On Monday 03 March 2003 22:33, Peter Milleson wrote: > Hi Rob and everyone, > > I was about to give up running the 2.4 kernel on my SS20 with dual > HyperSparc 150s but I downloaded the 2.4.21-pre5 from your site to try some > more testing. I can say that I wasn't expecting much success with all my > past failures so imagine my suprise when not only did it get to booting the > kernel on the first attempt, but it also passed init with no problems. I've > rebooted the machine a couple of times and can report only one hang on > init, which is great!
Same here! > But one question though, why does it appear that 2.4 is noticably slower > than 2.2? I've seen this on other sparcs when I tried using the 2.4 kernel, > but I was wondering does anybody know what is causing this? And why exactly > does /proc/cpuinfo show the HyperSparcs as 100s and not 150s? I'd like to > try and configure a test of the difference it takes to start apache with > 2.4 versus 2.2 and then the next test would to actually test the > performance of the system Didn't notice anything yet. But since I'm just using VI on the console I won't notice if they replace the cpu with a 40 MHz one :-) If there would be a performance hit, this would be a real shame imho... -- Frank Van Damme http://www.openstandaarden.be