>
>
> On Sat, 22 Sep 2001, David Chow wrote:
>
> > Dear all,
> >
> > I have one question from using vfs_symlink, is it usual after I called
> > vfs_symlink that will return a dentry with dentry->d_inode == NULL???
>
> vfs_symlink() returns an integer, not dentry.
>
> > The link was sucessfully created but I receive a null inode pointer.
> > When creating a symlink it should also create an inode. But according to
> > the documentation about VFS from Richard, the symlink call of
> > inode_operations, should self call d_instantiate() this also means it
> > should automatically create an inode pointer. This shouldn't be done by
> > the caller???? Any hints? the vfs_create works fine and return with a
> > proper inode number, why vfs_symlink doesn't? Thanks.
>
> No, it doesn't. vfs_create() returns 0 in case of success and small negative
> number in case of error. Neither has any relation to inode numbers.
>
> dentry you've passed to it is modified - it gets non-NULL ->d_inode _if_
> operation succeeded. Which is not guaranteed - depends on kind of filesystem,
> length of symlink body, permissions, etc.
>
> So unless you tell what you are passing to vfs_symlink(), what does it
> return, etc. - there's nothing anyone could do. We might be smart, by
> I'm not aware of anybody here being telepathic.
>
I shouldn't say return, it is the dentry that I pass to vfs_symlink() and
vfs_create(). The dentry passed to vfs_create did pass back a proper
dentry->d_inode. For the vfs_symlink() it didn't?? Even the link was
sucessfully created on the filesystem. The dentry->d_inode is NULL. That's
the story... thanks.
regards,
David Chow
-
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/