Return to BSD News archive
Newsgroups: comp.os.386bsd.bugs Path: sserve!newshost.anu.edu.au!munnari.oz.au!network.ucsd.edu!usc!zaphod.mps.ohio-state.edu!uwm.edu!caen!saimiri.primate.wisc.edu!usenet.coe.montana.edu!nate From: nate@cs.montana.edu (Nate Williams) Subject: Re: Backspace problem, xterm scroll bar problem. Message-ID: <1993May19.234613.19017@coe.montana.edu> Keywords: Backspace problem, xterm scroll bar problem. Sender: usenet@coe.montana.edu (USENET News System) Organization: CS References: <C76sxB.3Lq@ra.nrl.navy.mil> <1t9k8aINNo9h@srvr1.engin.umich.edu> <1993May18.232043.18617@coe.montana.edu> <C7A9nB.KDA@hip-hop.suvl.ca.us> Date: Wed, 19 May 1993 23:46:13 GMT Lines: 28 In article <C7A9nB.KDA@hip-hop.suvl.ca.us> dfox@hip-hop.suvl.ca.us (David Fox) writes: >nate@cs.montana.edu (Nate Williams) wrote: >>In article <1t9k8aINNo9h@srvr1.engin.umich.edu> scottl@engin.umich.edu (scott allen long) writes: >>> >>>You probably have an FPU-less system (386 or 486SX). XFree 1.2 was compiled >>>with gcc 2.3.3 with the assumption that an FPU would be present. As a >> >>The problem is in the math emulation code, not in assuming you need >>a FP processor. >> > >gcc 2.3.3 does FP differently - and BTW, a lot more efficiently - than >does gcc 1.39. The basic problem is that gcc is able to generate >some coprocessor opcodes (such as fsin) directly, and these are not >supported in the emulator. If you have a coprocessor, performance can >go way up: but if you don't, you're stuck. I don't know about GCC2 doing things better, but even I have had more problems with the GCC2 compiled binaries than I have had with the GCC1 binaries on my 486 (real one with a FPU even :) Nate -- osynw@terra.oscs.montana.edu | Still trying to find a good reason for nate@cs.montana.edu | these 'computer' things. Personally, work #: (406) 994-4836 | I don't think they'll catch on - home #: (406) 586-0579 | Don Hammerstrom