I attempted to quickly try the 1.1.0 DirectFB today, but it
seems that it may require the 7.0.1 version of fusion.
I reached this conclusion based on the "make" of
1.1.0 DirectFB, where fails with this message:

 gcc -DHAVE_CONFIG_H -I. -I../../.. -I../../../include -I../../../lib
-I../../../include -I../../../lib -DDATADIR=\"/usr/local/share/directfb-
1.1.0\" -DMODULEDIR=\"/usr/local/lib/directfb-1.1-0\" -D_REENTRANT -O3
-ffast-math -pipe -D_GNU_SOURCE -Werror-implicit-function-declaration -MT
heap.lo -MD -MP -MF .deps/heap.Tpo -c heap.c  -fPIC -DPIC -o .libs/heap.o
In file included from ../../../lib/fusion/conf.h:33,
                 from heap.c:68:
../../../lib/fusion/types.h:39:2: error: #error Need major API version 7!
make[4]: *** [heap.lo] Error 1
make[4]: Leaving directory `/home/jmiskinis/dfb_1_1_0/DirectFB-1.1.0
/lib/fusion/shm'

I've been asked by my boss to verify that fusion 7.0.1 is
really needed, and that is what the error is stating.  He is
a little concerned about the jump in version numbers
between 3.2.4 and 7.0.1 and want to ensure we (I) am
heading in the right direction.

On Dec 3, 2007 2:17 PM, Denis Oliver Kropp <[EMAIL PROTECTED]> wrote:

> John Miskinis wrote:
> > Hi Denis,
> >
> > First, thanks for the reply, as I was unsure if my emails were reaching
> the
> > list.
> >
> > On Dec 3, 2007 1:34 PM, Denis Oliver Kropp <[EMAIL PROTECTED]> wrote:
> >
> >> John Miskinis wrote:
> >>> Hi Folks,
> >>>
> >>> I've been using directfb along with GTK+ for a couple months, and I've
> >>> been ignoring the "fusion call return" errors until now.  I am
> referring
> >>> to the messages such as:
> >>>
> >>> (!) FUSION_CALL_RETURN    --> Input/output error
> >> Which versions of DirectFB, GTK+ and especially linux-fusion are you
> >> using?
> >
> >
> > We are currently using DirectFB components:
> > linux-fusion-3.2.4  DirectFB-1.0.0  DFBTerm-0.8.15  LiTE-0.8.3
>
> Ok, the 1.0 branch does not have the new call serials for out of order
> execution.
>
> > Thaks for confirming that the error messages are indeed indicating that
> > something is
>
> In the 1.0 version the code is different and EIO there means that no
> remaining
> call execution has been found for returning, but I'm not sure at the
> moment how
> that could happen and what could happen afterwards.
>
> > really wrong.  I can try out the latest (1.1.0?) DirectFB later this
> week,
> > to see if the errors
> > vanish.  Perhaps whatever *is* wrong here, is also contributing to the
> > machine freeze-ups
> > that we see about twice-per-day on average.
>
> In such a case please send the content of /proc/fusion/0/skirmishs,
> fusionees and calls.
>
> --
> Best regards,
>  Denis Oliver Kropp
>
> .------------------------------------------.
> | DirectFB - Hardware accelerated graphics |
> | http://www.directfb.org/                 |
> "------------------------------------------"
>
_______________________________________________
directfb-dev mailing list
directfb-dev@directfb.org
http://mail.directfb.org/cgi-bin/mailman/listinfo/directfb-dev

Reply via email to