[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180228125629.3062ac1d.cohuck@redhat.com>
Date: Wed, 28 Feb 2018 12:56:29 +0100
From: Cornelia Huck <cohuck@...hat.com>
To: Arvind Yadav <arvind.yadav.cs@...il.com>
Cc: jwi@...ux.vnet.ibm.com, ubraun@...ux.vnet.ibm.com,
davem@...emloft.net, linux-kernel@...r.kernel.org,
linux-s390@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: [PATCH] net: iucv: Free memory obtained by kzalloc
On Wed, 28 Feb 2018 17:14:55 +0530
Arvind Yadav <arvind.yadav.cs@...il.com> wrote:
> On Wednesday 28 February 2018 04:00 PM, Cornelia Huck wrote:
> > On Wed, 28 Feb 2018 15:24:16 +0530
> > Arvind Yadav <arvind.yadav.cs@...il.com> wrote:
> >
> >> Free memory, if afiucv_iucv_init is not successful and
> >> removing a IUCV driver.
> >>
> >> Signed-off-by: Arvind Yadav <arvind.yadav.cs@...il.com>
> >> ---
> >> net/iucv/af_iucv.c | 5 ++++-
> >> 1 file changed, 4 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c
> >> index 1e8cc7b..eb0995a 100644
> >> --- a/net/iucv/af_iucv.c
> >> +++ b/net/iucv/af_iucv.c
> >> @@ -2433,9 +2433,11 @@ static int afiucv_iucv_init(void)
> >> af_iucv_dev->driver = &af_iucv_driver;
> >> err = device_register(af_iucv_dev);
> >> if (err)
> >> - goto out_driver;
> >> + goto out_iucv_dev;
> >> return 0;
> >>
> >> +out_iucv_dev:
> >> + kfree(af_iucv_dev);
> >> out_driver:
> >> driver_unregister(&af_iucv_driver);
> >> out_iucv:
> >> @@ -2496,6 +2498,7 @@ static void __exit afiucv_exit(void)
> >> {
> >> if (pr_iucv) {
> >> device_unregister(af_iucv_dev);
> >> + kfree(af_iucv_dev);
> >> driver_unregister(&af_iucv_driver);
> >> pr_iucv->iucv_unregister(&af_iucv_handler, 0);
> >> symbol_put(iucv_if);
> > No, you must not use kfree() after you called device_register() (even
> > if it was not successful!) -- see the comment for device_register().
> Yes, Your are right. First we need to call put_device() then kfree().
> I will send updated patch.
No, that's not correct, either. device_register() will give up any
reference it obtained, and the caller did not obtain any additional
reference, so a put_device() would be wrong. A kfree() on a refcounted
structure is wrong as well.
Powered by blists - more mailing lists