Hi! > > I'd like to get some testing and comments on this. I know that > > userland part is messy and will not work on x86-64 etc, and I should > > obviously remove some extra printk's... but otherwise it should be okay. > > > > Testing would be nice, too, but be a bit careful. It should not be > > more dangerous than /sys/power/resume, but... If you suspect something > > unusual, be sure to force fsck. > > Could we please postpone it for some time? Well, it is probably not going into -mm before your changes are in mainline. That means 2.6.17 earliest. > First, the patch doesn't apply to the current -mm and the userland part is > incompatible with the code there to the extent of a serious breakage > (let alone the x86-64 issue). Of course I'm going to fix up breakage before the merge. At least all the kernel parts and enough userland to test that it still works. > Second, if I had some time to get my solution ready, we could compare the > alternatives instead of just discussing one of them. Well, I'm trying to gently push you. I'd like to get some solution in 1-2 months timeframe, so that work on userland parts can start. Pavel -- Thanks, Sharp!
