I just finished installing Debian 6.0.2.1 to my new OCZ Vertex 3 SSD and built all of my GPT partitions and ext4 filesystems in such a way to ensure that they were properly aligned to both the 4KiB pages and the 512KiB erase blocks within the SSD. I also specified the discard mount option for my SSD filesystems to ensure that online block discard support was enabled, but it doesn't appear to actually be working.

I tried the test at http://sites.google.com/site/lightru...bledandworking and confirmed that after deleting the temporary file and sync'ing all filesystems, the block discard code doesn't wipe the LBA sector of the file - I have to use the wiper.sh script to do an online manual block discard to clean up the SSD. Has anyone experienced this problem too and been able to solve it?

I doubt that takes into consideration the batched discard support.

I was aware of that patch but I didn't really stop and consider the implications of it until I actually read the code. Now I think I understand why the LBAs weren't trimmed immediately.