Changes between Version 3 and Version 4 of library_locks

Timestamp:

Dec 2, 2014, 6:12:26 PM (10 years ago)

Author:

alain

Comment:

--

library_locks

@@ -1 @@
-= The spin_lock library =
+= The user_lock library =
 
-The [source:soft/giet_vm/giet_libs/spin_lock.c spin_lock.c] and [source:soft/giet_vm/giet_libs/spin_lock.h spin_lock.h] 
-files define a user-level spin-lock service to obtain exclusive access to a shared resource in a parallel multi-tasks application.
+[[PageOutline]]
+
+The [source:soft/giet_vm/giet_libs/user_lock.c user_lock.c] and [source:soft/giet_vm/giet_libs/user_lock.h user_lock.h] 
+files define two types of services:
+ * user-level atomic access functions, to increment a shared counter.
+ * user-level spin-locks, to obtain exclusive access to a shared resource in a parallel multi-tasks application.
+The proposed spin-lock implement a waiting queue service, to enforce fairness and avoid live-lock situations.
 Each lock (giet_lock_t object) occupies a complete 64 bytes cache line to avoid false sharing.
-If the hardware architecture supports hardware cache coherence, it is recommanded to store the lock in a cachable segment architecture
 
  == Lock placement ==
 
-The lock being shared by several tasks, is defined as a global variable in the application code.
+The lock being shared by several tasks, should be defined as a global variable of type '''user_lock_t''' in the application code.
 It is possible to control precisely the placement of a specific lock by defining a specific vseg in the application mapping. 
 
  == Access functions ==
 
- * '''void lock_acquire( giet_lock_t * lock )'''
-This blocking function uses an atomic LL/SC mechanism to take the lock, and return only when the lock has been successfully taken.
+ === unsigned int '''atomic_increment'''( unsigned int* shared, unsigned int increment) ===
+This function uses a LL/SC to atomically increment a shared variable.
+ * '''shared''' : pointer on the shared variable
+ * '''increment''' : increment value.
 
- * '''void lock_release( giet_lock_t * lock )'''
-This function uses a simple SW to release the lock. It can be used for lock initialization.
+ === '''void lock_init'''( user_lock_t * lock ) ===
+This function should be called by one single task.
+ * '''lock''' : pointer on the user_lock_t structure.
+
+ === '''void lock_acquire'''( giet_lock_t * lock ) ===
+This blocking function returns only when the lock has been successfully taken.
+ * '''lock''' : pointer on the user_lock_t structure.
+
+ === '''void lock_release'''( giet_lock_t * lock ) ===
+This function releases the lock. It must be called by a task after a successful lock_acquire().
+ * '''lock''' : pointer on the user_lock_t structure.