RFC: Pass TRIM through GELI

Matthew D. Fuller fullermd at over-yonder.net
Wed Jul 15 02:17:17 UTC 2015


On Tue, Jul 14, 2015 at 05:48:13PM -0400 I heard the voice of
kpneal at pobox.com, and lo! it spake thus:
> 
> Who says that a write to a logical block in an SSD will result in a
> write to the same physical block? I thought it would typically or at
> least frequently go to a different physical block. Wear levelling
> and all that.

Yes, it wouldn't be useful on a SSD (or a zvol, or any other backing
that acts in a COW-ish way).  And it's meaningless combining TRIM and
shredding on a traditional platter drive, since it wouldn't do the
TRIM part.  Still, there can be some situations that would be both
directly overwriting, and supporting unused-space-notifications; some
virtualized SAN setups for instance, or or a md/VM/etc block device
that supported sparse usage.


-- 
Matthew Fuller     (MF4839)   |  fullermd at over-yonder.net
Systems/Network Administrator |  http://www.over-yonder.net/~fullermd/
           On the Internet, nobody can hear you scream.


More information about the freebsd-geom mailing list