Re: [ptxdist] [ANNOUNCE] OSELAS.Toolchain() 2024.11.1 released

[Michael Olbrich <m.olbrich@pengutronix.de>]

Hi,

On Thu, Mar 13, 2025 at 10:36:25AM +0000, Ian Abbott wrote:
> On 13/03/2025 08:47, Michael Olbrich wrote:
> > I'm happy to announce that I've just released OSELAS.Toolchain-2024.11.1.
> > 
> > This is a bugfix only release. There are no major update. The relevant
> > changes are:
> 
> I'm guessing that this won't have the 64-bit time_t support in 32-bit
> libstdc++ since it's a bugfix only release. Is that correct?

Correct. And to be honest, I'm not sure how I should handle this. I've been
digging some more and there is a lot of code like this:

#ifdef _GLIBCXX_USE_CLOCK_GETTIME_SYSCALL
      syscall(SYS_clock_gettime, CLOCK_REALTIME, &tp);
#else
      clock_gettime(CLOCK_REALTIME, &tp);
#endif

And the problem with that is, that -D_TIME_BITS=64 only works correctly if
clock_gettime() is used and not the syscall directly. And I'm not sure how
to guarantee that.
And so far, I've not really found any indication that gcc upstream is
actually working on that in any way.

So I don't know what I will do for the next major release. Just building
with -D_TIME_BITS=64 for the official toolchains is not something that I'm
comfortable with right now.
Maybe an option to enable it and keep it disabled by default? That way, you
could at least build the toolchain with it in a defined way.

Michael

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |