On Wed, Aug 09, 2006 at 11:39:14PM -0700, Andrew Morton wrote: > One the fs has been copied-and-pasted, it's much harder to address these > things: either need to do it twice, or allow the filesystems to diverge, or > not do it. > Also, JBD is presently feeding into submit_bh() buffer_heads which span two > machine pages, and some device drivers spit the dummy. It'd be better to > fix that once, rather than twice.. Well, it's harder no matter when we do it, right? Whether we do it before we submit or after, we still before us have the choice of whether to let the filesystems diverge (and make it harder to do maintenance), or not. But Linus and others have spoken pretty clearly that they don't want ext3 to be touched, and it's not even clear that people would be happy with cleanups. If we are going to do the cleanups in both, it would actually be better to fix up ext3 *first*, and then do the copy... - Ted - To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
