On 05/24/2010 10:32 AM, Nick Piggin wrote:
I wonder, though. If we no longer trust block devices to give the correct data back, should we provide a meta block device to do error detection?
Some block devices do provide space for end-to-end checksums. For the ones that don't, I see no efficient way of adding it (either we turn one access into two, or we have a non-power-of-two block size).
No production filesystem on Linux has checksums (well, ext4 has a few). Of the ones that add checksumming, I'd say most will not do data checksumming (and for direct IO it is not done).
I believe btrfs checksums direct IO. Unfortunately it has some way to go before it can be used in production.
-- Do not meddle in the internals of kernels, for they are subtle and quick to panic. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxxx For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>