Messages in this thread | | | Date | Thu, 11 Mar 2021 13:58:33 +0000 | From | Luis Chamberlain <> | Subject | Re: Re: [PATCH] lib/test_kmod: Fix a use after free in register_test_dev_kmod |
| |
On Thu, Mar 11, 2021 at 09:31:55PM +0800, lyl2019@mail.ustc.edu.cn wrote: > > > > > -----原始邮件----- > > 发件人: "Luis Chamberlain" <mcgrof@kernel.org> > > 发送时间: 2021-03-11 21:01:08 (星期四) > > 收件人: "Lv Yunlong" <lyl2019@mail.ustc.edu.cn> > > 抄送: linux-kernel@vger.kernel.org > > 主题: Re: [PATCH] lib/test_kmod: Fix a use after free in register_test_dev_kmod > > > > On Thu, Mar 11, 2021 at 12:02:46AM -0800, Lv Yunlong wrote: > > > In register_test_dev_kmod, it calls free_test_dev_kmod() to free > > > test_dev. But free_test_dev_kmod() can't set the original pointer > > > test_dev to NULL, because the test_dev was passed by it's value > > > not reference. > > > > Did you actually get a crash or something? If so can you supply the > > actual log? If this is just an observation and you think this is an > > issue, specifying that would help during patch review. > > > > Luis > > > > > Signed-off-by: Lv Yunlong <lyl2019@mail.ustc.edu.cn> > > > --- > > > lib/test_kmod.c | 2 +- > > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > > > diff --git a/lib/test_kmod.c b/lib/test_kmod.c > > > index 38c250fbace3..aa8a2a563d7e 100644 > > > --- a/lib/test_kmod.c > > > +++ b/lib/test_kmod.c > > > @@ -1124,7 +1124,6 @@ static void free_test_dev_kmod(struct kmod_test_device *test_dev) > > > free_test_dev_info(test_dev); > > > kmod_config_free(test_dev); > > > vfree(test_dev); > > > - test_dev = NULL; > > > } > > > } > > > > > > @@ -1149,6 +1148,7 @@ static struct kmod_test_device *register_test_dev_kmod(void) > > > if (ret) { > > > pr_err("could not register misc device: %d\n", ret); > > > free_test_dev_kmod(test_dev); > > > + test_dev = NULL; > > > goto out; > > > } > > > > > > -- > > > 2.25.1 > > > > > > > > This problem was reported by source code analyzers developed by our Security Lab(Loccs).
This should be included in the commit log, please.
> We have confirmed this issue before submiting the patch.
How was this confirmed exactly.
Luis
| |