Re: kobject_init_and_add is easy to misuse

[Matthew Wilcox]

On Tue, Jun 02, 2020 at 04:04:04PM +0200, Greg Kroah-Hartman wrote:
> On Tue, Jun 02, 2020 at 05:10:35AM -0700, Matthew Wilcox wrote:
> > On Tue, Jun 02, 2020 at 07:50:33PM +0800, Wang Hai wrote:
> > > syzkaller reports for memory leak when kobject_init_and_add()
> > > returns an error in the function sysfs_slab_add() [1]
> > > 
> > > When this happened, the function kobject_put() is not called for the
> > > corresponding kobject, which potentially leads to memory leak.
> > > 
> > > This patch fixes the issue by calling kobject_put() even if
> > > kobject_init_and_add() fails.
> > 
> > I think this speaks to a deeper problem with kobject_init_and_add()
> > -- the need to call kobject_put() if it fails is not readily apparent
> > to most users.  This same bug appears in the first three users of
> > kobject_init_and_add() that I checked --
> > arch/ia64/kernel/topology.c
> > drivers/firmware/dmi-sysfs.c
> > drivers/firmware/efi/esrt.c
> > drivers/scsi/iscsi_boot_sysfs.c
> > 
> > Some do get it right --
> > arch/powerpc/kernel/cacheinfo.c
> > drivers/gpu/drm/ttm/ttm_bo.c
> > drivers/gpu/drm/ttm/ttm_memory.c
> > drivers/infiniband/hw/mlx4/sysfs.c
> 
> Why are random individual drivers calling kobject* functions?  That
> speaks to a larger problem here...

There's around 120 callers in the kernel today ... large, indeed.

> Anyway, yes, it's a tricky function, but the issue usually is that the
> kobject is embedded in something else and if you call init_and_add() you
> want to tear things down _before_ the final put happens.
> 
> The good thing is, that function is really hard to get to fail except if
> you abuse it with syzkaller :)

Yes ;-)

> > I'd argue that the current behaviour is wrong, that kobject_init_and_add()
> > should call kobject_put() if the add fails.  This would need a tree-wide
> > audit.  But somebody needs to do that anyway because based on my random
> > sampling, half of the users currently get it wrong.
> 
> As said above, this is "tricky", and might break things.

My audit may not be correct then.  The kobject_put() may be appropriately
being called at a higher level rather than in the same function as the
kobject_init_and_add().