Received-SPF: softfail (google.com: domain of transitioning linux-kernel-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) client-ip=23.128.96.19;
To:     trix@redhat.com, jpoimboe@redhat.com, jikos@kernel.org,
        mbenes@suse.cz, pmladek@suse.com, nathan@kernel.org,
        ndesaulniers@google.com
Cc:     live-patching@vger.kernel.org, linux-kernel@vger.kernel.org,
        llvm@lists.linux.dev
References: <20220320015143.2208591-1-trix@redhat.com>
From:   Joe Lawrence <joe.lawrence@redhat.com>
Subject: Re: [PATCH v2] livepatch: Reorder to use before freeing a pointer
Message-ID: <3ee9826e-b770-d015-0251-e9770172d973@redhat.com>
Date:   Mon, 21 Mar 2022 09:39:34 -0400
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.10.1
MIME-Version: 1.0
In-Reply-To: <20220320015143.2208591-1-trix@redhat.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 3/19/22 9:51 PM, trix@redhat.com wrote:
> From: Tom Rix <trix@redhat.com>
> 
> Clang static analysis reports this issue
> livepatch-shadow-fix1.c:113:2: warning: Use of
>   memory after it is freed
>   pr_info("%s: dummy @ %p, prevented leak @ %p\n",
>   ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 
> The pointer is freed in the previous statement.
> Reorder the pr_info to report before the free.
> 
> Similar issue in livepatch-shadow-fix2.c
> 
> Signed-off-by: Tom Rix <trix@redhat.com>
> ---
> v2: Fix similar issue in livepatch-shadow-fix2.c
> 
>  samples/livepatch/livepatch-shadow-fix1.c | 2 +-
>  samples/livepatch/livepatch-shadow-fix2.c | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/samples/livepatch/livepatch-shadow-fix1.c b/samples/livepatch/livepatch-shadow-fix1.c
> index 918ce17b43fda..6701641bf12d4 100644
> --- a/samples/livepatch/livepatch-shadow-fix1.c
> +++ b/samples/livepatch/livepatch-shadow-fix1.c
> @@ -109,9 +109,9 @@ static void livepatch_fix1_dummy_leak_dtor(void *obj, void *shadow_data)
>  	void *d = obj;
>  	int **shadow_leak = shadow_data;
>  
> -	kfree(*shadow_leak);
>  	pr_info("%s: dummy @ %p, prevented leak @ %p\n",
>  			 __func__, d, *shadow_leak);
> +	kfree(*shadow_leak);
>  }
>  
>  static void livepatch_fix1_dummy_free(struct dummy *d)
> diff --git a/samples/livepatch/livepatch-shadow-fix2.c b/samples/livepatch/livepatch-shadow-fix2.c
> index 29fe5cd420472..361046a4f10cf 100644
> --- a/samples/livepatch/livepatch-shadow-fix2.c
> +++ b/samples/livepatch/livepatch-shadow-fix2.c
> @@ -61,9 +61,9 @@ static void livepatch_fix2_dummy_leak_dtor(void *obj, void *shadow_data)
>  	void *d = obj;
>  	int **shadow_leak = shadow_data;
>  
> -	kfree(*shadow_leak);
>  	pr_info("%s: dummy @ %p, prevented leak @ %p\n",
>  			 __func__, d, *shadow_leak);
> +	kfree(*shadow_leak);
>  }
>  
>  static void livepatch_fix2_dummy_free(struct dummy *d)
> 

Hi Tom,

Ordering doesn't matter for the example, so let's clean up the static
analysis.

Acked-by: Joe Lawrence <joe.lawrence@redhat.com>

But for my sanity, isn't this a false positive?  There shouldn't be harm
in printing the pointer itself, even after what it points to has been
freed, i.e.

	int *i = malloc(sizeof(*i));
	free(i);
	printf("%p\n", i);      << ok
	printf("%d\n", *i);     << NOT ok

But I suppose clang doesn't know that the passed pointer isn't getting
dereferenced by the function, so it throws up a warning?  Just curious
what your experience has been with respect to these reports.

Thanks,
-- 
Joe