __timerfd_gettime64 implementation

[Arnd Bergmann <arnd@arndb.de>]

On Tue, Jan 7, 2020 at 3:25 PM Lukasz Majewski <lukma@denx.de> wrote:
> > On 07/01/2020 06:27, Lukasz Majewski wrote:
> >
> > >> As a side note, now that arch-syscall patch is upstream should we
> > >> assume that for !__ASSUME_TIME64_SYSCALLS the
> > >> __NR_timerfd_gettime64 should be defined (meaning that Linux
> > >> supports time64 for all 32-bit architectures)?
> > >
> > > Only Linux version >= 5.1 supports 64 bit time on archs with
> > > __WORDSIZE = 32. I do guess (but I may be wrong here) that the
> > > arch-syscall is supposed to reflect the exact syscalls provided by
> > > kernel headers used for building (to help with validation of Y2038
> > > patches).
> >
> > The arch-syscall is now autogenerated from the latest kernel release
> > defined in build-many-glibcs.py. So the question is whether Linux
> > support and enforces time64 support on all and future 32-bit
> > architectures or if there is still some missing ones (as it has
> > happen on some syscall additions, where some architecture lag
> > behind some releases).
>
> This question would be best answered by Arnd (CC'ed) IMHO. From what I
> know all 32 bit architectures gained syscalls covered by
> __ASSUME_TIME64_SYSCALLS from Linux 5.1+.

Yes, we intentionally converted all architectures at the same time to
have a reliable baseline, i.e. once a future glibc requires linux-5.1 as
the minimum kernel all the backwards-compatibility support for
old kernels can be dropped.

New 32-bit architectures (if any) will only support the time64 syscalls
and not time time32 ones.

For some ioctl interfaces, you also need to use the latest kernel
headers, e.g. sound/asound.h from kernels before 5.6 has some
bugs with time64. For the ioctl implementation I hope to wrap
up the final bits in linux-5.6 as well, earlier kernels may return
-EINVAL on some of the ioctls that pass a time_t.

        Arnd