Rewrite ethread library

Large parts of the ethread library have been rewritten. The
ethread library is an Erlang runtime system internal, portable
thread library used by the runtime system itself.

Most notable improvement is a reader optimized rwlock
implementation which dramatically improve the performance of
read-lock/read-unlock operations on multi processor systems by
avoiding ping-ponging of the rwlock cache lines. The reader
optimized rwlock implementation is used by miscellaneous
rwlocks in the runtime system that are known to be read-locked
frequently, and can be enabled on ETS tables by passing the
`{read_concurrency, true}' option upon table creation. See the
documentation of `ets:new/2' for more information.

The ethread library can now also use the libatomic_ops library
for atomic memory accesses. This makes it possible for the
Erlang runtime system to utilize optimized atomic operations
on more platforms than before. Use the
`--with-libatomic_ops=PATH' configure command line argument
when specifying where the libatomic_ops installation is
located. The libatomic_ops library can be downloaded from:
http://www.hpl.hp.com/research/linux/atomic_ops/

The changed API of the ethread library has also caused
modifications in the Erlang runtime system. Preparations for
the to come "delayed deallocation" feature has also been done
since it depends on the ethread library.

Note: When building for x86, the ethread library will now use
instructions that first appeared on the pentium 4 processor. If
you want the runtime system to be compatible with older
processors (back to 486) you need to pass the
`--enable-ethread-pre-pentium4-compatibility' configure command
line argument when configuring the system.

1 files changed, 185 insertions, 0 deletions

diff --git a/erts/include/internal/ethr_optimized_fallbacks.h b/erts/include/internal/ethr_optimized_fallbacks.h
new file mode 100644
index 0000000000..2f9f987d0b
--- /dev/null
+++ b/erts/include/internal/ethr_optimized_fallbacks.h
@@ -0,0 +1,185 @@
+/*
+ * %CopyrightBegin%
+ *
+ * Copyright Ericsson AB 2010. All Rights Reserved.
+ *
+ * The contents of this file are subject to the Erlang Public License,
+ * Version 1.1, (the "License"); you may not use this file except in
+ * compliance with the License. You should have received a copy of the
+ * Erlang Public License along with this software. If not, it can be
+ * retrieved online at http://www.erlang.org/.
+ *
+ * Software distributed under the License is distributed on an "AS IS"
+ * basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See
+ * the License for the specific language governing rights and limitations
+ * under the License.
+ *
+ * %CopyrightEnd%
+ */
+
+/*
+ * Description: "Optimized" fallbacks used when native ops are missing
+ * Author: Rickard Green
+ */
+
+#ifndef ETHR_OPTIMIZED_FALLBACKS_H__
+#define ETHR_OPTIMIZED_FALLBACKS_H__
+
+#ifdef ETHR_HAVE_NATIVE_ATOMICS
+#define ETHR_HAVE_OPTIMIZED_ATOMIC_OPS 1
+#endif
+
+#ifdef ETHR_HAVE_NATIVE_SPINLOCKS
+#define ETHR_HAVE_OPTIMIZED_SPINLOCKS 1
+#elif defined(ETHR_HAVE_PTHREAD_SPIN_LOCK)
+/* --- Optimized spinlocks using pthread spinlocks -------------------------- */
+#define ETHR_HAVE_OPTIMIZED_SPINLOCKS 1
+
+typedef pthread_spinlock_t ethr_opt_spinlock_t;
+
+#if defined(ETHR_TRY_INLINE_FUNCS) || defined(ETHR_AUX_IMPL__)
+
+static ETHR_INLINE int
+ethr_opt_spinlock_init(ethr_opt_spinlock_t *lock)
+{
+    return pthread_spin_init((pthread_spinlock_t *) lock, 0);
+}
+
+static ETHR_INLINE int
+ethr_opt_spinlock_destroy(ethr_opt_spinlock_t *lock)
+{
+    return pthread_spin_destroy((pthread_spinlock_t *) lock);
+}
+
+
+static ETHR_INLINE int
+ethr_opt_spin_unlock(ethr_opt_spinlock_t *lock)
+{
+    return pthread_spin_unlock((pthread_spinlock_t *) lock);
+}
+
+static ETHR_INLINE int
+ethr_opt_spin_lock(ethr_opt_spinlock_t *lock)
+{
+    return pthread_spin_lock((pthread_spinlock_t *) lock);
+}
+
+#endif
+
+#elif defined(ETHR_HAVE_NATIVE_ATOMICS)
+/* --- Native spinlocks using native atomics -------------------------------- */
+#define ETHR_HAVE_NATIVE_SPINLOCKS 1
+#define ETHR_HAVE_OPTIMIZED_SPINLOCKS 1
+
+typedef ethr_native_atomic_t ethr_native_spinlock_t;
+
+#if defined(ETHR_TRY_INLINE_FUNCS) || defined(ETHR_AUX_IMPL__)
+
+static ETHR_INLINE void
+ethr_native_spinlock_init(ethr_native_spinlock_t *lock)
+{
+    ethr_native_atomic_init((ethr_native_atomic_t *) lock, 0);
+}
+
+static ETHR_INLINE void
+ethr_native_spin_unlock(ethr_native_spinlock_t *lock)
+{
+    ETHR_COMPILER_BARRIER;
+    ETHR_ASSERT(ethr_native_atomic_read((ethr_native_atomic_t *) lock) == 1);
+    ethr_native_atomic_set_relb((ethr_native_atomic_t *) lock, 0);
+}
+
+static ETHR_INLINE void
+ethr_native_spin_lock(ethr_native_spinlock_t *lock)
+{
+    while (ethr_native_atomic_cmpxchg_acqb((ethr_native_atomic_t *) lock,
+					   (long) 1, (long) 0) != 0) {
+	ETHR_SPIN_BODY;
+    }
+    ETHR_COMPILER_BARRIER;
+}
+
+#endif
+
+#endif
+
+
+#ifdef ETHR_HAVE_NATIVE_RWSPINLOCKS
+#define ETHR_HAVE_OPTIMIZED_RWSPINLOCKS 1
+#elif defined(ETHR_HAVE_NATIVE_ATOMICS)
+/* --- Native rwspinlocks using native atomics ------------------------------ */
+#define ETHR_HAVE_NATIVE_RWSPINLOCKS 1
+#define ETHR_HAVE_OPTIMIZED_RWSPINLOCKS 1
+
+typedef ethr_native_atomic_t ethr_native_rwlock_t;
+
+#if defined(ETHR_TRY_INLINE_FUNCS) || defined(ETHR_AUX_IMPL__)
+
+#define ETHR_WLOCK_FLAG__ (((long) 1) << 30)
+
+static ETHR_INLINE void
+ethr_native_rwlock_init(ethr_native_rwlock_t *lock)
+{
+    ethr_native_atomic_init((ethr_native_atomic_t *) lock, 0);
+}
+
+static ETHR_INLINE void
+ethr_native_read_unlock(ethr_native_rwlock_t *lock)
+{
+    ETHR_COMPILER_BARRIER;
+#ifdef DEBUG
+    ETHR_ASSERT(ethr_native_atomic_read((ethr_native_atomic_t *) lock) >= 0);
+#endif
+    ethr_native_atomic_dec_relb((ethr_native_atomic_t *) lock);
+}
+
+static ETHR_INLINE void
+ethr_native_read_lock(ethr_native_rwlock_t *lock)
+{
+    long act, exp = 0;
+    while (1) {
+	act = ethr_native_atomic_cmpxchg_acqb((ethr_native_atomic_t *) lock,
+					      exp+1, exp);
+	if (act == exp)
+	    break;
+	ETHR_SPIN_BODY;
+	exp = (act & ETHR_WLOCK_FLAG__) ? 0 : act;
+    }
+    ETHR_COMPILER_BARRIER;
+}
+
+static ETHR_INLINE void
+ethr_native_write_unlock(ethr_native_rwlock_t *lock)
+{
+    ETHR_COMPILER_BARRIER;
+    ETHR_ASSERT(ethr_native_atomic_read((ethr_native_atomic_t *) lock)
+		== ETHR_WLOCK_FLAG__);
+    ethr_native_atomic_set_relb((ethr_native_atomic_t *) lock, 0);
+}
+
+static ETHR_INLINE void
+ethr_native_write_lock(ethr_native_rwlock_t *lock)
+{
+    long act, exp = 0;
+    while (1) {
+	act = ethr_native_atomic_cmpxchg_acqb((ethr_native_atomic_t *) lock,
+					      exp|ETHR_WLOCK_FLAG__, exp);
+	if (act == exp)
+	    break;
+	ETHR_SPIN_BODY;
+	exp = act & ~ETHR_WLOCK_FLAG__;
+    }
+    act |= ETHR_WLOCK_FLAG__;
+    /* Wait for readers to leave */
+    while (act != ETHR_WLOCK_FLAG__) {
+	ETHR_SPIN_BODY;
+	act = ethr_native_atomic_read_acqb((ethr_native_atomic_t *) lock);
+    }
+    ETHR_COMPILER_BARRIER;
+}
+
+#endif
+
+#endif
+
+#endif