> >:I would like to see "the PIIX problem" caught on camera, personally. > >:We're aware of one errata for it already, and we work around it. If > >:there's another problem, or ideally if someone has some relatively quick > >:code to test it, that would be much better. > > > > Holy shit. We are screwed. It's a free-running counter with NO > > synchronization whatsoever. None. Zip. Zero. > > Yes, there is an errata for just that on early chipsets. > > Does the ..._slow patch I sent work for you ?
Matt's problem (look-ahead carry) will break the three-read algorithm because it can generate a sequence of three reads that appear to be in succession, but which are all wrong. We need three different algorithms; "works", "ripple" and "look-ahead". Of those, "works" should be based exclusively off a list of known-good chipsets, "look-ahead" seems to be easily enough detected (but we should probably have a blacklist anyway) and "ripple" is hard to detect and should be the default case. I really, really hate hardware. -- To announce that there must be no criticism of the president, or that we are to stand by the president, right or wrong, is not only unpatriotic and servile, but is morally treasonable to the American public. - Theodore Roosevelt To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
