* [dm-crypt] How will ATA-TRIM be handled?
@ 2009-10-03 23:23 Christian Tschabuschnig
2009-10-04 6:32 ` Heinz Diehl
0 siblings, 1 reply; 2+ messages in thread
From: Christian Tschabuschnig @ 2009-10-03 23:23 UTC (permalink / raw)
To: dm-crypt
Hello,
I don't know much about the internals of filesystems or devicemapper but
I know that the kernel supports ATA-TRIM and that ext4 supports or will
soon support it.
My question is: What happens, if a dm-crypt-layer is between the
filesystem and the real block-device? Will the trim be passed through,
or will it be discarded?
Optimally, i think, both should be possible. Passing through the
information will improve performance and lifetime of a SSD and
discarding it would give less information to an attacker since the SSD
will return zeroes on LBAs that are not mapped.
Regards,
Christian
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [dm-crypt] How will ATA-TRIM be handled?
2009-10-03 23:23 [dm-crypt] How will ATA-TRIM be handled? Christian Tschabuschnig
@ 2009-10-04 6:32 ` Heinz Diehl
0 siblings, 0 replies; 2+ messages in thread
From: Heinz Diehl @ 2009-10-04 6:32 UTC (permalink / raw)
To: dm-crypt
On 04.10.2009, Christian Tschabuschnig wrote:
> My question is: What happens, if a dm-crypt-layer is between the
> filesystem and the real block-device? Will the trim be passed through,
> or will it be discarded?
It guess it will be passed through. I'm using btrfs on a SSD which is
encrypted with LUKS/dmcrypt, and it works great. Btrfs uses ata trim quite
a while.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2009-10-04 6:32 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-10-03 23:23 [dm-crypt] How will ATA-TRIM be handled? Christian Tschabuschnig
2009-10-04 6:32 ` Heinz Diehl
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.