Re: [autofs] [RFC PATCH]autofs4: hang and proposed fix

[Will Taber <wtaber@us.ibm.com>]

Ian Kent wrote:
> On Thu, 1 Dec 2005, William H. Taber wrote:
> 
> 
>>Ian Kent wrote:
>>
>>>On Wed, 30 Nov 2005, William H. Taber wrote:
>>>
>>>
>>>
>>>>Trond Myklebust wrote:
>>>>
>>>>
>>>>>On Wed, 2005-11-30 at 15:32 -0500, William H. Taber wrote:
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>>Not only is there this case, but the original premise is wrong as
>>>>>>well.
>>>>>>There is a second case in which a d_revalidate function is called with
>>>>>>the
>>>>>>parent i_sem and that is when it is called from inside of
>>>>>>lookup_one_len.
>>>>>>What makes this tricky is that lookup_one_len is called from
>>>>>>nfs_sillyrename from inside of nfs_rename which is called, naturally
>>>>>>enough by sys_rename.  The rename code is very careful about the order
>>>>>>in
>>>>>>which it obtains the parent semaphores because it needs to get two of
>>>>>>them.  It must always obtain the locks in the same order so that does
>>>>>>not
>>>>>>get into a deadly embrace.  If we start arbitrarily releasing a parent
>>>>>>semaphore in cached_lookup and taking it again after the revalidate,
>>>>>>we
>>>>>>risk breaking the lock ordering and creating a deadly embrace.
>>>>>>
>>>>>>When I started writing this I thought that it would be safe for the
>>>>>>autofs
>>>>>>revalidate code to release the parent semaphore because they do not
>>>>>>have a
>>>>>>rename callback.  But I looked again at the rename code and it calls
>>>>>>lookup_hash on the final source and destination files after locking
>>>>>>the
>>>>>>parents so the potential for a deadly embrace still exists unless
>>>>>>there is
>>>>>>some other assurance that these final lookups will never pend waiting
>>>>>>on
>>>>>>the automounter in either their revalidate or lookup routines.
>>>>>>(Actually
>>>>>>the requirement is that they never give up the parent i_sem lock, but
>>>>>>the
>>>>>>lookup code has to give up the lock so that the autofs demon can run
>>>>>>and
>>>>>>perform the mount so it amounts to the same thing.)
>>>>>>
>>>>>>The same issue exists for devfs which also releases the parent i_sem
>>>>>>lock
>>>>>>so that it can wait inside its revalidation routine.
>>>>>
>>>>>
>>>>>So exactly why does autofs4 want to hold the dir->i_sem in d_revalidate
>>>>>in the first place? Can't we move any code that requires dir->i_sem to
>>>>>be held into a ->lookup() method?
>>>>
>>>>It's not that d_revalidate wants or doesn't want to hold the lock.  The
>>>>caller
>>>>of lookup_one_len is required to get the lock and this function calls
>>>>lookup_hash which calls cached_lookup which calls d_revalidate.
>>>>
>>>>
>>>>
>>>>>Trivially, if you have a d_revalidate that does something like
>>>>>
>>>>>int autofs_revalidate(struct dentry *dentry, struct nameidata *nd)
>>>>>{
>>>>> d_drop(dentry);
>>>>> return 0;
>>>>>}
>>>>>
>>>>>then the VFS will currently allocate a new dentry with the same name,
>>>>>and call ->lookup() on it without dropping dir->i_sem. If you still need
>>>>>to reference the old dentry, then put it on a private list somewhere.
>>>>>That would also allow you to return the old dentry as the result of the
>>>>>->lookup() operation if that is desirable.
>>>>
>>>>Problem with that, as I understand it and Ian Kent knows better than I, is
>>>>that the autofs lookup code creates the dentry and fills it in partially
>>>>and
>>>>marks it as waiting for mounting and wakes up the automount demon.  The
>>>>demon
>>>>completes the mount and finishes filling in the dentry.  So we cannot have
>>>>some other lookup coming in and removing the dentry on us.  At least that
>>>>is
>>>>what I understand from Ian's answer when I proposed the same sort of thing
>>>>to
>>>>him.   Even if  they end up doing something like that in a future version
>>>>of
>>>>the automounter, I would still like a simple patch that can be applied to
>>>>existing systems as an interim fix.
>>>
>>>
>>>Lets see if I can keep this explaination simple.
>>>
>>>The user space process using the autofs filesystem (autodir or automount)
>>>needs to be able to call mkdir at mount time as a result of a callback from
>>>revalidate. Sometimes this comes indirectly from lookup (if the directory
>>>does not already exist).
>>>
>>>lookup_one_len requires the i_sem to be held so two instances of a
>>>filesystem calling it lead to a deadlock when mkdir is called from userspace
>>>(the third process). In the case we are discussing this happens because the
>>>first process calls lookup which releases the i_sem and calls revalidate
>>>itself. The second calls revalidate which doesn't release the i_sem and is
>>>places on a wait queue for mount completion. Consequently the mkdir blocks.
>>>
>>>So the requirement is that autofs release the i_sem during the callback, not
>>>obtain it.
>>>
>>>Will believes that it is not safe for autofs to release i_sem for the
>>>callback to user space because it is possible that path that aquired it may
>>>not be the path that has called revalidate and I can see his point.
>>>
>>>Never the less I'm still not convinced that this is possible given the
>>>restrictions of autofs.
>>>
>>>Let me try and describe this, hopefully more clearly than I've done so far.
>>>
>>>The only operations defined for autofs are:
>>>
>>>mkdir, rmdir, symlink and unlink 
>>>and the only processes that can do these operations must be in the same
>>>process group that mounted the filesystem. EACCESS is returned for all other
>>>processes attempting these operations.
>>>
>>>The other functionality is read-only (and perhaps triggers a mount) being
>>>lookup, revalidate and readdir.
>>>
>>>So the question is, can anyone provide an example of a path that, upon
>>>calling autofs revalidate or lookup with the i_sem held, not be the path
>>>that aquired it?
> 
> 
> So still no counter example!
> 
> 
>>Any other process calling lookup_one_len on a file in /net.
> 
> 
> I'm afraid this is not an example it's an assertion.
> "Any other process" is a little broad I think.
> You'll need to be more specific.
> 
> Consider the example reported by yourself and Ram.
> 
> In that example we have processes P1, P2 and lets call the user space 
> callback P1(mount). Also assume there is a mechamism to check the 
> semaphore, release it if held and later re-take it if previously held, 
> like the patch I offered before.
> 
> Correct me if I'm wrong but, with the assumption above, you report 
> goes like:
> 
> P1 - calls lookup_one_len, takes i_sem and eventually calls 
> autofs4_lookup and indirectly autofs4_revalidate.
> 
> P2 - comes along and waits on i_sem.
And what happens if P3 comes in with a normal lookup without i_sem held 
and calls autofs4_revalidate from do_lookup and wakes up P2? Think both 
about what will happen later in your code path and also what happens 
when P2 tries to release the lock that was no longer held.

> 
> P1 - autofs4_revalidate releases i_sem and posts a user space callback.
> 
> P2 - aquires i_sem and eventually calls autofs4_revalidate, releases 
> i_sem and is posted to the wait queue for mount completion.
> 
> P1(mount) - calls mkdir, aquires i_sem, and calls autofs4_dir_mkdir,  
> i_sem is then released.
> 
> Mount completion is signaled back to autofs4 and the waiters are released.
> 
> P1, P2, in any order each (one after the other due to the semaphore) 
> re-take i_sem and each complete their lookup_one_len calls.
> 
> On both calls to autofs4_revalidate the calling process is itself the 
> holder of i_sem.
> 
> Further, any other process that does a path walk during this time has two 
> possible paths.
> 
> First case, the dentry exists, the process is placed on the wait queue 
> along with P1 and P2 awaiting mount completion without taking i_sem.
> 
> Second case, the dentry does not yet exist, this process either aquires 
> the i_sem in do_lookup and follows a similar path to P1 and waits on the 
> queue for mount completion or it waits on the i_sem while P1 does 
> the lookup and triggers the mount request, it the aquires i_sem find the 
> dentry exists, releases i_sem and calls autofs4_revalidate without i_sem 
> held and is sent to the wait queue to wait for mount completion.
> 
> Again in both these cases a process that enters autofs4_revalidate when 
> the i_sem is held is the process that aquired it.

But a regular lookup can enter autofs4_revalidate at anytime without 
holding i_sem.

The main lookup path does not hold i_sem and Trond was pretty clear 
about why it cannot.  That is why devfs has the code which tries to 
guess whether it is the person holding the lock before it releases it. 
If you put similar code into autofs4_revalidate before you release i_sem 
it would probably work.  This of course makes your code sensitive to 
changes in the lookup code because the devfs code makes assumptions 
about what flags are set on different lookups.  The best fix would be to 
move all of the waiting into autofs4_lookup and not hash the dentry 
until the mount was ready to run.  That is necessarily a large piece of 
coding and would require a lot of testing.  That is why I am suggesting 
for now a patch that determines if the lock was held by the caller or 
not and releasing i_sem if it was, before waiting in autofs4_revalidate. 
  And of course remembering whether or not it needs to retake the lock 
after the wait completes.

Will