Path: utzoo!dciem!array!colin From: colin@array.UUCP (Colin Plumb) Newsgroups: comp.arch Subject: Re: What *should* architectural pointers point at? Message-ID: <643@array.UUCP> Date: 31 Aug 90 15:26:38 GMT References: <0887@sheol.UUCP> <41167@mips.mips.COM> <3318@awdprime.UUCP> Organization: Array Systems Computing, Inc., Toronto, Ontario, CANADA Lines: 15 In article <3318@awdprime.UUCP> tif@reed.UUCP (Paul Chamberlain/32767) writes: > Now there's an interesting thought. Why not order these 64 bits so that > the 3 on the left are the bit offset. Mere mortals would use it just like > a byte addressed machine but wizards could use those 3 bits anyway they like. Because it's unnecessary. Where is some code that depends on the fact that incrementing a character pointer adds 1 to the value of its bit pattern, if read as an integer? I'm sure there is some, buried somewhere, but for the most part, the only code that does such pointer diddling in in C and C hides the representation of pointers sufficiently, unless you really try. If you're extra paranoid, define the cast from a pointer to a long and back to do a shift by 3 bits; then you only have to worry about unions. -- -Colin