> From: Greg Smith [mailto:rys@;epaibm.rtpnc.epa.gov] > Ross Patterson wrote: > > s390 port had been successful, we wouldn't have this problem. Linas' > port > > of GCC for "Bigfoot" had the stack growing *upward*, not *downward* as > on > > almost every other platform. > > I've always been curious. Why is a top down stack used anyways ?? > Of course I've been using a bottom up stack for almost 30 yrs so I > might be biased ;-)
If the heap grows up, and the stack grows down, then one can have, in theory, arbitrarily large stacks. Handy for CPU's that have a single flat memory space that is not very big (e.g. a few MBytes), as otherwise one is forced to limit the size of the stack. But this is a 4 decades old theory, and clearly there are other options these days, and other considerations to take into account. Its time the stack-growth-direction bug got fixed; the architectural limitations that caused it to grow down are now gone, and the stack-overrun attacks that it engenders are a great threat to computer security. --linas p.s. while I'm on this topic: I would like to call to action, again, the need for a fundamental shift in the way UNIX handles shared libraries, in order to make room for, and leverage, the way that the 390 cpu architecture is able to protect different memory spaces. If implemented, the 390-style memory architecture can offer a radical rethinking of how client/server systems are implemented, and how security and trust is done. For this to catch on in the mainstream, other CPU architectures would need to add similar features as well. But given the recent burbling from microsoft and intel about palladium and how cpu arch changes can enhance security, (which intel seems to be actually working on) I do not think that it is too wild, too early or too impractical to engage in this task. -- pub 1024D/01045933 2001-02-01 Linas Vepstas (Labas!) <[EMAIL PROTECTED]> PGP Key fingerprint = 8305 2521 6000 0B5E 8984 3F54 64A9 9A82 0104 5933
msg09195/pgp00000.pgp
Description: PGP signature
