Re: New Software Suspend Patch for testing.

[Alan Cox]

On Sad, 2003-04-05 at 00:03, Pavel Machek wrote:
> > Yes. Under 2.4, some people found that if the writeback cache isn't
> > flushed before we powerdown, the image isn't completely saved. This was
> > the fix (the first test is the original one, the second is based upon
> > hdparm's tests for writeback cache).
> 
> But this looks like you are fixing generic bug, which could make
> kernel do something very wrong even without swsusp, right? If so,
> submit it to Alan, ASAP.

I'll grab the older ATA spec and take a look. The other changes are
broken, but this one looks quite beliavable