Re: [PATCH] mm/slab: make __free(kfree) accept error pointers

[Vlastimil Babka]

On 4/30/24 2:50 PM, Dan Carpenter wrote:
> On Tue, Apr 30, 2024 at 02:09:10PM +0200, Vlastimil Babka wrote:
>> On 4/29/24 5:03 AM, Matthew Wilcox wrote:
>> > On Sun, Apr 28, 2024 at 05:26:44PM +0300, Dan Carpenter wrote:
>> >> Currently, if an automatically freed allocation is an error pointer that
>> >> will lead to a crash.  An example of this is in wm831x_gpio_dbg_show().
>> >> 
>> >>    171	char *label __free(kfree) = gpiochip_dup_line_label(chip, i);
>> >>    172	if (IS_ERR(label)) {
>> >>    173		dev_err(wm831x->dev, "Failed to duplicate label\n");
>> >>    174		continue;
>> >>    175  }
>> >> 
>> >> The auto clean up function should check for error pointers as well,
>> >> otherwise we're going to keep hitting issues like this.
>> >> 
>> >> Fixes: 54da6a092431 ("locking: Introduce __cleanup() based infrastructure")
>> >> Cc: <stable@xxxxxxxxxxxxxxx>
>> >> Signed-off-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
>> >> ---
>> >> Obviously, the fixes tag isn't very fair but it will tell the -stable
>> >> tools how far to backport this.
>> >> 
>> >>  include/linux/slab.h  | 4 ++--
>> >>  1 files changed, 2 insertions(+), 2 deletions(-)
>> >> 
>> >> diff --git a/include/linux/slab.h b/include/linux/slab.h
>> >> index 4cc37ef22aae..5f5766219375 100644
>> >> --- a/include/linux/slab.h
>> >> +++ b/include/linux/slab.h
>> >> @@ -279,7 +279,7 @@ void kfree(const void *objp);
>> >>  void kfree_sensitive(const void *objp);
>> >>  size_t __ksize(const void *objp);
>> >>  
>> >> -DEFINE_FREE(kfree, void *, if (_T) kfree(_T))
>> >> +DEFINE_FREE(kfree, void *, if (!IS_ERR_OR_NULL(_T)) kfree(_T))
>> > 
>> > Wait, why do we check 'if (_T)' at all?  kfree() already handles NULL
>> > pointers just fine.  I wouldn't be averse to making it handle error
>> > pointers either.
>> 
>> Making kfree() handle IS_ERR() is perhaps a discussion for something else
>> than a stable fix. But Christoph has a point that kfree() checks
>> ZERO_OR_NULL_PTR. Here we check IS_ERR_OR_NULL. How about we checked only
>> IS_ERR here so it makes some sense?
>> 
> 
> I wondered why Peter Z wrote it like this as well...  I think he did
> it so the compiler can figure out which calls to kfree() are unnecessary
> and remove them.  These functions are inline and kfree() is not.  I
> haven't measured to see if it actually results in a space savings but
> the theory is sound.

Hmm that makes sense. There seem to be places that initialize the
__free(kfree) variable to NULL and only at some point actually allocate, and
between those there are possible returns, i.e. ice_init_hw().

OK, patch applied as-is to slab/for-6.9-rc7/fixes, thanks.

> regards,
> dan carpenter
>