Re: [BUG] Panic when systemd boot do mkdir on tmpfs mounted path with smack enabled environment

[Al Viro <viro@ZenIV.linux.org.uk>]

On Fri, May 27, 2016 at 12:01:05PM -0700, Linus Torvalds wrote:

> Al, if you want Casey to help test, I think you should write out the
> full git repository address, rather than just say "See
> vfs.git#smack-fix".
> 
> Anybody who isn't used to pulling for you will just wonder where you
> keep your tree. And even I, who _am_ used to pulling from you, would
> have to look it up, so it's a lot more convenient if you actually
> write out the whole thing,

Point taken.

> Casey, Al is talking about
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs smack-fix
> 
> and Al, please make your commit messages more informative than just
> "switch ->setxattr() to passing dentry and inode separately". You can
> see that from the patch. Please add a _why_ something is done, not
> just what it does.

Umm...  Would something along the lines of

switch ->setxattr() to passing inode and dentry separately

smack ->d_instantiate() uses ->setxattr(), so to be able to call it before
we'd hashed the new dentry and attached it to inode, we ->setxattr() instances
get the inode as an explicit argument rather than obtaining it from dentry.
Similar change for ->getxattr() had been done commit ce23e64.  Unlike
->getxattr() (which is used both by selinux and smack instances of
->d_instantiate()) ->setxattr() is used only by smack one and unfortunately
it had been missed back then.

be detailed enough for the second one with

switch xattr_handler->set() to passing inode and dentry separately

preparation for similar switch in ->setxattr() (see the next commit for
rationale)

for the first one?