> function while Windows NT uses the TrgLookupFileAllocation function. MS
> actually stores the file allocations in the VM Cache, and then transacts
> them by completely bypassing the file system driver. In essence, the file
yes, if i understand you corrently then maybe you have missed this detail,
the new page cache does something like this as well: we store all
logical->on_disk_physical file data mappings in the page cache, this means
that if we revisit a file then we do not have to go into the lowlevel
filesystem code anymore like previous kernels. Relevant filesystem
metadata can even be freed 'under' the inode (this will eventually happen
if memory pressure increases), still file contents will be cached, the
file can be read and written, and asynchron 'store to disk' IO will still
happen, completely parallel, and without involving _any_ filesystem
metadata at all.
filesystems actively have to take advantage of this functionality, but
it's rather straightforward and the caching is automatic.
> What would really make this easy to understand from a Linux standpoint were
> if I open sourced the Windows NT portions of FENRIS that were removed so you
> could see the differences between FENRIS NT and FENRIS Linux. I will ask MS
> if they have no objections (MS worked with us on this, and they have some
> ownership rights of the Windows NT portions of FENRIS not released). I do
> not think they would object, but I will ask. I think FENRIS is probably
> oneof the first file systems that has been single-sourced between NT and
> Linux.
nice to hear this. Please keep us updated.
-- mingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/