Date: 08 Apr 2001 20:18:26 +0100 From: Rolf Neugebauer <neugebar@dcs.gla.ac.uk> To: freebsd-alpha@FreeBSD.ORG Subject: Re: 64bits mixups (Was: Re: Vinum status) Message-ID: <ysqr8z32oak.fsf@therese.dcs.gla.ac.uk> In-Reply-To: Andrew Gallatin's message of "Sun, 8 Apr 2001 11:58:47 -0400 (EDT)" References: <20010408030913.213D1380A@overcee.netplex.com.au> <3AD05711.7000002@videotron.ca> <15056.35511.810402.576637@grasshopper.cs.duke.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
Andrew Gallatin <gallatin@cs.duke.edu> writes: > tcn writes: > > It is an error to keep ints of 32 bits on 64 bits machine like it > > would have been an error on ia32 to keep ints of 16bits. (I come from a > > dos world...) > > It is not an error. It is standard industry practice, every major > vendor with a 64-bit platform follows lp64. Almost nobody chose > ilp64. Go read http://www.opengroup.org/public/tech/aspen/lp64_wp.htm > > I think the pointer size manipulation you're referring to in Compaq's > CXX is the old Tru64 taso (Truncated Address Space) trick where they > use 32-bit pointers and make sure that the entire program's address > space fits in 4G. This can indeed be useful to run old, buggy > code like your aliance program. taso actually is a linker option, ie independent of the compiler. There is also a version of the GNU ld which implements this flag. IIRC this is not (yet?) part of the standard binutils distro, but is in a separate Linux branch. I once installed it on an Alpha/Linux box but can't remember were I downloaded it from :( If people are interested, there was a discussion about this on the alpha linux mailing list a while back which I could dig out my archive. Rolf To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?ysqr8z32oak.fsf>