Re: [PATCH] iget_locked [6/6]

[Alexander Viro]

On Fri, 10 May 2002, Jan Harkes wrote:

> 
> As of the previous patch the inode_hashtable doesn't really need to be
> indexed by i_ino anymore, the only reason we still have to keep the
> hashvalue and i_ino identical is because of insert_inode_hash.
> 
> Here we simply add an argument to insert_inode_hash. If at some point a
> FS specific getattr method is implemented it will be possible to
> completely remove all uses of i_ino in the VFS.

How about

static inline void insert_inode_hash(struct inode *inode)
{
	__insert_inode_hash(inode, inode->i_hash);
}

in fs.h and switching those who want something different to direct use
of __insert_inode_hash()?

It's going to remain a very common case and IMO it makes a lot of sense
to keep a simple helper for it.  That has a nice property of getting
patch way smaller ;-)