linux-kernel - Re: [PATCH v2] mm/gup: Fix memory leak in __gup_benchmark

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <450b93d7-9629-6c00-ea75-d68369435931@redhat.com>
Date:   Mon, 16 Dec 2019 13:18:45 +0100
From:   David Hildenbrand <david@...hat.com>
To:     Navid Emamdoost <navid.emamdoost@...il.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        "Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
        Keith Busch <keith.busch@...el.com>, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Cc:     emamd001@....edu, jhubbard@...dia.com
Subject: Re: [PATCH v2] mm/gup: Fix memory leak in __gup_benchmark_ioctl

On 13.12.19 23:37, Navid Emamdoost wrote:
> In the implementation of __gup_benchmark_ioctl() the allocated pages
> should be released before returning in case of an invalid cmd. Release
> pages via kvfree() by goto done.
> 
> Fixes: 714a3a1ebafe ("mm/gup_benchmark.c: add additional pinning methods")
> Signed-off-by: Navid Emamdoost <navid.emamdoost@...il.com>
> ---
> Changes in v2:
> 	-- added goto and ret value instead of return -1.
> ---
>  mm/gup_benchmark.c | 9 +++++----
>  1 file changed, 5 insertions(+), 4 deletions(-)
> 
> diff --git a/mm/gup_benchmark.c b/mm/gup_benchmark.c
> index b160638f647e..b773b2568544 100644
> --- a/mm/gup_benchmark.c
> +++ b/mm/gup_benchmark.c
> @@ -24,7 +24,7 @@ static int __gup_benchmark_ioctl(unsigned int cmd,
>  {
>  	ktime_t start_time, end_time;
>  	unsigned long i, nr_pages, addr, next;
> -	int nr;
> +	int nr, ret = 0;
>  	struct page **pages;
>  
>  	if (gup->size > ULONG_MAX)
> @@ -63,8 +63,8 @@ static int __gup_benchmark_ioctl(unsigned int cmd,
>  					    NULL);
>  			break;
>  		default:
> -			kvfree(pages);
> -			return -1;
> +			ret = -EINVAL;
> +			goto done;
>  		}
>  
>  		if (nr <= 0)
> @@ -85,8 +85,9 @@ static int __gup_benchmark_ioctl(unsigned int cmd,
>  	end_time = ktime_get();
>  	gup->put_delta_usec = ktime_us_delta(end_time, start_time);
>  
> +done:
>  	kvfree(pages);
> -	return 0;
> +	return ret;
>  }
>  
>  static long gup_benchmark_ioctl(struct file *filep, unsigned int cmd,
> 

You should really CC people that give review feedback in previous
iterations (IOW me :) ) and properly tag versions/mention code change
since the last version.

(v1 was already <20191122224117.2372-1-navid.emamdoost@...il.com>, this
here would be v3)

Also, this patch here is not a standalone patch, it is an addon on top
of <20191211174653.4102-1-navid.emamdoost@...il.com>

Enough rambling :) This squashed into
<20191211174653.4102-1-navid.emamdoost@...il.com>:

Reviewed-by: David Hildenbrand <david@...hat.com>


I agree that the label could have a better name (e.g., out_free)

-- 
Thanks,

David / dhildenb