On Sat, 3 May 2008 08:22:50 +0200, Jean Delvare wrote: > I think I have a old SuperMicro board with a ServerWorks OSB4, I'll do > some tests with it. I think it has only seen 2.4 kernels so far, so > HZ=100, where the initial delay was at least 10 ms. I'll test it at > HZ=1000 and see if I can reproduce the problem. If I can, then I guess > we are better increasing the delay for all the old ServerWorks chips > (OSB4, CSB5 and CSB6.) If I can't then maybe just increase the delay > for the CSB5?
I've tested my OSB4 for 2 hours continuously at HZ=1000 without your patch and it worked perfectly. So I suggest that we only increase the initial delay for the CSB5 for now - if that's what the reporter has. -- Jean Delvare _______________________________________________ i2c mailing list [email protected] http://lists.lm-sensors.org/mailman/listinfo/i2c
