> JonY <[email protected]> wrote:
> > This a little more serious - those references are, indeed, not defined
> > in libmpc.a. This smells to me like a bug in mpc itself. (Again, I'll
> > have to take a closer look when time permits.)
> >
>
> Hi,
> this looks like a real problem, I haven't encountered it before.
>
> I'll investigate further once I get the time; or until xeno shows up on
> irc, he's been doing trying to make a win64 toolchain too. I'm sure he
> has something to say.
I hacked my way past this problem as follows:
In both src/set_x.c and src/set_x_x.c I changed:
#if HAVE_INTTYPES_H
to
#if 1
and I changed
#ifdef _MPC_H_HAVE_INTMAX_T
to
#if 1
Then ran 'make distclean' and started afresh. All was then fine, and all tests
built and passed.
It seems that, somehow, HAVE_INTTYPES_H is untrue and _MPC_H_HAVE_INTMAX_T is
undefined at the 'make' stage, but HAVE_INTTYPES_H is true and
_MPC_H_HAVE_INTMAX_T is defined during the 'make check' stage.
I figured (perhaps incorrectly) that this was a problem with the mpc source,
and posted to the mpc-discuss mailing list. My post there still awaits
moderator approval as the post is just over their allowable 40kb in size. (I
think it's the config.log which I sent as an attachment with that post that's
providing most of that 40kb.)
Cheers,
Rob
------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
Mingw-w64-public mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mingw-w64-public
