[PATCH 05/82] examples/ip_pipeline: remove unnecessary NULL checks

[Stephen Hemminger]

On Mon, 24 Jan 2022 10:17:23 +0000
"Dumitrescu, Cristian" <cristian.dumitrescu at intel.com> wrote:

> > -----Original Message-----
> > From: Stephen Hemminger <stephen at networkplumber.org>
> > Sent: Monday, January 24, 2022 12:04 AM
> > To: dev at dpdk.org
> > Cc: Stephen Hemminger <stephen at networkplumber.org>; Dumitrescu,
> > Cristian <cristian.dumitrescu at intel.com>
> > Subject: [PATCH 05/82] examples/ip_pipeline: remove unnecessary NULL
> > checks
> > 
> > Remove redundant NULL pointer checks before free functions
> > found by nullfree.cocci
> > 
> > Signed-off-by: Stephen Hemminger <stephen at networkplumber.org>
> > ---
> >  examples/ip_pipeline/cli.c       | 12 ++++--------
> >  examples/ip_pipeline/cryptodev.c |  6 ++----
> >  examples/ip_pipeline/thread.c    |  6 ++----
> >  3 files changed, 8 insertions(+), 16 deletions(-)
> >   
> 
> Hi Stephen,
> 
> The rte_ring_free() and rte_mempool_free() do not state in their API description that freeing a NULL pointer is harmless. Before pushing these changes, please add the necessary note in the API header files for these functions.


Okay, will fix the documentation then.
The implemeentation has always done the right thing.

> In the absence of the clear note in their API description, the user is forced to check for the NULL pointer. I agree that the implementation of these functions does the right think and exits early when the input pointer is NULL, but there is no guarantee that the implementation is not going to change. Agree?


No. I don't agree. All XXX_free() functions should accept NULL and handle it in the same way.
Any inconsistency is an obvious source of bugs.

> 
> The stdlib free() and the rte_free() do have the clear API description note that freeing a NULL object is harmless, so removing the NULL check before their call is indeed safe.

Will add first patch to fix the documentation to match the current implementation.