On Fri, Jun 24, 2011 at 12:02:05PM -0400, Thor Lancelot Simon wrote: > On Fri, Jun 24, 2011 at 07:50:35AM +0000, Michael van Elst wrote: > > > > Similar to raw devices in performance, I/O caching, and without the > > hardware constraints regarding block sizes or alignment. I.e. something > > that lets you treat disk contents like a regular file. > > At least for NetBSD, that's never been true. The most glaring problem > is that there's no protection against causing the same underlying disk > blocks to be multiply cached by accessing the buffer cache with a different > stride.
Well, the code assumes that disklabel (used by the block device) and filesystem agree on block size. But there are probably even bugs with that and without disklabel there is somewhere a 2kB default. -- Michael van Elst Internet: mlel...@serpens.de "A potential Snark may lurk in every tree."