> > > > Yes, you are absolutely right. This was done in the beginning when > > automake understanding was not what it is now. > > If you open a bug at sourceforge.net, I will position it so that it > > gets fixed in one of the 2.11.x releases.
The bug is created as http://sourceforge.net/tracker/index.php?func=detail&aid=1939985&group_id=71730&atid=532251 and the removal of CFLAGS and COV_CFLAGS from AM_CFLAGS is done. Now I am working on the "NOCFLAGS" - the CFLAGS with filtered-out optimization. Is it really necessary to have such flags stored in separate variable and used only in the tests? Because it is quite hard to replace the content of user-supplied CFLAGS (eg. set in environment when running the configure script). There can be 2 ways to have a possibility to compile without optimization - rely on the user-supplied CFLAGS or add an option to configure (--disable-optimization) that will filter -Ox from CFLAGS as it is done now when --enable-testcover is supplied. What do you think about that? Dan ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone _______________________________________________ Openhpi-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/openhpi-devel
