F17 may corrupt user data when hibernating/resuming?

[cgrim]

Hi,

I saw the same thing on my system today. Yesterday I used hibernate on Lenovo T520 and today this has occurred for the first time:
kernel: [43990.079414] EXT4-fs error (device dm-2) in ext4_new_inode:941: IO failure

Than FS was in readonly state and run of fsck was required.

As I read comments for the bug it will be the best to wait for 3.4 kernel and than start using hibernation ... for now it's safer to power off/on :-(

[cunio]

Sorry to inform you but kernel 3.4 is no better.
I observe ext4 fs corruption every time I hibernate my Lenovo G550.

The smolt profile is here http://www.smolts.org/client/show/pu...e-014b47bd254d

[hostace]

AdamW, this is basic functionality of system. Drivers must work correctly. And without tweaks problem is persists. I discovered the problem with ACPI and HDD cache, but it's errors with hibernation algorythm. now I'm discovering kernel structure.

3.4 kernels will be more better, but I think, this is something with ext4 driver, or Intel ICH hardware. I'm not kernel programmer, but I,m reading now sources of kernel. In 3.4 kernels with hibernate fixed some intel grafics, fixed atl1c driver, etc. This is some cosmetical changes to the compressing image and etc. With acpi=off my filesystem isn't crashing now, but I have some quirks in my ACPI BIOS. I can't downgrade it for testing purposes at this point due to vendor restrictions.