Re: [Q21.5] KKCC backtrace

Marcus> You're right, it should all either be conditioned or not, Marcus> otherwise it doesn't really make much sense. For speed and Marcus> low memory usage, I'd better condition all. What do you Marcus> propose? Unfortunately, DEBUG vs ERROR_CHECK has two semantics. 1) DEBUG gives tools to inspect the damage after it happens; ERROR_CHECK tries to catch it before, or just when, it happens. 2) DEBUG adds a memory burden but usually not speed. ERROR_CHECK may cost a lot of speed (eg, in Mule where it can add a factor of (buffer-size) to many operations). I would say since you've already got it conditioned on ERROR_CHECKING_GC you may as well just use that. -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Ask not how you can "do" free software business; ask what your business can "do for" free software.