Messages in this thread | | | From | Muchun Song <> | Date | Tue, 24 May 2022 10:30:01 +0800 | Subject | Re: [PATCH v2] sysctl: handle table->maxlen properly for proc_dobool |
| |
On Tue, May 24, 2022 at 1:27 AM Luis Chamberlain <mcgrof@kernel.org> wrote: > > On Sun, May 22, 2022 at 01:26:24PM +0800, Muchun Song wrote: > > Setting ->proc_handler to proc_dobool at the same time setting ->maxlen > > to sizeof(int) is counter-intuitive, it is easy to make mistakes. For > > robustness, fix it by reimplementing proc_dobool() properly. > > > > Signed-off-by: Muchun Song <songmuchun@bytedance.com> > > Cc: Luis Chamberlain <mcgrof@kernel.org> > > Cc: Kees Cook <keescook@chromium.org> > > Cc: Iurii Zaikin <yzaikin@google.com> > > --- > > Thanks for your patch Muchun! > > Does this fix an actualy issue? Because the comit log suggest so.
Thanks for taking a look.
I think it is an improvement not a real bug fix. When I first use proc_dobool in my driver, I assign sizeof(variable) to table->maxlen. Then I found it was wrong, it should be sizeof(int) which was counter-intuitive. So it is very easy to make mistakes. Should I add those into the commit log?
Thanks.
> If so is there a bug which is known or a reproducer which can be > implemented to showcase that bug? > > The reason I ask is that we have automatic scrapers for bug fixes, > and I tend to prefer to avoid giving those automatic scrapers > the idea that a patch is a fix for a kernel bug when it it is not. > If what you are change is an optimization then your commit log should > clarify that. > > If you are fixing something then you must be clear about about the > details I mentioned. And then, if it does fix an issue, how long > has the issue been know, what are the consequences of it? And up > to what kernel is this issue present for? > > Luis
| |