Re: nfs invalidates lose pte dirty bits

From: Trond Myklebust
Date: Thu Dec 22 2005 - 18:29:53 EST


On Thu, 2005-12-22 at 18:55 +0100, Andrea Arcangeli wrote:
> Hello Andrew,
>
> this is yet another problem exposed by the new invalidate_inode_pages2
> semantics.
>
> If a mapping has dirty data (pte dirty and page clean), and an
> invalidate triggers, the filemap_write_and_wait will do nothing, but the
> invalidate_inode_pages2 will destroy the pte dirty bit and discard the
> pagecache.

See the latest git release where we introduce the nfs_sync_mapping()
helper.

Cheers,
Trond

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/