Re: AC_C_BIGENDIAN vs. Darwin

2001-07-02 Thread Ed L Cashin
Russ Allbery <[EMAIL PROTECTED]> writes: > Mike Castle <[EMAIL PROTECTED]> writes: > > > Personally, I'd vote for getting rid of AC_C_BIGENDIAN. No need to in > > supporting poor programming. > > I use it to allow more optimized versions of algorithms like MD5 where you > can skip additional m

cross-compiling ac_c_bigendian again (Re: AC_C_BIGENDIAN vs. Darwin)

2001-06-26 Thread Guido Draheim
Paul Eggert wrote: > [...] > does OpenSSH 2.9p1. So AC_C_BIGENDIAN works only for native compiles. > [...] > Also, if you want to throw in cross-compiles for Solaris 8 while > you're at it, append "|| defined _BIG_ENDIAN". > The topic of cross-compiling with ac_c_bigendian has been on the list

Re: AC_C_BIGENDIAN vs. Darwin

2001-06-26 Thread Mo DeJong
On Mon, 25 Jun 2001, Peter Eisentraut wrote: > Mike Castle writes: > > > On Sun, Jun 24, 2001 at 01:51:27PM -0700, Matt Watson wrote: > > > I'm trying to figure out the best way to fix this. Is the > > > > Fix the code so it doesn't require knowledge of the endianess of the > > machine. > > > >

Re: AC_C_BIGENDIAN vs. Darwin

2001-06-26 Thread Peter Eisentraut
Mike Castle writes: > On Sun, Jun 24, 2001 at 01:51:27PM -0700, Matt Watson wrote: > > I'm trying to figure out the best way to fix this. Is the > > Fix the code so it doesn't require knowledge of the endianess of the > machine. > > Personally, I'd vote for getting rid of AC_C_BIGENDIAN. No need

Re: AC_C_BIGENDIAN vs. Darwin

2001-06-26 Thread Paul Eggert
> From: Matt Watson <[EMAIL PROTECTED]> > Date: Mon, 25 Jun 2001 14:55:26 -0700 > > My vote too. How often does configure come across a system which doesn't > have a macro defining the byte order? Fairly often. Solaris 8 has two macros (_BIG_ENDIAN and _LITTLE_ENDIAN), but autoconf 2.50 doesn'

Re: AC_C_BIGENDIAN vs. Darwin

2001-06-25 Thread Russ Allbery
Mike Castle <[EMAIL PROTECTED]> writes: > Personally, I'd vote for getting rid of AC_C_BIGENDIAN. No need to in > supporting poor programming. I use it to allow more optimized versions of algorithms like MD5 where you can skip additional memory copies on hosts with some endianness. I'd be pret

Re: AC_C_BIGENDIAN vs. Darwin

2001-06-25 Thread Matt Watson
My vote too. How often does configure come across a system which doesn't have a macro defining the byte order? I've done this for now in the Darwin OpenSSH sources: #if (defined(BYTE_ORDER) && defined(BIG_ENDIAN) && (BYTE_ORDER == BIG_ENDIAN)) || defined(WORDS_BIGENDIAN) ... matt. On Monday,

Re: AC_C_BIGENDIAN vs. Darwin

2001-06-25 Thread Mike Castle
On Sun, Jun 24, 2001 at 01:51:27PM -0700, Matt Watson wrote: > I'm trying to figure out the best way to fix this. Is the Fix the code so it doesn't require knowledge of the endianess of the machine. Personally, I'd vote for getting rid of AC_C_BIGENDIAN. No need to in supporting poor programmi

Re: AC_C_BIGENDIAN vs. Darwin

2001-06-25 Thread Lars J. Aas
On Sun, Jun 24, 2001 at 01:51:27PM -0700, Matt Watson wrote: : (This issue caused OpenSSH to fail to configure properly on Darwin, so : it's not just a contrived example...) : : Darwin's (g)cc driver has the ability to build for multiple : architectures simultaneously: : : cc -arch i386 -arch

AC_C_BIGENDIAN vs. Darwin

2001-06-24 Thread Matt Watson
(This issue caused OpenSSH to fail to configure properly on Darwin, so it's not just a contrived example...) Darwin's (g)cc driver has the ability to build for multiple architectures simultaneously: cc -arch i386 -arch ppc foo.c This will invoke the compiler once for each architecture and cre