Path: chuka.playstation.co.uk!tjs From: tjs@cs.monash.edu.au (Toby Sargeant) Newsgroups: scee.yaroze.freetalk.english,scea.yaroze.freetalk Subject: Debugger performance [Was: Re: TestCard() issue] Date: 18 Sep 1998 04:21:37 GMT Organization: PlayStation Net Yaroze (SCEE) Lines: 17 Message-ID: References: <35FC10F2.3841CEE6@bigfoot.com> <35FD49AC.7FA76186@augsburg.baynet.de> <3600CE30.98A32F71@easynet.co.uk> NNTP-Posting-Host: longford.cs.monash.edu.au X-Newsreader: slrn (0.9.4.3 UNIX) Xref: chuka.playstation.co.uk scee.yaroze.freetalk.english:2375 scea.yaroze.freetalk:954 On Thu, 17 Sep 1998 22:09:59 -0500, Sean Kennedy wrote: >HSS.EXE is used in conjunction with GDB {Gnu DeBugger} outside of SIOCONS. >It is a SECOND application that runs real-time above your code, and it had >severe performance penalties. All debuggers have this really, it only is >good for single stepping through problematic portions of your code. Interesting question. As I see it, the only way that a debugger stub interferes with the running of code is by modifying exception vectors. Thus, only those things that cause exceptions (or interrupts in the case of the R3000) will be affected. Additional latency in a system which already has quite a large latency shouldn't be all that noticeable, especially if the per-frame interrupt count is reasonably low, which I would expect. I'd be interested to hear other views though. I haven't written a debugger stub before, so I'm mostly going on gut feeling. Toby.