On Tue, Jul 07, 2009 at 11:07:58AM -0400, Christoph Hellwig wrote: > On Tue, Jul 07, 2009 at 05:02:57PM +0200, Nick Piggin wrote: > > That's kind of why I liked it in inode_setattr better. > > > > But if the filesystem defines its own ->setattr, then it could simply > > not define a ->setsize and do the right thing in setattr. So this > > calling convention seems not too bad. > > Or the filesystem could just call into it's own setattr method > internally. For that we'd switch back to passing the iattr to > ->setsize. For a filesystem that doesn't do anything special for > ATTR_SIZE ->setsize could point to the same function as ->setattr. > > For filesystem where's it's really different they could be separate or > share helpers. OK, so what do you suggest? If the filesystem defines ->setsize then do not pass ATTR_SIZE changes into setattr? But then do you also not pass in ATTR_TIME cchanges to setattr iff they are together with ATTR_SIZE change? It sees also like quite a difficult calling convention. -- 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
