duke@435: /*
duke@435:  * Copyright 2006-2007 Sun Microsystems, Inc.  All Rights Reserved.
duke@435:  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
duke@435:  *
duke@435:  * This code is free software; you can redistribute it and/or modify it
duke@435:  * under the terms of the GNU General Public License version 2 only, as
duke@435:  * published by the Free Software Foundation.
duke@435:  *
duke@435:  * This code is distributed in the hope that it will be useful, but WITHOUT
duke@435:  * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
duke@435:  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
duke@435:  * version 2 for more details (a copy is included in the LICENSE file that
duke@435:  * accompanied this code).
duke@435:  *
duke@435:  * You should have received a copy of the GNU General Public License version
duke@435:  * 2 along with this work; if not, write to the Free Software Foundation,
duke@435:  * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
duke@435:  *
duke@435:  * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
duke@435:  * CA 95054 USA or visit www.sun.com if you need additional information or
duke@435:  * have any questions.
duke@435:  *
duke@435:  */
duke@435: 
duke@435: # include "incls/_precompiled.incl"
duke@435: # include "incls/_copy.cpp.incl"
duke@435: 
duke@435: 
duke@435: // Copy bytes; larger units are filled atomically if everything is aligned.
duke@435: void Copy::conjoint_memory_atomic(void* from, void* to, size_t size) {
duke@435:   address src = (address) from;
duke@435:   address dst = (address) to;
duke@435:   uintptr_t bits = (uintptr_t) src | (uintptr_t) dst | (uintptr_t) size;
duke@435: 
duke@435:   // (Note:  We could improve performance by ignoring the low bits of size,
duke@435:   // and putting a short cleanup loop after each bulk copy loop.
duke@435:   // There are plenty of other ways to make this faster also,
duke@435:   // and it's a slippery slope.  For now, let's keep this code simple
duke@435:   // since the simplicity helps clarify the atomicity semantics of
duke@435:   // this operation.  There are also CPU-specific assembly versions
duke@435:   // which may or may not want to include such optimizations.)
duke@435: 
duke@435:   if (bits % sizeof(jlong) == 0) {
duke@435:     Copy::conjoint_jlongs_atomic((jlong*) src, (jlong*) dst, size / sizeof(jlong));
duke@435:   } else if (bits % sizeof(jint) == 0) {
duke@435:     Copy::conjoint_jints_atomic((jint*) src, (jint*) dst, size / sizeof(jint));
duke@435:   } else if (bits % sizeof(jshort) == 0) {
duke@435:     Copy::conjoint_jshorts_atomic((jshort*) src, (jshort*) dst, size / sizeof(jshort));
duke@435:   } else {
duke@435:     // Not aligned, so no need to be atomic.
duke@435:     Copy::conjoint_bytes((void*) src, (void*) dst, size);
duke@435:   }
duke@435: }
duke@435: 
duke@435: 
duke@435: // Fill bytes; larger units are filled atomically if everything is aligned.
duke@435: void Copy::fill_to_memory_atomic(void* to, size_t size, jubyte value) {
duke@435:   address dst = (address) to;
duke@435:   uintptr_t bits = (uintptr_t) to | (uintptr_t) size;
duke@435:   if (bits % sizeof(jlong) == 0) {
duke@435:     jlong fill = (julong)( (jubyte)value ); // zero-extend
duke@435:     if (fill != 0) {
duke@435:       fill += fill << 8;
duke@435:       fill += fill << 16;
duke@435:       fill += fill << 32;
duke@435:     }
duke@435:     //Copy::fill_to_jlongs_atomic((jlong*) dst, size / sizeof(jlong));
duke@435:     for (uintptr_t off = 0; off < size; off += sizeof(jlong)) {
duke@435:       *(jlong*)(dst + off) = fill;
duke@435:     }
duke@435:   } else if (bits % sizeof(jint) == 0) {
duke@435:     jint fill = (juint)( (jubyte)value ); // zero-extend
duke@435:     if (fill != 0) {
duke@435:       fill += fill << 8;
duke@435:       fill += fill << 16;
duke@435:     }
duke@435:     //Copy::fill_to_jints_atomic((jint*) dst, size / sizeof(jint));
duke@435:     for (uintptr_t off = 0; off < size; off += sizeof(jint)) {
duke@435:       *(jint*)(dst + off) = fill;
duke@435:     }
duke@435:   } else if (bits % sizeof(jshort) == 0) {
duke@435:     jshort fill = (jushort)( (jubyte)value ); // zero-extend
duke@435:     fill += fill << 8;
duke@435:     //Copy::fill_to_jshorts_atomic((jshort*) dst, size / sizeof(jshort));
duke@435:     for (uintptr_t off = 0; off < size; off += sizeof(jshort)) {
duke@435:       *(jshort*)(dst + off) = fill;
duke@435:     }
duke@435:   } else {
duke@435:     // Not aligned, so no need to be atomic.
duke@435:     Copy::fill_to_bytes(dst, size, value);
duke@435:   }
duke@435: }