I'm not entirely sure if the current implementation supports it or even if it could be seen as a real tail packing, but isn't it possible to simply format a partition with different block and fragment size? If so, the last few bytes of the file, if they do not fill more than one fragment, will be stored in a fragment rather than a block. If the hard disk uses 4KB blocks, it could be possible to create fragment sizes of 512/1024/2048/4096 bytes. This is actually in the design of the ext2, so if the implementation fits the original specification, it should be possible to use such feature right away. Hope this helps, EKS - Dave Poirier On Thu, Apr 04, 2002 at 11:29:59PM +0200, Erik Smit wrote: > On Thu, 2002-04-04 at 23:10, IT3 Stuart Blake Tener, USNR-R wrote: > > Mr. Smit: > > > > With regard to the tail packing, yes, I need it; I am using a > > laptop with limited space. Any ideas on whether it has become a stable > > option to use yet in EXT3? > > I personally do not think it would ever be included in EXT3 as one of > the main features of EXT3 is backwardscompatibility with EXT2 and I do > not think tailpacking would be backwardscompatible. But I'm not a > developer so I'm not sure about this. > > > With regard to turning off the journaling via tune2fs, I think > > there is a bug in tune2fs. It seems that it did not reverse whatever > > needed to be reversed in order for Partition Magic to think the > > filesystem was EXT2 again and not EXT3. Any further ideas? > > I would think this is a PQ problem because as far as I am aware when a > EXT3 filesystem is unmounted cleanly every EXT2-"only" application > should work smoothly with it. > > Perhaps resize2fs is a solution for you (should be included in > e2fsprogs)? It's not as point/click as PQ but should otherwise work just > nicely. Or maybe even parted > (http://www.gnu.org/software/parted/parted.html). I've never worked with > it myself but I've heard good things about it. > > Regards, > > Erik Smit > > PS: Why the excessively large signature? Your signature is almost just > as large as the message itself. :)
