That's odd; I thought %I64 was only used by the Microsoft C library.
Maybe with -mno-cygwin, gcc is linking scanf to a version from the
Microsoft runtime library? Just as an experiment, what happens if you
do use the Cygwin library (i.e., don't give -mno-cygwin)?
The compiler warnings probably need to be fixed by using the correct
variable types. I'm surprised they aren't bugs... however, I haven't
seen the actual warnings so I don't really know what the problem is
there.
--Tim
On Tue, 02 Feb 2010 17:42:39 +0100, "h.g. muller" <[email protected]> wrote:
> The current source does not work correctly when compiled with gcc+Cygwin,
> which I use (with the -mno-cygwin flag) to generate the binary distribution
> and
> installer package.
>
> The u64Display format that is used (%lld) will not work to read all 64-bits
> in scanf,
> and lead to a phase error of any following arguments in printf. This leads
> to incorrect
> operation of node-based time controls, as they use the node count from the
> thinking
> output, which now has an un-initialized most-significant word.
>
> It seems that %I64u is needed here. This will produce compiler warnings for
> formats
> not matching the variable, but it does read and print the u64 nodes count
> correctly.
>
> Btw, I upgraded Cygwin, and the upgraded make now encounters a fatal error
> in the
> makefile. A dependency of wgamelist.c on "config.h." is listed there (with
> an extra
> trailing period!), and make now chokes on that. I see makefile.ms contains
> the same
> error (for as long as they exist).
>
>
> _______________________________________________
> Bug-XBoard mailing list
> [email protected]
> http://lists.gnu.org/mailman/listinfo/bug-xboard
>
--
Tim Mann [email protected] http://tim-mann.org/
_______________________________________________
Bug-XBoard mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-xboard
