Re: kernel BUG at kernel/cred.c:434!

["chengjian (D)" <cj.chengjian@huawei.com>]

On 2019/4/16 11:40, Kees Cook wrote:
> On Mon, Apr 15, 2019 at 11:20 AM Paul Moore <paul@paul-moore.com> wrote:
>> On Mon, Apr 15, 2019 at 11:05 AM Oleg Nesterov <oleg@redhat.com> wrote:
>>> On 04/15, Paul Moore wrote:
>>>> On Mon, Apr 15, 2019 at 9:43 AM Oleg Nesterov <oleg@redhat.com> wrote:
>>>>> Well, acct("/proc/self/attr/current") doesn't look like a good idea, but I do
>>>>> not know where should we put the additional check... And probably
>>>>> "echo /proc/self/attr/current > /proc/sys/kernel/core_pattern" can hit the
>>>>> same problem, do_coredump() does override_creds() too.
>>>>>
>>>>> May be just add
>>>>>
>>>>>          if (current->cred != current->real_cred)
>>>>>                  return -EACCES;
>>>>>
>>>>> into proc_pid_attr_write(), I dunno.
>>>> Is the problem that do_acct_process() is calling override_creds() and
>>>> the returned/old credentials are being freed before do_acct_process()
>>>> can reinstall the creds via revert_creds()?  Presumably because the
>>>> process accounting is causing the credentials to be replaced?
>>> Afaics, the problem is that do_acct_process() does override_creds() and
>>> then __kernel_write(). Which calls proc_pid_attr_write(), which in turn calls
>>> selinux_setprocattr(), which does another prepare_creds() + commit_creds();
>>> and commit_creds() hits
>>>
>>>          BUG_ON(task->cred != old);
>> Gotcha.  In the process of looking at the backtrace I forgot about the
>> BUG_ON() at the top of the oops message.
>>
>> I wonder what terrible things would happen if we changed the BUG_ON()
>> in commit_creds to simple returning an error an error code to the
>> caller.  There is a warning/requirement in commit_creds() function
>> header comment that it should always return 0.
> Would callers be expected to call abort_creds() on failure? There are
> a number of places where it'd need fixing up. And would likely be best
> with a __must_check marking.
>
> It seems like avoiding the pathological case might be simpler?


Yeah, Avoiding this pathological case is a better solution.

It seems like that we can't commit_creds() during

override_creds() and revert_creds().

So how about just put commit_creds outside !


just like:

     override_creds()   // cred  -=> new

     // may BUG_ON if commit_creds done.

     revert_creds()     //  cred -=> old                         
<-----------|

     commit_creds   //  cred = real_cred = new                |

[revert_creds]--------------------------------------------------|


[1]--Before we call commit_creds in selinux_setprocattr(),

if we find that cred != real_cred, it may have been overridden

before, we should revert it.

[2]--The same to revert_creds, when we found someone have committed,

orig_cred != current->real_cred may hits, this means that

we have reverted before(see [1]).

[3]--Sometimes new and old are the same, then we need to consider this

situation specially.


The code just like:


From: Yang Yingliang <yangyingliang@huawei.com>
Date: Sat, 13 Apr 2019 21:56:01 +0800
Subject: [PATCH] fix cred bug_on

Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
---
  kernel/acct.c            | 3 ++-
  kernel/cred.c            | 6 ++++++
  security/selinux/hooks.c | 2 ++
  3 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/kernel/acct.c b/kernel/acct.c
index addf7732fb56..f2065f899eee 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -522,7 +522,8 @@ static void do_acct_process(struct bsd_acct_struct 
*acct)
      }
  out:
      current->signal->rlim[RLIMIT_FSIZE].rlim_cur = flim;
-    revert_creds(orig_cred);
+    if (orig_cred == current->real_cred)    // [2]
+        revert_creds(orig_cred);
  }

  /**
diff --git a/kernel/cred.c b/kernel/cred.c
index ecf03657e71c..c4d5ba92fb9b 100644
--- a/kernel/cred.c
+++ b/kernel/cred.c
@@ -522,6 +522,9 @@ const struct cred *override_creds(const struct cred 
*new)
  {
      const struct cred *old = current->cred;

+    if (old == new)    //  [3]
+        return old;
+
      kdebug("override_creds(%p{%d,%d})", new,
             atomic_read(&new->usage),
             read_cred_subscribers(new));
@@ -551,6 +554,9 @@ void revert_creds(const struct cred *old)
  {
      const struct cred *override = current->cred;

+    if (override == old)    // [3]
+        return;
+
      kdebug("revert_creds(%p{%d,%d})", old,
             atomic_read(&old->usage),
             read_cred_subscribers(old));
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index b5017beb4ef7..bc8108e4e90f 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -6590,6 +6590,8 @@ static int selinux_setprocattr(const char *name, 
void *value, size_t size)
          goto abort_change;
      }

+    if (current->cred != current->real_cred)    // [1]
+        revert_creds(current->real_cred);
      commit_creds(new);
      return size;

-- 
2.17.1


We have tested this patch for 3 days and it works well.

Are there any cases that are not covered here ?


Thanks.

     Cheng Jian