Sometimes one doesn't want to respect CFLAGS.  For example, the
bytesex.c file in the devel/hdf package only tries to determine the
endianness of the host machine and isn't part of the final compiled
package.  Using optimization might cause the compiler to be too clever
and give us a wrong result.

It should be alright in this case, but there are others where it would
cause the package to malfunction.  This is especially true when trying
to determine machine parameters, eg. math/slatec contains three files
which compile with no optimization and these end up in the final
library.

Thanks a lot for tidying up those Makefiles, but watch out
for this kind of thing.

Jason.

On Sat, Mar 31, 2001 at 02:14:32PM +0300, Tomasz Luchowski wrote:
> 
> Module Name:	pkgsrc
> Committed By:	zuntum
> Date:		Sat Mar 31 11:14:31 UTC 2001
> 
> Modified Files:
> 	pkgsrc/devel/gnome-objc: Makefile
> 	pkgsrc/devel/hdf: Makefile
> 	pkgsrc/lang/siod: Makefile
> 	pkgsrc/net/mouse-pppoe: Makefile
> 	pkgsrc/security/gnupg: Makefile
> 	pkgsrc/sysutils/sysinfo: Makefile
> 	pkgsrc/sysutils/tload: Makefile
> 	pkgsrc/x11/XmHTML: Makefile
> 	pkgsrc/x11/xlockmore: Makefile
> 
> Log Message:
> o Fix/add quoting
> o Respect ${CFLAGS}
> 
> 
> To generate a diff of this commit:
> cvs rdiff -r1.4 -r1.5 pkgsrc/devel/gnome-objc/Makefile
> cvs rdiff -r1.6 -r1.7 pkgsrc/devel/hdf/Makefile
> cvs rdiff -r1.3 -r1.4 pkgsrc/lang/siod/Makefile
> cvs rdiff -r1.6 -r1.7 pkgsrc/net/mouse-pppoe/Makefile
> cvs rdiff -r1.23 -r1.24 pkgsrc/security/gnupg/Makefile
> cvs rdiff -r1.2 -r1.3 pkgsrc/sysutils/sysinfo/Makefile
> cvs rdiff -r1.2 -r1.3 pkgsrc/sysutils/tload/Makefile
> cvs rdiff -r1.19 -r1.20 pkgsrc/x11/XmHTML/Makefile
> cvs rdiff -r1.41 -r1.42 pkgsrc/x11/xlockmore/Makefile
> 
> Please note that diffs are not public domain; they are subject to the
> copyright notices on the relevant files.
> 
>