Hello friend,
As far as I understand, `fractions` is used to compute a scalar variable `cs` that describes the volume fraction of each cell and a face vector `fs` that describes the face-based quantity based on a vertex scalar `phi`. So I imagine if you check `cs` and `fs` after the macro call, the fields must be fully defined.
On a second note,
“Or at least, removing this line allows the code to compile without floating point exceptions.“
I imagine you mean you can run the code. It should compile in any case. Floating point exceptions usually infer a division by 0. I guess you could try calling `fractions_cleanup` just to try and avoid very small volume fraction values.
Hope this helps,
Yiannis
--
You received this message because you are subscribed to the Google Groups "basilisk-fr" group.
To unsubscribe from this group and stop receiving emails from it, send an email to basilisk-fr...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/basilisk-fr/a38d5de9-40c6-4f51-8ce1-e2f6dc09ade3n%40googlegroups.com.
make sessile.tst
/basilisk/src/qcc -MD -o sessile.s.d sessile.c
/basilisk/src/qcc -MD -tags -o sessile.c.tags.d sessile.c
updating Makefile.deps
qcc -autolink -I/opt/local/include -O2 -I/opt/local/include/ -I/opt/X11/include/ -I/usr/local/include/ -g -Wall -pipe -o sessile/sessile sessile.c -lm
[sessile.tst]
/basilisk/src/runtest: line 51: 34887 Floating point exception: 8 $EXEC ./$test 2> $log > $out
make: *** [sessile.tst] Error 1