Re: [PATCH 2/4] carl9170: fix debugfs crashes

[Christian Lamparter]

On Sunday, September 18, 2016 6:44:08 PM CEST Greg KH wrote:
> On Sun, Sep 18, 2016 at 02:49:33PM +0200, Christian Lamparter wrote:
> > On Sunday, September 18, 2016 12:14:55 PM CEST Greg KH wrote:
> > > On Sun, Sep 18, 2016 at 10:54:18AM +0300, Kalle Valo wrote:
> > > > Greg KH <gregkh@xxxxxxxxxxxxxxxxxxx> writes:
> > > > 
> > > > > On Sat, Sep 17, 2016 at 09:43:02PM +0200, Christian Lamparter wrote:
> > > > >> Ben Greear reported:
> > > > >> > I see lots of instability as soon as I load up the carl9710 NIC.
> > > > >> > My application is going to be poking at it's debugfs files...
> > > > >> >
> > > > >> > BUG: KASAN: slab-out-of-bounds in carl9170_debugfs_read+0xd5/0x2a0
> > > > >> > [carl9170] at addr ffff8801bc1208b0
> > > > >> > Read of size 8 by task btserver/5888
> > > > >> > =======================================================================
> > > > >> > BUG kmalloc-256 (Tainted: G        W      ): kasan: bad access detected
> > > > >> > -----------------------------------------------------------------------
> > > > >> >
> > > > >> > INFO: Allocated in seq_open+0x50/0x100 age=2690 cpu=2 pid=772
> > > > >> >...
> > > > >> 
> > > > >> This breakage was caused by the introduction of intermediate
> > > > >> fops in debugfs by commit 9fd4dcece43a
> > > > >> ("debugfs: prevent access to possibly dead file_operations at file open")
> > > > >
> > > > > Because of this, these should all be backported to 4.7-stable, and
> > > > > 4.8-stable, right?
> > Ok, only b43legacy has debugfs enabled by default. For b43 and carl9170 
> > debugfs support is usually disabled.
> > 
> > Greg, would you take these four patches "as is" for -stable
> > or do you want a "minimal version" which just replaces the
> > 
> > dfops = container_of(file->f_op, ...
> > 
> > with
> > 
> > dfops = container_of(file->f_path.dentry->d_fsdata, ...
> > 
> > in the three drivers for -stable?
> 
> No, I'll take this as is, we want things to remain as close as possible
> to Linus's tree.  When we are not, is when things break.
> 
> > > > Via which tree should these go, Greg's or mine?
> > > 
> > > I'll take it if you ack it, as it's a debugfs issue.
> > For carl9170: Ben Greear has reported:
> > "I have verified this fixes my problem in the 4.7 kernel."
> > 
> > But this was with a preliminary/minimal version so I didn't
> > add the tested-by tag.
> > 
> > As for b43, I'll see if I have a working b43 in my collection
> > somewhere to confirm the issue and the fix. Question is, do
> > you want to wait or not?
> 
> I'll queue these up this week, no rush.

I was able to sucessfully test the b43 patch on my iBook G4's BCM4306.

Thanks,

Christian