Hi! > > Well, for example disk driver certainly needs interrupt controller and > > timer devices to be working... Disk itself does not care about timer > > devices, but driver certainly needs them. > > > > And it is this kind of dependencies that would make it "interesting". > > I'm not disputing those dependencies, but there seems to be something > wrong. If we're writing the image with interrupts enabled, we've either > not suspended the system devices (which means we're not saving their > state), or we've re-enabled interrupts (which may cause some bad things if > some system devices are disabled). If we're writing with interrupts > disabled, then the timers and IRQ seem to be irrelevant. What am I > missing? Currently, we always freeze/unfreeze the whole tree. (We write with interrupts enabled). We do not see any of these issues. Someone suggested that we should only unfreeze swap device. That does not work, see above. > > > Hrm. Maybe we should do: > > > > > > - Stop > > > - Save > > > - Restart > > > > > > On each device as we walk the tree. That way we would never have to worry > > > about the swap device being stopped before we allocated memory for all > > > devices. We would also have the ability to separate that action from > > > actually powering the device down, too. > > > > Eh? If you have Save after Stop, how are you going to allocate memory > > for save? > > How can you Save before you Stop? For each device, it needs to be in a > consistent and quiescent state before you save its state. For example ipw2100 would like to have pre-stop phase, where it could read firmware from filesystem and store it to RAM. I thought your "Save" refers to this one. Pavel -- People were complaining that M$ turns users into beta-testers... ...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!
