On Thu, 12 Mar 1998, Matt Vail wrote: > Last question, could these problems be interrelated? The missed > pulses happen much less often than the latter problem. Any reponse > would be greatly appreciated. My suggestion to this problem is to take a look at possibly using a profiler. Most likely if you use gcc you can compile the mud using the -pg flag. (i.e. in the MYFLAGS= section of the makefile add -pg) then let the mud run for a while (maybe a day or so) and then restart it, and run gprof or your other preferred profiler on the output file (usually gmon.out). I'm not 100% sure of the syntax (I haven't used it in quite a while) but it does produce useful information such as how long in time/cpu cycles it spends in each function. Its only drawback is that it doesn't do C++ really well, atleast the version I had didn't. Good Luck, -- Steve quickey@genocide.net smoret@uci.edu +------------------------------------------------------------+ | Ensure that you have read the CircleMUD Mailing List FAQ: | | http://democracy.queensu.ca/~fletcher/Circle/list-faq.html | +------------------------------------------------------------+
This archive was generated by hypermail 2b30 : 12/15/00 PST