my laptop coming back from a hibernate function, 2 gigs of ram, takes about twice as long as the initial boot. i dont use hibernate anymore, i just shutdown and then restart.
and, nothing ever works right after coming back from hibernate. Again, ive not used newer macs, but both my classic II and my power mac had the same issue. I never used hibernate, as it just didn't work right. On Wed, Dec 31, 2008 at 12:36 PM, Jed Rothwell <[email protected]> wrote: > Terry Blanton wrote: > >> >> My point is that nothing happens between system clock pulses. The state >> of a computer remains unchanged. With fast and cheap NV memory, you could >> build a PC that could literally suspend its state between processor pulses >> and instantly restore the PC to the state before suspension. > > I do not see how this would be any faster than a conventional battery-backed > up computer in a wait state. What I mean is, we can accomplish the same > thing today. > > It would draw no power and be immune to power outages, of course. For some > applications that would be a tremendous advantage. Especially dedicated > control and data collection devices. But a regular PC "wakes up" from > hibernation and reloads 4 GB of RAM quickly. It is not a bit annoying. > > >> >> Many believe that the memristor could provide such an operation. BIOS, OS, >> video, processor cache, all memories would be NV and totally freezable such >> that no bootstrapping would ever occur (except with MicroSoft products >> which would have to be restarted after the blue screen of death) :-) > > Even without a complete blue-screen of death crash, and even with other > operating systems, garbage collection is imperfect and programs fail partly, > so from time to time you have to completely reboot. Especially programs such > as voice input, Acrobat reader and anything to do with graphics or video. > > A process control computer, such as the one driving a Prius or a Boeing 747, > must have an emergency reboot and reload RAM procedure. It has to happen in > a fraction of a second. So it cannot load from a disk, obviously. As far as > I know, it goes back to Byte 0 of a ROM chip and starts from scratch, and > whatever is in RAM is g-o-n-e. > > I suppose the hardest part for the programmer is to determine that the > program has crashed and it is time to issue an emergency interrupt and > reset. I recall that Data General computers used in critical apps used to > have an auxiliary microcomputer checking the main computer, standing by > ready to goose the main computer when it did not respond. The aux computer > spend all day asking: "You okay? Still okay? Still there? Still okay? . . ." > No response -- bam! If they both go out to lunch what do you do then? I > guess the main one can check the aux one from time to time. Not likely they > will both go out at the same moment. > > Years ago, Toyota issued a recall for Prius computers that were locking up > and crashing while the vehicle was underway at high speed. That's scary! > > - Jed > >
