Messages in this thread | | | Subject | Re: [PATCH 3/6] vfs: add __d_instantiate() helper | From | Trond Myklebust <> | Date | Wed, 15 Oct 2008 16:47:47 -0400 |
| |
On Thu, 2008-10-16 at 05:39 +0900, OGAWA Hirofumi wrote: > Christoph Hellwig <hch@infradead.org> writes: > > > On Wed, Oct 15, 2008 at 10:58:10PM +0900, OGAWA Hirofumi wrote: > >> > >> This adds __d_instantiate() for users which is already taking > >> dcache_lock, and replace with it. > >> > >> The part of d_add_ci() isn't equivalent. But it should be needed > >> fsnotify_d_instantiate() actually, because the path is to add the > >> inode to negative dentry. fsnotify_d_instantiate() should be called > >> after change from negative to positive. > >> > >> __d_instantiate_unique() and d_materialise_unique() does opencoded > >> optimized version. From history, it seems a intent, so just add comment. > > > > Looks good, but I think those "optimized' version should also be > > converted, a single if shouldn't matter with todays branch predictors. > > Me too. > > Trond, do you care the following convert? E.g. > > in d_materialise_unique(): > > from > dentry->d_inode = NULL; > goto found_lock; > to > __d_instantiate(dentry, NULL); > goto found_lock;
That would be fine by me.
Cheers Trond
-- Trond Myklebust Linux NFS client maintainer
NetApp Trond.Myklebust@netapp.com www.netapp.com
| |