Re: performance of FreeBSD-current as SMP

2000-01-08 Thread Garrett Wollman
< said: > Well, some difference is to be expected. But 22% Flaws in the Intel SMP design make it more expensive than it ought to be. Things like atomic RMW cycles are very, very expensive. (We had a discussion about six months ago about how expensive something like a locked xchg instruct

Re: performance of FreeBSD-current as SMP

2000-01-08 Thread Mohit Aron
> > Yes it should. SMP support enables inter-processor locking code which > does not exist in non-SMP kernels. Ergo, non-SMP kernels run > uniprocessor tasks faster. > Well, some difference is to be expected. But 22% - Mohit To Unsubscribe: send mail to [EMAIL PROTECTED] with "un

Re: performance of FreeBSD-current as SMP

2000-01-08 Thread Garrett Wollman
[Not network-related; moved to -current.] < said: > Good Lord! This is the second time now. I even SAID in my last two mails that > there is only ONE processor. Theortically then, FreeBSD configured with/without > SMP support shouldn't make any difference. Yes it should. SMP support enables in