eCryptfs: Use notify_change for truncating lower inodes
On 02/16/2012 08:08 AM, Colin Ian King wrote:
On 16/02/12 15:03, Tim Gardner wrote:
Colin - in your soak testing prior to the application of this patch,
were you able to detect corrupted or leaked inodes ? If so, can you
distill your tests and attach the reproducer to the bug report ?
That's a good question. As yet, I have not yet been able to figure out a
simple way to detect this and spot any theoretical corruption or leakage.
I've got this mammoth Emerald with lots of CPUs and spindles. Lemme run
some soak testing with the patch applied.
Tim Gardner firstname.lastname@example.org
kernel-team mailing list