add

1 files changed, 205 insertions, 0 deletions

diff --git a/liboil/jpeg/zigzag8x8_s16.c b/liboil/jpeg/zigzag8x8_s16.c
new file mode 100644
index 0000000..f490ec7
--- /dev/null
+++ b/liboil/jpeg/zigzag8x8_s16.c
@@ -0,0 +1,205 @@
+/* liboil - Library of Optimized Inner Loops
+ * Copyright (C) 2003  David A. Schleef <ds@schleef.org>
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of version 2.1 of the GNU Lesser General
+ * Public License as published by the Free Software Foundation.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Library General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, write to the
+ * Free Software Foundation, Inc., 59 Temple Place, Suite 330,
+ * Boston, MA 02111-1307 USA.
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include <liboil/liboilfunction.h>
+#include <liboil/simdpack/simdpack.h>
+
+static const unsigned char zigzag_order[64] = {
+	0,
+	8, 1,
+	2, 9, 16,
+	24, 17, 10, 3, 
+	4, 11, 18, 25, 32,
+	40, 33, 26, 19, 12, 5, 
+	6, 13, 20, 27, 34, 41, 48,
+	56, 49, 42, 35, 28, 21, 14, 7,
+	15, 22, 29, 36, 43, 50, 57,
+	58, 51, 44, 37, 30, 23,
+	31, 38, 45, 52, 59,
+	60, 53, 46, 39,
+	47, 54, 61,
+	62, 55,
+	63
+};
+
+static void zigzag8x8_s16_ref(int16_t *dest, int16_t *src, int sstr)
+{
+	int i,j;
+	unsigned int z;
+
+	for(j=0;j<8;j++){
+		for(i=0;i<8;i++){
+			z = zigzag_order[j*8+i];
+			dest[j*8+i] = OIL_GET(src, sstr*(z>>3)+(z&7), int16_t);
+		}
+	}
+}
+OIL_DEFINE_IMPL (zigzag8x8_s16_ref, zigzag8x8_s16_class);
+
+
+#ifdef HAVE_CPU_POWERPC
+/* 00 indicates that the element can't be handled by vperm, and needs
+ * to be fixed up later. */
+u8 mangle[128] __attribute__ ((__aligned__ (16))) = {
+	00,00, 2, 3, 0, 1,16,17,18,19,20,21,00,00,00,00, /* 1, 2 */
+	 2, 3, 0, 1,16,17,18,19,20,21,22,23,24,25,00,00, /* 3, 4 */
+	 8, 9, 6, 7, 4, 5, 2, 3, 0, 1,16,17,18,19,20,21, /* 5, 6 */
+	 6, 7, 8, 9,10,11,12,13,30,31,28,29,26,27,24,25, /* 6, 7 */
+	 6, 7, 4, 5, 2, 3, 0, 1,18,19,20,21,22,23,24,25, /* 7, 0 */
+	10,11,12,13,14,15,30,31,28,29,26,27,24,25,22,23, /* 0, 1 */
+	00,00, 6, 7, 8, 9,10,11,12,13,14,15,30,31,28,29, /* 2, 3 */
+	00,00,00,00,10,11,12,13,14,15,30,31,28,29,00,00, /* 4, 5 */
+};
+
+/* IMPL zigzag8x8_s16_a16_altivec defined(SIMDPACK_USE_ALTIVEC) */
+SL_zigzag8x8_s16_storage
+void zigzag8x8_s16_a16_altivec(int16_t *dest, int16_t *src, int sstr)
+{
+	sl_altivec_load8_0(src,sstr);
+
+	/* "slide" vectors to right */
+	__asm__ __volatile__(
+		"vsldoi 0,0,0,0\n"
+		"vsldoi 1,1,1,14\n"
+		"vsldoi 2,2,2,12\n"
+		"vsldoi 3,3,3,10\n"
+		"vsldoi 4,4,4,8\n"
+		"vsldoi 5,5,5,6\n"
+		"vsldoi 6,6,6,4\n"
+		"vsldoi 7,7,7,2\n"
+	);
+
+	/* transpose */
+	__asm__ __volatile__(
+		"vmrghh    8, 0, 4\n"
+		"\tvmrglh  9, 0, 4\n"
+		"\tvmrghh 10, 1, 5\n"
+		"\tvmrglh 11, 1, 5\n"
+		"\tvmrghh 12, 2, 6\n"
+		"\tvmrglh 13, 2, 6\n"
+		"\tvmrghh 14, 3, 7\n"
+		"\tvmrglh 15, 3, 7\n"
+
+		"\tvmrghh 16,  8, 12\n"
+		"\tvmrglh 17,  8, 12\n"
+		"\tvmrghh 18,  9, 13\n"
+		"\tvmrglh 19,  9, 13\n"
+		"\tvmrghh 20, 10, 14\n"
+		"\tvmrglh 21, 10, 14\n"
+		"\tvmrghh 22, 11, 15\n"
+		"\tvmrglh 23, 11, 15\n"
+
+		"\tvmrghh  0, 16, 20\n"
+		"\tvmrglh  1, 16, 20\n"
+		"\tvmrghh  2, 17, 21\n"
+		"\tvmrglh  3, 17, 21\n"
+		"\tvmrghh  4, 18, 22\n"
+		"\tvmrglh  5, 18, 22\n"
+		"\tvmrghh  6, 19, 23\n"
+		"\tvmrglh  7, 19, 23\n"
+	);
+
+	sl_altivec_load8_8(mangle,16);
+
+	__asm__ __volatile__(
+		"\n"
+		"\tvperm 16,1,2,8\n"
+		"\tvperm 17,3,4,9\n"
+		"\tvperm 18,5,6,10\n"
+		"\tvperm 19,6,7,11\n"
+		"\tvperm 20,7,0,12\n"
+		"\tvperm 21,0,1,13\n"
+		"\tvperm 22,2,3,14\n"
+		"\tvperm 23,4,5,15\n"
+	);
+
+	sl_altivec_store8_16(dest,16);
+
+	/* fix up the elements that were missed */
+
+	block8x8_s16(dest,16,0,0) = block8x8_s16(src,sstr,0,0);
+	block8x8_s16(dest,16,0,6) = block8x8_s16(src,sstr,3,0);
+	block8x8_s16(dest,16,0,7) = block8x8_s16(src,sstr,2,1);
+	block8x8_s16(dest,16,1,7) = block8x8_s16(src,sstr,5,0);
+
+	block8x8_s16(dest,16,6,0) = block8x8_s16(src,sstr,2,7);
+	block8x8_s16(dest,16,7,0) = block8x8_s16(src,sstr,5,6);
+	block8x8_s16(dest,16,7,1) = block8x8_s16(src,sstr,4,7);
+	block8x8_s16(dest,16,7,7) = block8x8_s16(src,sstr,7,7);
+}
+#endif
+
+
+#ifdef TEST_zigzag8x8_s16
+int TEST_zigzag8x8_s16(void)
+{
+	int i;
+	int pass;
+	int failures = 0;
+	int16_t *src, *dest_ref, *dest_test;
+	struct sl_profile_struct t;
+
+#ifdef ALIGNED16
+	src = sl_malloc_s16_a16(64);
+	dest_ref = sl_malloc_s16_a16(64);
+	dest_test = sl_malloc_s16_a16(64);
+#else
+	src = sl_malloc_s16(64);
+	dest_ref = sl_malloc_s16(64);
+	dest_test = sl_malloc_s16(64);
+#endif
+	
+	sl_profile_init(t);
+	srand(20020326);
+
+	printf("I: " sl_stringify(zigzag8x8_s16_FUNC) "\n");
+
+	for(pass=0;pass<N_PASS;pass++){
+		for(i=0;i<64;i++)src[i] = sl_rand_s16();
+
+		zigzag8x8_s16_ref(dest_test, src, 16);
+		sl_profile_start(t);
+		zigzag8x8_s16_FUNC(dest_ref, src, 16);
+		sl_profile_stop(t);
+
+		for(i=0;i<64;i++){
+			if(dest_test[i]!=dest_ref[i]){
+				failures++;
+			}
+		}
+	}
+
+	sl_free(src);
+	sl_free(dest_ref);
+	sl_free(dest_test);
+
+	if(failures){
+		printf("E: %d failures\n",failures);
+	}
+
+	sl_profile_print(t);
+
+	return failures;
+}
+#endif
+
+