From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mateusz Guzik Date: Fri, 25 Apr 2014 19:05:32 +0000 Subject: Re: Clarification needed on use of put_user inside a loop Message-Id: <20140425190531.GD3636@mguzik.redhat.com> List-Id: References: <535A88D5.3030008@gmail.com> In-Reply-To: <535A88D5.3030008@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Kumar Gaurav Cc: kernel-janitors@vger.kernel.org, Dan Carpenter , Greg KH , linux-kernel@vger.kernel.org On Fri, Apr 25, 2014 at 09:39:57PM +0530, Kumar Gaurav wrote: > Hence when transferring data involves loops then checking permission > (using access_ok()) once should be good to go then after we can > simply transfer data using __put_user(), instead of using put_user() > itself in loop. > Well, I can't tell you whether this is a good idea, but: This looks correct and other code is doing this already. However, put_user calls might_fault, but __put_user consumers I found (e.g. copy_siginfo_to_user) don't do that. While it has only debugging purposes and would not change anything for those consumers, it seems to be a bug to not include it. Thus I suggest adding access_ok variant which calls might_fault. -- Mateusz Guzik