Messages in this thread | | | Date | Wed, 09 Aug 2006 03:18:39 -0600 | From | Hans Reiser <> | Subject | Re: partial reiser4 review comments |
| |
Christoph Hellwig wrote:
> I must admit that standalone code snipplet doesn't really tell me a lot. > >Do you mean the possibility to pass around a filesystem-defined structure >to multiple allocator calls? I'm pretty sure can add that, I though it >would be useful multiple times in the past but always found ways around >it. > > > Assuming I understand your discussion, I see two ways to go, one is to pass around fs specific state and continue to call into the FS many times, and the other is to instead provide the fs with helper functions that accomplish readahead calculation, page allocation, etc., and let the FS keep its state naturally without having to preserve it in some fs defined structure. The second approach would be cleaner code design, that would also ease cross-os porting of filesystems, in my view.
As a philosophy of design issue, the current VFS and generic code leaves me with the feeling that people are trying to write the FS rather than trying to help the FS writer with some useful library functions. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |