linux-kernel - Re: [GIT PULL] notification tree - fsnotify assumes incorrectly positive parent dentry

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <alpine.LFD.2.01.0906291801310.3605@localhost.localdomain>
Date:	Mon, 29 Jun 2009 18:02:09 -0700 (PDT)
From:	Linus Torvalds <torvalds@...ux-foundation.org>
To:	Jeremy Kerr <jk@...abs.org>
cc:	Eric Paris <eparis@...hat.com>, Al Viro <viro@...iv.linux.org.uk>,
	Stephen Rothwell <sfr@...b.auug.org.au>,
	Benjamin Herrenschmidt <benh@...nel.crashing.org>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [GIT PULL] notification tree - fsnotify assumes incorrectly
 positive parent dentry



On Tue, 30 Jun 2009, Jeremy Kerr wrote:
> 
> > Hmm. It does sound like a reasonable assumption, though. Maybe spufs
> > should be fixed to have an inode for all directories?
> 
> We have inodes for all directories, it's just the order which we set 
> things up. When a new 'spu context' (ie, a directory with a bunch of 
> files) is created, we add the parent dentry, populate it with files 
> (positive dentries), then instantiate the parent.
> 
> There's no specific need to do it in this order, it just makes the code 
> a little simpler - we just 'stitch the parent in' once everything else 
> has completed successfully, so less stuff to do in the error path.
> 
> > A NULL d_inode means that something is a negative dentry, and a
> > negative dentry shouldn't have children.
> 
> OK. If this is a general invariant, then I'll get a change to spufs 
> going to do the setup in the right order.

Hmm. I doubt it matters a ton, but if it's easy to change it so that it 
populates the parent inode first, I think that would be a good thing. 
Thanks,

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