Proper maintenance and care of multi-threading locks

The following strategies are used to ensure that the code is dead-lock free (generally by addressing the 4th Coffman condition: circular wait).

1. structure code such that only one lock will need to be acquired at a time
2. always acquire shared locks in the same order, as given by the table below
3. avoid constructs that expect to need unrestricted recursion

Locks

Below are all of the locks that exist in the system and the mechanisms for using them that avoid the potential for deadlocks (no Ostrich algorithm allowed here):

The following are definitely leaf locks (level 1), and must not try to acquire any other lock:

• safepoint

  Note that this lock is acquired implicitly by JL_LOCK and JL_UNLOCK. use the _NOGC variants to avoid that for level 1 locks.

  While holding this lock, the code must not do any allocation or hit any safepoints. Note that there are safepoints when doing allocation, enabling / disabling GC, entering / restoring exception frames, and taking / releasing locks.

• shared_map

• finalizers

• pagealloc

• gcpermlock

• flisp

• jlinstackwalk (Win32)

• ResourcePool<?>::mutex

• RLST_mutex

• jllockedstream::mutex

• debuginfo_asyncsafe

• inferencetimingmutex

  flisp itself is already threadsafe, this lock only protects the jl_ast_context_list_t pool likewise, the ResourcePool<?>::mutexes just protect the associated resource pool

The following is a leaf lock (level 2), and only acquires level 1 locks (safepoint) internally:

• typecache
• Module->lock
• JLDebuginfoPlugin::PluginMutex
• newlyinferredmutex

The following is a level 3 lock, which can only acquire level 1 or level 2 locks internally:

• Method->writelock

The following is a level 4 lock, which can only recurse to acquire level 1, 2, or 3 locks:

• MethodTable->writelock

No Julia code may be called while holding a lock above this point.

orc::ThreadSafeContext (TSCtx) locks occupy a special spot in the locking hierarchy. They are used to protect LLVM's global non-threadsafe state, but there may be an arbitrary number of them. By default, all of these locks may be treated as level 5 locks for the purposes of comparing with the rest of the hierarchy. Acquiring a TSCtx should only be done from the JIT's pool of TSCtx's, and all locks on that TSCtx should be released prior to returning it to the pool. If multiple TSCtx locks must be acquired at the same time (due to recursive compilation), then locks should be acquired in the order that the TSCtxs were borrowed from the pool.

The following are a level 6 lock, which can only recurse to acquire locks at lower levels:

• codegen
• jlmodulesmutex

The following is an almost root lock (level end-1), meaning only the root look may be held when trying to acquire it:

• typeinf

  this one is perhaps one of the most tricky ones, since type-inference can be invoked from many points

  currently the lock is merged with the codegen lock, since they call each other recursively

The following lock synchronizes IO operation. Be aware that doing any I/O (for example, printing warning messages or debug information) while holding any other lock listed above may result in pernicious and hard-to-find deadlocks. BE VERY CAREFUL!

• iolock

• Individual ThreadSynchronizers locks

  this may continue to be held after releasing the iolock, or acquired without it, but be very careful to never attempt to acquire the iolock while holding it

The following is the root lock, meaning no other lock shall be held when trying to acquire it:

• toplevel

  this should be held while attempting a top-level action (such as making a new type or defining a new method): trying to obtain this lock inside a staged function will cause a deadlock condition!

  additionally, it's unclear if any code can safely run in parallel with an arbitrary toplevel expression, so it may require all threads to get to a safepoint first

Broken Locks

The following locks are broken:

• toplevel

  doesn't exist right now

  fix: create it

• Module->lock

  This is vulnerable to deadlocks since it can't be certain it is acquired in sequence. Some operations (such as import_module) are missing a lock.

  fix: replace with jl_modules_mutex?

• loading.jl: require and register_root_module

  This file potentially has numerous problems.

  fix: needs locks

Shared Global Data Structures

These data structures each need locks due to being shared mutable global state. It is the inverse list for the above lock priority list. This list does not include level 1 leaf resources due to their simplicity.

MethodTable modifications (def, cache) : MethodTable->writelock

Type declarations : toplevel lock

Type application : typecache lock

Global variable tables : Module->lock

Module serializer : toplevel lock

JIT & type-inference : codegen lock

MethodInstance/CodeInstance updates : Method->writelock, codegen lock

• These are set at construction and immutable:
  • specTypes
  • sparam_vals
  • def

• These are set by jl_type_infer (while holding codegen lock):
  • cache
  • rettype
  • inferred

    * valid ages

• inInference flag:
  • optimization to quickly avoid recurring into jl_type_infer while it is already running
  • actual state (of setting inferred, then fptr) is protected by codegen lock

• Function pointers:

  • these transition once, from NULL to a value, while the codegen lock is held

• Code-generator cache (the contents of functionObjectsDecls):

  • these can transition multiple times, but only while the codegen lock is held
  • it is valid to use old version of this, or block for new versions of this, so races are benign, as long as the code is careful not to reference other data in the method instance (such as rettype) and assume it is coordinated, unless also holding the codegen lock

LLVMContext : codegen lock

Method : Method->writelock

• roots array (serializer and codegen)
• invoke / specializations / tfunc modifications