Hi,
I would like to share my experience for compiling and running
Lilipond-1.2.9 under S.u.S.E-6.1 and S.u.S.E-6.2.
They identify themselves under YaST more precise as :
S.u.S.E-6.1.1-1
S.u.S.E-6.2.0-4
I use the standard libraries coming with both distributions.
Only two packages packages are different from standard :
o guile-1.3.4
o texinfo-4.0 ( for Lilypond 1.2.17 )
Intentionally I didn't change the line of code in
scalar.cc, to see if this would trigger a segfault.
Nothing the like happened. It compiled and runs without
crashing under S.u.S.E-6.1 and S.u.S.E-6.2.
Léon Kirsch.
---------
Andreas Schwab wrote:
> Jan Nieuwenhuizen <[EMAIL PROTECTED]> writes:
>
> |> On Monday, 29 November 1999, Karl Eichwalder writes:
> |>
> |> > "Howard E. Motteler" <[EMAIL PROTECTED]> writes:
> |> >
> |> > | Ekkehard Ellmann suggested simply
> |> > | reverting to SuSE 6.1, and I'll give that a try, at least on a
> |> > | temporary partition, to compare libraries.
> |> >
> |> > I've had working lilypond binaries on 6.1, 6.2 and 6.3pre --
> |> > unfortunately on the final release of 6.3 it segfaults. Until now, I
> |> > don't know the reason why.
> |>
> |> Which version? Do you have a stack trace? We were talking about a
> |> segfault without stack trace, before main () is entered.
>
> This is just due to a stupid bug in lilypond:
>
> --- scalar.cc 1999/12/10 14:29:36 1.1
> +++ scalar.cc 1999/12/10 14:30:56
> @@ -54,7 +54,7 @@
> if (length_i ())
> {
> long l =0;
> - conv = sscanf (strh_.ch_C (), "%lf", &l);
> + conv = sscanf (strh_.ch_C (), "%ld", &l);
> }
> return length_i () && conv;
> }
>
> BTW, the compiler has warned you. :-)
>
> Andreas.
>
> --
> Andreas Schwab "And now for something
> SuSE Labs completely different."
> [EMAIL PROTECTED]
> SuSE GmbH, Schanzäckerstr. 10, D-90443 Nürnberg
