linux-kernel - Re: [GIT PULL] f2fs update for 6.8-rc1

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

Message-ID: <ZaF0p0nEOeW48H2l@google.com>
Date: Fri, 12 Jan 2024 09:19:35 -0800
From: Jaegeuk Kim <jaegeuk@...nel.org>
To: Al Viro <viro@...iv.linux.org.uk>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Linux F2FS Dev Mailing List <linux-f2fs-devel@...ts.sourceforge.net>
Subject: Re: [GIT PULL] f2fs update for 6.8-rc1

Posted this.
https://lore.kernel.org/lkml/20240112171645.3929428-1-jaegeuk@kernel.org/T/#u

On 01/12, Jaegeuk Kim wrote:
> On 01/12, Al Viro wrote:
> > On Thu, Jan 11, 2024 at 09:05:51PM -0800, Linus Torvalds wrote:
> > > On Thu, 11 Jan 2024 at 10:28, Jaegeuk Kim <jaegeuk@...nel.org> wrote:
> > > >
> > > >   git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git tags/f2fs-for-6.8-rc1
> > > 
> > > Hmm. I got a somewhat confusing conflict in f2fs_rename().
> > > 
> > > And honestly, I really don't know what the right resolution is. What I
> > > ended up with was this:
> > > 
> > >         if (old_is_dir) {
> > >                 if (old_dir_entry)
> > >                         f2fs_set_link(old_inode, old_dir_entry,
> > >                                                 old_dir_page, new_dir);
> > >                 else
> > >                         f2fs_put_page(old_dir_page, 0);
> > 
> > Where would you end up with old_dir_page != NULL and old_dir_entry == NULL?
> > old_dir_page is initialized to NULL and the only place where it's altered
> > is
> >                 old_dir_entry = f2fs_parent_dir(old_inode, &old_dir_page);
> > Which is immediately followed by
> >                 if (!old_dir_entry) {
> >                         if (IS_ERR(old_dir_page))
> >                                 err = PTR_ERR(old_dir_page);
> >                         goto out_old;
> >                 }
> > so we are *not* going to end up at that if (old_is_dir) in that case.
> 
> It seems [1] changed the condition of getting old_dir_page reference as below,
> which made f2fs_put_page(old_dir_page, 0) voided.
> 
> -       if (S_ISDIR(old_inode->i_mode)) {
> +       if (old_is_dir && old_dir != new_dir) {
>                 old_dir_entry = f2fs_parent_dir(old_inode, &old_dir_page);
>                 if (!old_dir_entry) {
>                         if (IS_ERR(old_dir_page))
> 
> [1] 7deee77b993a ("f2fs: Avoid reading renamed directory if parent does not change")
> 
> > 
> > Original would have been more clear as
> > 	if (old_is_dir) {
> > 		if (old_dir != new_dir) {
> > 			/* we have .. in old_dir_page/old_dir_entry */
> > 			if (!whiteout)
> > 	                        f2fs_set_link(old_inode, old_dir_entry,
> >                                                 old_dir_page, new_dir);
> > 			else
> > 	                        f2fs_put_page(old_dir_page, 0);
> > 		}
> >                 f2fs_i_links_write(old_dir, false);
> > 	}
> > - it is equivalent to what that code used to do.  And "don't update ..
> > if we are leaving a whiteout behind" was teh bug fixed by commit
> > in f2fs tree...
> > 
> > The bottom line: your variant is not broken, but only because
> > f2fs_put_page() starts with
> > static inline void f2fs_put_page(struct page *page, int unlock)
> > {
> >         if (!page)
> >                 return;
> > 
> > IOW, you are doing f2fs_put_page(NULL, 0), which is an explicit no-op.