[llvm-commits] [vector_llvm] CVS: llvm/examples/SIMD/MADFilter/Makefile mad.h mad_filter.altivec.handwritten.c mad_filter.sse.handwritten.c mad_filter.vectorc.c main.c
Robert L. Bocchino Jr.
bocchino at persephone.cs.uiuc.edu
Sun Oct 23 15:50:20 PDT 2005
Changes in directory llvm/examples/SIMD/MADFilter:
Makefile added (r1.1.2.1)
mad.h added (r1.1.2.1)
mad_filter.altivec.handwritten.c added (r1.1.2.1)
mad_filter.sse.handwritten.c added (r1.1.2.1)
mad_filter.vectorc.c added (r1.1.2.1)
main.c added (r1.1.2.1)
---
Log message:
Examples to illustrate Vector LLVM's SIMD support.
---
Diffs of the changes: (+1109 -0)
Makefile | 4
mad.h | 932 +++++++++++++++++++++++++++++++++++++++
mad_filter.altivec.handwritten.c | 15
mad_filter.sse.handwritten.c | 16
mad_filter.vectorc.c | 13
main.c | 129 +++++
6 files changed, 1109 insertions
Index: llvm/examples/SIMD/MADFilter/Makefile
diff -c /dev/null llvm/examples/SIMD/MADFilter/Makefile:1.1.2.1
*** /dev/null Sun Oct 23 17:50:00 2005
--- llvm/examples/SIMD/MADFilter/Makefile Sun Oct 23 17:49:40 2005
***************
*** 0 ****
--- 1,4 ----
+ NAME= mad_filter
+
+ include ../Makefile.common
+
Index: llvm/examples/SIMD/MADFilter/mad.h
diff -c /dev/null llvm/examples/SIMD/MADFilter/mad.h:1.1.2.1
*** /dev/null Sun Oct 23 17:50:17 2005
--- llvm/examples/SIMD/MADFilter/mad.h Sun Oct 23 17:49:40 2005
***************
*** 0 ****
--- 1,932 ----
+ /*
+ * libmad - MPEG audio decoder library
+ * Copyright (C) 2000-2001 Robert Leslie
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program 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 General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * If you would like to negotiate alternate licensing terms, you may do
+ * so by contacting the author: Robert Leslie <rob at mars.org>
+ */
+
+ # ifdef __cplusplus
+ extern "C" {
+ # endif
+
+ # define FPM_INTEL
+
+ # define SIZEOF_INT 4
+ # define SIZEOF_LONG 4
+ # define SIZEOF_LONG_LONG 8
+
+ /* Id: version.h,v 1.20 2001/10/27 22:47:32 rob Exp */
+
+ # ifndef LIBMAD_VERSION_H
+ # define LIBMAD_VERSION_H
+
+ # define MAD_VERSION_MAJOR 0
+ # define MAD_VERSION_MINOR 14
+ # define MAD_VERSION_PATCH 2
+ # define MAD_VERSION_EXTRA " (beta)"
+
+ # define MAD_VERSION_STRINGIZE(str) #str
+ # define MAD_VERSION_STRING(num) MAD_VERSION_STRINGIZE(num)
+
+ # define MAD_VERSION MAD_VERSION_STRING(MAD_VERSION_MAJOR) "." \
+ MAD_VERSION_STRING(MAD_VERSION_MINOR) "." \
+ MAD_VERSION_STRING(MAD_VERSION_PATCH) \
+ MAD_VERSION_EXTRA
+
+ # define MAD_PUBLISHYEAR "2000-2001"
+ # define MAD_AUTHOR "Robert Leslie"
+ # define MAD_EMAIL "rob at mars.org"
+
+ extern char const mad_version[];
+ extern char const mad_copyright[];
+ extern char const mad_author[];
+ extern char const mad_build[];
+
+ # endif
+
+ /* Id: fixed.h,v 1.30 2001/11/02 09:51:06 rob Exp */
+
+ # ifndef LIBMAD_FIXED_H
+ # define LIBMAD_FIXED_H
+
+ # if SIZEOF_INT >= 4
+ typedef signed int mad_fixed_t;
+
+ typedef signed int mad_fixed64hi_t;
+ typedef unsigned int mad_fixed64lo_t;
+ # else
+ typedef signed long mad_fixed_t;
+
+ typedef signed long mad_fixed64hi_t;
+ typedef unsigned long mad_fixed64lo_t;
+ # endif
+
+ # if defined(_MSC_VER)
+ # define mad_fixed64_t signed __int64
+ # elif 1 || defined(__GNUC__)
+ # define mad_fixed64_t signed long long
+ # endif
+
+ # if defined(FPM_FLOAT)
+ typedef double mad_sample_t;
+ # else
+ typedef mad_fixed_t mad_sample_t;
+ # endif
+
+ /*
+ * Fixed-point format: 0xABBBBBBB
+ * A == whole part (sign + 3 bits)
+ * B == fractional part (28 bits)
+ *
+ * Values are signed two's complement, so the effective range is:
+ * 0x80000000 to 0x7fffffff
+ * -8.0 to +7.9999999962747097015380859375
+ *
+ * The smallest representable value is:
+ * 0x00000001 == 0.0000000037252902984619140625 (i.e. about 3.725e-9)
+ *
+ * 28 bits of fractional accuracy represent about
+ * 8.6 digits of decimal accuracy.
+ *
+ * Fixed-point numbers can be added or subtracted as normal
+ * integers, but multiplication requires shifting the 64-bit result
+ * from 56 fractional bits back to 28 (and rounding.)
+ *
+ * Changing the definition of MAD_F_FRACBITS is only partially
+ * supported, and must be done with care.
+ */
+
+ # define MAD_F_FRACBITS 28
+
+ # if MAD_F_FRACBITS == 28
+ # define MAD_F(x) ((mad_fixed_t) (x##L))
+ # else
+ # if MAD_F_FRACBITS < 28
+ # warning "MAD_F_FRACBITS < 28"
+ # define MAD_F(x) ((mad_fixed_t) \
+ (((x##L) + \
+ (1L << (28 - MAD_F_FRACBITS - 1))) >> \
+ (28 - MAD_F_FRACBITS)))
+ # elif MAD_F_FRACBITS > 28
+ # error "MAD_F_FRACBITS > 28 not currently supported"
+ # define MAD_F(x) ((mad_fixed_t) \
+ ((x##L) << (MAD_F_FRACBITS - 28)))
+ # endif
+ # endif
+
+ # define MAD_F_MIN ((mad_fixed_t) -0x80000000L)
+ # define MAD_F_MAX ((mad_fixed_t) +0x7fffffffL)
+
+ # define MAD_F_ONE MAD_F(0x10000000)
+
+ # define mad_f_tofixed(x) ((mad_fixed_t) \
+ ((x) * (double) (1L << MAD_F_FRACBITS) + 0.5))
+ # define mad_f_todouble(x) ((double) \
+ ((x) / (double) (1L << MAD_F_FRACBITS)))
+
+ # define mad_f_intpart(x) ((x) >> MAD_F_FRACBITS)
+ # define mad_f_fracpart(x) ((x) & ((1L << MAD_F_FRACBITS) - 1))
+ /* (x should be positive) */
+
+ # define mad_f_fromint(x) ((x) << MAD_F_FRACBITS)
+
+ # define mad_f_add(x, y) ((x) + (y))
+ # define mad_f_sub(x, y) ((x) - (y))
+
+ # if defined(FPM_FLOAT)
+ # error "FPM_FLOAT not yet supported"
+
+ # undef MAD_F
+ # define MAD_F(x) mad_f_todouble(x)
+
+ # define mad_f_mul(x, y) ((x) * (y))
+ # define mad_f_scale64
+
+ # undef ASO_ZEROCHECK
+
+ # elif defined(FPM_64BIT)
+
+ /*
+ * This version should be the most accurate if 64-bit types are supported by
+ * the compiler, although it may not be the most efficient.
+ */
+ # if defined(OPT_ACCURACY)
+ # define mad_f_mul(x, y) \
+ ((mad_fixed_t) \
+ ((((mad_fixed64_t) (x) * (y)) + \
+ (1L << (MAD_F_SCALEBITS - 1))) >> MAD_F_SCALEBITS))
+ # else
+ # define mad_f_mul(x, y) \
+ ((mad_fixed_t) (((mad_fixed64_t) (x) * (y)) >> MAD_F_SCALEBITS))
+ # endif
+
+ # define MAD_F_SCALEBITS MAD_F_FRACBITS
+
+ /* --- Intel --------------------------------------------------------------- */
+
+ # elif defined(FPM_INTEL)
+
+ # if defined(_MSC_VER)
+ # pragma warning(push)
+ # pragma warning(disable: 4035) /* no return value */
+ static __forceinline
+ mad_fixed_t mad_f_mul_inline(mad_fixed_t x, mad_fixed_t y)
+ {
+ enum {
+ fracbits = MAD_F_FRACBITS
+ };
+
+ __asm {
+ mov eax, x
+ imul y
+ shrd eax, edx, fracbits
+ }
+
+ /* implicit return of eax */
+ }
+ # pragma warning(pop)
+
+ # define mad_f_mul mad_f_mul_inline
+ # define mad_f_scale64
+ # else
+ /*
+ * This Intel version is fast and accurate; the disposition of the least
+ * significant bit depends on OPT_ACCURACY via mad_f_scale64().
+ */
+ # define MAD_F_MLX(hi, lo, x, y) \
+ asm ("imull %3" \
+ : "=a" (lo), "=d" (hi) \
+ : "%a" (x), "rm" (y) \
+ : "cc")
+
+ # if defined(OPT_ACCURACY)
+ /*
+ * This gives best accuracy but is not very fast.
+ */
+ # define MAD_F_MLA(hi, lo, x, y) \
+ ({ mad_fixed64hi_t __hi; \
+ mad_fixed64lo_t __lo; \
+ MAD_F_MLX(__hi, __lo, (x), (y)); \
+ asm ("addl %2,%0\n\t" \
+ "adcl %3,%1" \
+ : "=rm" (lo), "=rm" (hi) \
+ : "r" (__lo), "r" (__hi), "0" (lo), "1" (hi) \
+ : "cc"); \
+ })
+ # endif /* OPT_ACCURACY */
+
+ # if defined(OPT_ACCURACY)
+ /*
+ * Surprisingly, this is faster than SHRD followed by ADC.
+ */
+ # define mad_f_scale64(hi, lo) \
+ ({ mad_fixed64hi_t __hi_; \
+ mad_fixed64lo_t __lo_; \
+ mad_fixed_t __result; \
+ asm ("addl %4,%2\n\t" \
+ "adcl %5,%3" \
+ : "=rm" (__lo_), "=rm" (__hi_) \
+ : "0" (lo), "1" (hi), \
+ "ir" (1L << (MAD_F_SCALEBITS - 1)), "ir" (0) \
+ : "cc"); \
+ asm ("shrdl %3,%2,%1" \
+ : "=rm" (__result) \
+ : "0" (__lo_), "r" (__hi_), "I" (MAD_F_SCALEBITS) \
+ : "cc"); \
+ __result; \
+ })
+ # else
+ # define mad_f_scale64(hi, lo) \
+ ({ mad_fixed_t __result; \
+ asm ("shrdl %3,%2,%1" \
+ : "=rm" (__result) \
+ : "0" (lo), "r" (hi), "I" (MAD_F_SCALEBITS) \
+ : "cc"); \
+ __result; \
+ })
+ # endif /* OPT_ACCURACY */
+
+ # define MAD_F_SCALEBITS MAD_F_FRACBITS
+ # endif
+
+ /* --- ARM ----------------------------------------------------------------- */
+
+ # elif defined(FPM_ARM)
+
+ /*
+ * This ARM V4 version is as accurate as FPM_64BIT but much faster. The
+ * least significant bit is properly rounded at no CPU cycle cost!
+ */
+ # if 1
+ /*
+ * There's a bug somewhere, possibly in the compiler, that sometimes makes
+ * this necessary instead of the default implementation via MAD_F_MLX and
+ * mad_f_scale64. It may be related to the use (or lack) of
+ * -finline-functions and/or -fstrength-reduce.
+ *
+ * This is also apparently faster than MAD_F_MLX/mad_f_scale64.
+ */
+ # define mad_f_mul(x, y) \
+ ({ mad_fixed64hi_t __hi; \
+ mad_fixed64lo_t __lo; \
+ mad_fixed_t __result; \
+ asm ("smull %0, %1, %3, %4\n\t" \
+ "movs %0, %0, lsr %5\n\t" \
+ "adc %2, %0, %1, lsl %6" \
+ : "=&r" (__lo), "=&r" (__hi), "=r" (__result) \
+ : "%r" (x), "r" (y), \
+ "M" (MAD_F_SCALEBITS), "M" (32 - MAD_F_SCALEBITS) \
+ : "cc"); \
+ __result; \
+ })
+ # endif
+
+ # define MAD_F_MLX(hi, lo, x, y) \
+ asm ("smull %0, %1, %2, %3" \
+ : "=&r" (lo), "=&r" (hi) \
+ : "%r" (x), "r" (y))
+
+ # define MAD_F_MLA(hi, lo, x, y) \
+ asm ("smlal %0, %1, %2, %3" \
+ : "+r" (lo), "+r" (hi) \
+ : "%r" (x), "r" (y))
+
+ # define MAD_F_MLN(hi, lo) \
+ asm ("rsbs %0, %2, #0\n\t" \
+ "rsc %1, %3, #0" \
+ : "=r" (lo), "=r" (hi) \
+ : "0" (lo), "1" (hi) \
+ : "cc")
+
+ # define mad_f_scale64(hi, lo) \
+ ({ mad_fixed_t __result; \
+ asm ("movs %0, %1, lsr %3\n\t" \
+ "adc %0, %0, %2, lsl %4" \
+ : "=r" (__result) \
+ : "r" (lo), "r" (hi), \
+ "M" (MAD_F_SCALEBITS), "M" (32 - MAD_F_SCALEBITS) \
+ : "cc"); \
+ __result; \
+ })
+
+ # define MAD_F_SCALEBITS MAD_F_FRACBITS
+
+ /* --- MIPS ---------------------------------------------------------------- */
+
+ # elif defined(FPM_MIPS)
+
+ /*
+ * This MIPS version is fast and accurate; the disposition of the least
+ * significant bit depends on OPT_ACCURACY via mad_f_scale64().
+ */
+ # define MAD_F_MLX(hi, lo, x, y) \
+ asm ("mult %2,%3" \
+ : "=l" (lo), "=h" (hi) \
+ : "%r" (x), "r" (y))
+
+ # if defined(HAVE_MADD_ASM)
+ # define MAD_F_MLA(hi, lo, x, y) \
+ asm ("madd %2,%3" \
+ : "+l" (lo), "+h" (hi) \
+ : "%r" (x), "r" (y))
+ # elif defined(HAVE_MADD16_ASM)
+ /*
+ * This loses significant accuracy due to the 16-bit integer limit in the
+ * multiply/accumulate instruction.
+ */
+ # define MAD_F_ML0(hi, lo, x, y) \
+ asm ("mult %2,%3" \
+ : "=l" (lo), "=h" (hi) \
+ : "%r" ((x) >> 12), "r" ((y) >> 16))
+ # define MAD_F_MLA(hi, lo, x, y) \
+ asm ("madd16 %2,%3" \
+ : "+l" (lo), "+h" (hi) \
+ : "%r" ((x) >> 12), "r" ((y) >> 16))
+ # define MAD_F_MLZ(hi, lo) ((mad_fixed_t) (lo))
+ # endif
+
+ # if defined(OPT_SPEED)
+ # define mad_f_scale64(hi, lo) \
+ ((mad_fixed_t) ((hi) << (32 - MAD_F_SCALEBITS)))
+ # define MAD_F_SCALEBITS MAD_F_FRACBITS
+ # endif
+
+ /* --- SPARC --------------------------------------------------------------- */
+
+ # elif defined(FPM_SPARC)
+
+ /*
+ * This SPARC V8 version is fast and accurate; the disposition of the least
+ * significant bit depends on OPT_ACCURACY via mad_f_scale64().
+ */
+ # define MAD_F_MLX(hi, lo, x, y) \
+ asm ("smul %2, %3, %0\n\t" \
+ "rd %%y, %1" \
+ : "=r" (lo), "=r" (hi) \
+ : "%r" (x), "rI" (y))
+
+ /* --- PowerPC ------------------------------------------------------------- */
+
+ # elif defined(FPM_PPC)
+
+ /*
+ * This PowerPC version is tuned for the 4xx embedded processors. It is
+ * effectively a tuned version of FPM_64BIT. It is a little faster and just
+ * as accurate. The disposition of the least significant bit depends on
+ * OPT_ACCURACY via mad_f_scale64().
+ */
+ # define MAD_F_MLX(hi, lo, x, y) \
+ asm ("mulhw %1, %2, %3\n\t" \
+ "mullw %0, %2, %3" \
+ : "=&r" (lo), "=&r" (hi) \
+ : "%r" (x), "r" (y))
+
+ # define MAD_F_MLA(hi, lo, x, y) \
+ ({ mad_fixed64hi_t __hi; \
+ mad_fixed64lo_t __lo; \
+ MAD_F_MLX(__hi, __lo, (x), (y)); \
+ asm ("addc %0, %2, %3\n\t" \
+ "adde %1, %4, %5" \
+ : "=r" (lo), "=r" (hi) \
+ : "%r" (__lo), "0" (lo), "%r" (__hi), "1" (hi)); \
+ })
+
+ # if defined(OPT_ACCURACY)
+ /*
+ * This is accurate and ~2 - 2.5 times slower than the unrounded version.
+ *
+ * The __volatile__ improves the generated code by another 5% (fewer spills
+ * to memory); eventually they should be removed.
+ */
+ # define mad_f_scale64(hi, lo) \
+ ({ mad_fixed_t __result; \
+ mad_fixed64hi_t __hi_; \
+ mad_fixed64lo_t __lo_; \
+ asm __volatile__ ("addc %0, %2, %4\n\t" \
+ "addze %1, %3" \
+ : "=r" (__lo_), "=r" (__hi_) \
+ : "r" (lo), "r" (hi), "r" (1 << (MAD_F_SCALEBITS - 1))); \
+ asm __volatile__ ("rlwinm %0, %2,32-%3,0,%3-1\n\t" \
+ "rlwimi %0, %1,32-%3,%3,31" \
+ : "=&r" (__result) \
+ : "r" (__lo_), "r" (__hi_), "I" (MAD_F_SCALEBITS)); \
+ __result; \
+ })
+ # else
+ # define mad_f_scale64(hi, lo) \
+ ({ mad_fixed_t __result; \
+ asm ("rlwinm %0, %2,32-%3,0,%3-1\n\t" \
+ "rlwimi %0, %1,32-%3,%3,31" \
+ : "=r" (__result) \
+ : "r" (lo), "r" (hi), "I" (MAD_F_SCALEBITS)); \
+ __result; \
+ })
+ # endif /* OPT_ACCURACY */
+
+ # define MAD_F_SCALEBITS MAD_F_FRACBITS
+
+ /* --- Default ------------------------------------------------------------- */
+
+ # elif defined(FPM_DEFAULT)
+
+ /*
+ * This version is the most portable but it loses significant accuracy.
+ * Furthermore, accuracy is biased against the second argument, so care
+ * should be taken when ordering operands.
+ *
+ * The scale factors are constant as this is not used with SSO.
+ *
+ * Pre-rounding is required to stay within the limits of compliance.
+ */
+ # if defined(OPT_SPEED)
+ # define mad_f_mul(x, y) (((x) >> 12) * ((y) >> 16))
+ # else
+ # define mad_f_mul(x, y) ((((x) + (1L << 11)) >> 12) * \
+ (((y) + (1L << 15)) >> 16))
+ # endif
+
+ /* ------------------------------------------------------------------------- */
+
+ # else
+ # error "no FPM selected"
+ # endif
+
+ /* default implementations */
+
+ # if !defined(mad_f_mul)
+ # define mad_f_mul(x, y) \
+ ({ mad_fixed64hi_t __hi; \
+ mad_fixed64lo_t __lo; \
+ MAD_F_MLX(__hi, __lo, (x), (y)); \
+ mad_f_scale64(__hi, __lo); \
+ })
+ # endif
+
+ # if !defined(MAD_F_MLA)
+ # define MAD_F_ML0(hi, lo, x, y) ((lo) = mad_f_mul((x), (y)))
+ # define MAD_F_MLA(hi, lo, x, y) ((lo) += mad_f_mul((x), (y)))
+ # define MAD_F_MLN(hi, lo) ((lo) = -(lo))
+ # define MAD_F_MLZ(hi, lo) ((void) (hi), (mad_fixed_t) (lo))
+ # endif
+
+ # if !defined(MAD_F_ML0)
+ # define MAD_F_ML0(hi, lo, x, y) MAD_F_MLX((hi), (lo), (x), (y))
+ # endif
+
+ # if !defined(MAD_F_MLN)
+ # define MAD_F_MLN(hi, lo) ((hi) = ((lo) = -(lo)) ? ~(hi) : -(hi))
+ # endif
+
+ # if !defined(MAD_F_MLZ)
+ # define MAD_F_MLZ(hi, lo) mad_f_scale64((hi), (lo))
+ # endif
+
+ # if !defined(mad_f_scale64)
+ # if defined(OPT_ACCURACY)
+ # define mad_f_scale64(hi, lo) \
+ ((((mad_fixed_t) \
+ (((hi) << (32 - (MAD_F_SCALEBITS - 1))) | \
+ ((lo) >> (MAD_F_SCALEBITS - 1)))) + 1) >> 1)
+ # else
+ # define mad_f_scale64(hi, lo) \
+ ((mad_fixed_t) \
+ (((hi) << (32 - MAD_F_SCALEBITS)) | \
+ ((lo) >> MAD_F_SCALEBITS)))
+ # endif
+ # define MAD_F_SCALEBITS MAD_F_FRACBITS
+ # endif
+
+ /* miscellaneous C routines */
+
+ mad_fixed_t mad_f_abs(mad_fixed_t);
+
+ # endif
+
+ /* Id: bit.h,v 1.8 2001/10/17 19:14:47 rob Exp */
+
+ # ifndef LIBMAD_BIT_H
+ # define LIBMAD_BIT_H
+
+ struct mad_bitptr {
+ unsigned char const *byte;
+ unsigned short cache;
+ unsigned short left;
+ };
+
+ void mad_bit_init(struct mad_bitptr *, unsigned char const *);
+
+ # define mad_bit_finish(bitptr) /* nothing */
+
+ unsigned int mad_bit_length(struct mad_bitptr const *,
+ struct mad_bitptr const *);
+
+ # define mad_bit_bitsleft(bitptr) ((bitptr)->left)
+ unsigned char const *mad_bit_nextbyte(struct mad_bitptr const *);
+
+ void mad_bit_skip(struct mad_bitptr *, unsigned int);
+ unsigned long mad_bit_read(struct mad_bitptr *, unsigned int);
+ void mad_bit_write(struct mad_bitptr *, unsigned int, unsigned long);
+
+ unsigned short mad_bit_crc(struct mad_bitptr, unsigned int, unsigned short);
+
+ # endif
+
+ /* Id: timer.h,v 1.12 2001/11/03 03:57:11 rob Exp */
+
+ # ifndef LIBMAD_TIMER_H
+ # define LIBMAD_TIMER_H
+
+ typedef struct {
+ signed long seconds; /* whole seconds */
+ unsigned long fraction; /* 1/MAD_TIMER_RESOLUTION seconds */
+ } mad_timer_t;
+
+ extern mad_timer_t const mad_timer_zero;
+
+ # define MAD_TIMER_RESOLUTION 352800000UL
+
+ enum mad_units {
+ MAD_UNITS_HOURS = -2,
+ MAD_UNITS_MINUTES = -1,
+ MAD_UNITS_SECONDS = 0,
+
+ /* metric units */
+
+ MAD_UNITS_DECISECONDS = 10,
+ MAD_UNITS_CENTISECONDS = 100,
+ MAD_UNITS_MILLISECONDS = 1000,
+
+ /* audio sample units */
+
+ MAD_UNITS_8000_HZ = 8000,
+ MAD_UNITS_11025_HZ = 11025,
+ MAD_UNITS_12000_HZ = 12000,
+
+ MAD_UNITS_16000_HZ = 16000,
+ MAD_UNITS_22050_HZ = 22050,
+ MAD_UNITS_24000_HZ = 24000,
+
+ MAD_UNITS_32000_HZ = 32000,
+ MAD_UNITS_44100_HZ = 44100,
+ MAD_UNITS_48000_HZ = 48000,
+
+ /* video frame/field units */
+
+ MAD_UNITS_24_FPS = 24,
+ MAD_UNITS_25_FPS = 25,
+ MAD_UNITS_30_FPS = 30,
+ MAD_UNITS_48_FPS = 48,
+ MAD_UNITS_50_FPS = 50,
+ MAD_UNITS_60_FPS = 60,
+
+ /* CD audio frames */
+
+ MAD_UNITS_75_FPS = 75,
+
+ /* video drop-frame units */
+
+ MAD_UNITS_23_976_FPS = -24,
+ MAD_UNITS_24_975_FPS = -25,
+ MAD_UNITS_29_97_FPS = -30,
+ MAD_UNITS_47_952_FPS = -48,
+ MAD_UNITS_49_95_FPS = -50,
+ MAD_UNITS_59_94_FPS = -60
+ };
+
+ # define mad_timer_reset(timer) ((void) (*(timer) = mad_timer_zero))
+
+ int mad_timer_compare(mad_timer_t, mad_timer_t);
+
+ # define mad_timer_sign(timer) mad_timer_compare((timer), mad_timer_zero)
+
+ void mad_timer_negate(mad_timer_t *);
+ mad_timer_t mad_timer_abs(mad_timer_t);
+
+ void mad_timer_set(mad_timer_t *, unsigned long, unsigned long, unsigned long);
+ void mad_timer_add(mad_timer_t *, mad_timer_t);
+ void mad_timer_multiply(mad_timer_t *, signed long);
+
+ signed long mad_timer_count(mad_timer_t, enum mad_units);
+ unsigned long mad_timer_fraction(mad_timer_t, unsigned long);
+ void mad_timer_string(mad_timer_t, char *, char const *,
+ enum mad_units, enum mad_units, unsigned long);
+
+ # endif
+
+ /* Id: stream.h,v 1.15 2001/11/08 23:28:03 rob Exp */
+
+ # ifndef LIBMAD_STREAM_H
+ # define LIBMAD_STREAM_H
+
+ # define MAD_BUFFER_GUARD 8
+ # define MAD_BUFFER_MDLEN (511 + 2048 + MAD_BUFFER_GUARD)
+
+ enum mad_error {
+ MAD_ERROR_NONE = 0x0000, /* no error */
+
+ MAD_ERROR_BUFLEN = 0x0001, /* input buffer too small (or EOF) */
+ MAD_ERROR_BUFPTR = 0x0002, /* invalid (null) buffer pointer */
+
+ MAD_ERROR_NOMEM = 0x0031, /* not enough memory */
+
+ MAD_ERROR_LOSTSYNC = 0x0101, /* lost synchronization */
+ MAD_ERROR_BADLAYER = 0x0102, /* reserved header layer value */
+ MAD_ERROR_BADBITRATE = 0x0103, /* forbidden bitrate value */
+ MAD_ERROR_BADSAMPLERATE = 0x0104, /* reserved sample frequency value */
+ MAD_ERROR_BADEMPHASIS = 0x0105, /* reserved emphasis value */
+
+ MAD_ERROR_BADCRC = 0x0201, /* CRC check failed */
+ MAD_ERROR_BADBITALLOC = 0x0211, /* forbidden bit allocation value */
+ MAD_ERROR_BADSCALEFACTOR = 0x0221, /* bad scalefactor index */
+ MAD_ERROR_BADFRAMELEN = 0x0231, /* bad frame length */
+ MAD_ERROR_BADBIGVALUES = 0x0232, /* bad big_values count */
+ MAD_ERROR_BADBLOCKTYPE = 0x0233, /* reserved block_type */
+ MAD_ERROR_BADSCFSI = 0x0234, /* bad scalefactor selection info */
+ MAD_ERROR_BADDATAPTR = 0x0235, /* bad main_data_begin pointer */
+ MAD_ERROR_BADPART3LEN = 0x0236, /* bad audio data length */
+ MAD_ERROR_BADHUFFTABLE = 0x0237, /* bad Huffman table select */
+ MAD_ERROR_BADHUFFDATA = 0x0238, /* Huffman data overrun */
+ MAD_ERROR_BADSTEREO = 0x0239 /* incompatible block_type for JS */
+ };
+
+ # define MAD_RECOVERABLE(error) ((error) & 0xff00)
+
+ struct mad_stream {
+ unsigned char const *buffer; /* input bitstream buffer */
+ unsigned char const *bufend; /* end of buffer */
+ unsigned long skiplen; /* bytes to skip before next frame */
+
+ int sync; /* stream sync found */
+ unsigned long freerate; /* free bitrate (fixed) */
+
+ unsigned char const *this_frame; /* start of current frame */
+ unsigned char const *next_frame; /* start of next frame */
+ struct mad_bitptr ptr; /* current processing bit pointer */
+
+ struct mad_bitptr anc_ptr; /* ancillary bits pointer */
+ unsigned int anc_bitlen; /* number of ancillary bits */
+
+ unsigned char (*main_data)[MAD_BUFFER_MDLEN];
+ /* Layer III main_data() */
+ unsigned int md_len; /* bytes in main_data */
+
+ int options; /* decoding options (see below) */
+ enum mad_error error; /* error code (see above) */
+ };
+
+ enum {
+ MAD_OPTION_IGNORECRC = 0x0001, /* ignore CRC errors */
+ MAD_OPTION_HALFSAMPLERATE = 0x0002 /* generate PCM at 1/2 sample rate */
+ # if 0 /* not yet implemented */
+ MAD_OPTION_LEFTCHANNEL = 0x0010, /* decode left channel only */
+ MAD_OPTION_RIGHTCHANNEL = 0x0020, /* decode right channel only */
+ MAD_OPTION_SINGLECHANNEL = 0x0030 /* combine channels */
+ # endif
+ };
+
+ void mad_stream_init(struct mad_stream *);
+ void mad_stream_finish(struct mad_stream *);
+
+ # define mad_stream_options(stream, opts) \
+ ((void) ((stream)->options = (opts)))
+
+ void mad_stream_buffer(struct mad_stream *,
+ unsigned char const *, unsigned long);
+ void mad_stream_skip(struct mad_stream *, unsigned long);
+
+ int mad_stream_sync(struct mad_stream *);
+
+ char const *mad_stream_errorstr(struct mad_stream const *);
+
+ # endif
+
+ /* Id: frame.h,v 1.16 2001/10/17 19:13:41 rob Exp */
+
+ # ifndef LIBMAD_FRAME_H
+ # define LIBMAD_FRAME_H
+
+ enum mad_layer {
+ MAD_LAYER_I = 1, /* Layer I */
+ MAD_LAYER_II = 2, /* Layer II */
+ MAD_LAYER_III = 3 /* Layer III */
+ };
+
+ enum mad_mode {
+ MAD_MODE_SINGLE_CHANNEL = 0, /* single channel */
+ MAD_MODE_DUAL_CHANNEL = 1, /* dual channel */
+ MAD_MODE_JOINT_STEREO = 2, /* joint (MS/intensity) stereo */
+ MAD_MODE_STEREO = 3 /* normal LR stereo */
+ };
+
+ enum mad_emphasis {
+ MAD_EMPHASIS_NONE = 0, /* no emphasis */
+ MAD_EMPHASIS_50_15_US = 1, /* 50/15 microseconds emphasis */
+ MAD_EMPHASIS_CCITT_J_17 = 3 /* CCITT J.17 emphasis */
+ };
+
+ struct mad_header {
+ enum mad_layer layer; /* audio layer (1, 2, or 3) */
+ enum mad_mode mode; /* channel mode (see above) */
+ int mode_extension; /* additional mode info */
+ enum mad_emphasis emphasis; /* de-emphasis to use (see above) */
+
+ unsigned long bitrate; /* stream bitrate (bps) */
+ unsigned int samplerate; /* sampling frequency (Hz) */
+
+ unsigned short crc_check; /* frame CRC accumulator */
+ unsigned short crc_target; /* final target CRC checksum */
+
+ int flags; /* flags (see below) */
+ int private_bits; /* private bits (see below) */
+
+ mad_timer_t duration; /* audio playing time of frame */
+ };
+
+ struct mad_frame {
+ struct mad_header header; /* MPEG audio header */
+
+ int options; /* decoding options (from stream) */
+
+ mad_fixed_t sbsample[2][36][32]; /* synthesis subband filter samples */
+ mad_fixed_t (*overlap)[2][32][18]; /* Layer III block overlap data */
+ };
+
+ # define MAD_NCHANNELS(header) ((header)->mode ? 2 : 1)
+ # define MAD_NSBSAMPLES(header) \
+ ((header)->layer == MAD_LAYER_I ? 12 : \
+ (((header)->layer == MAD_LAYER_III && \
+ ((header)->flags & MAD_FLAG_LSF_EXT)) ? 18 : 36))
+
+ enum {
+ MAD_FLAG_NPRIVATE_III = 0x0007, /* number of Layer III private bits */
+ MAD_FLAG_INCOMPLETE = 0x0008, /* header but not data is decoded */
+
+ MAD_FLAG_PROTECTION = 0x0010, /* frame has CRC protection */
+ MAD_FLAG_COPYRIGHT = 0x0020, /* frame is copyright */
+ MAD_FLAG_ORIGINAL = 0x0040, /* frame is original (else copy) */
+ MAD_FLAG_PADDING = 0x0080, /* frame has additional slot */
+
+ MAD_FLAG_I_STEREO = 0x0100, /* uses intensity joint stereo */
+ MAD_FLAG_MS_STEREO = 0x0200, /* uses middle/side joint stereo */
+ MAD_FLAG_FREEFORMAT = 0x0400, /* uses free format bitrate */
+
+ MAD_FLAG_LSF_EXT = 0x1000, /* lower sampling freq. extension */
+ MAD_FLAG_MC_EXT = 0x2000, /* multichannel audio extension */
+ MAD_FLAG_MPEG_2_5_EXT = 0x4000 /* MPEG 2.5 (unofficial) extension */
+ };
+
+ enum {
+ MAD_PRIVATE_HEADER = 0x0100, /* header private bit */
+ MAD_PRIVATE_III = 0x001f /* Layer III private bits (up to 5) */
+ };
+
+ void mad_header_init(struct mad_header *);
+
+ # define mad_header_finish(header) /* nothing */
+
+ int mad_header_decode(struct mad_header *, struct mad_stream *);
+
+ void mad_frame_init(struct mad_frame *);
+ void mad_frame_finish(struct mad_frame *);
+
+ int mad_frame_decode(struct mad_frame *, struct mad_stream *);
+
+ void mad_frame_mute(struct mad_frame *);
+
+ # endif
+
+ /* Id: synth.h,v 1.11 2001/11/08 23:28:03 rob Exp */
+
+ # ifndef LIBMAD_SYNTH_H
+ # define LIBMAD_SYNTH_H
+
+ struct mad_pcm {
+ unsigned int samplerate; /* sampling frequency (Hz) */
+ unsigned short channels; /* number of channels */
+ unsigned short length; /* number of samples per channel */
+ mad_fixed_t samples[2][1152]; /* PCM output samples [ch][sample] */
+ };
+
+ struct mad_synth {
+ mad_fixed_t filter[2][2][2][16][8]; /* polyphase filterbank outputs */
+ /* [ch][eo][peo][s][v] */
+
+ unsigned int phase; /* current processing phase */
+
+ struct mad_pcm pcm; /* PCM output */
+ };
+
+ /* single channel PCM selector */
+ enum {
+ MAD_PCM_CHANNEL_SINGLE = 0
+ };
+
+ /* dual channel PCM selector */
+ enum {
+ MAD_PCM_CHANNEL_DUAL_1 = 0,
+ MAD_PCM_CHANNEL_DUAL_2 = 1
+ };
+
+ /* stereo PCM selector */
+ enum {
+ MAD_PCM_CHANNEL_STEREO_LEFT = 0,
+ MAD_PCM_CHANNEL_STEREO_RIGHT = 1
+ };
+
+ void mad_synth_init(struct mad_synth *);
+
+ # define mad_synth_finish(synth) /* nothing */
+
+ void mad_synth_mute(struct mad_synth *);
+
+ void mad_synth_frame(struct mad_synth *, struct mad_frame const *);
+
+ # endif
+
+ /* Id: decoder.h,v 1.13 2001/11/03 03:57:11 rob Exp */
+
+ # ifndef LIBMAD_DECODER_H
+ # define LIBMAD_DECODER_H
+
+ enum mad_decoder_mode {
+ MAD_DECODER_MODE_SYNC = 0,
+ MAD_DECODER_MODE_ASYNC
+ };
+
+ enum mad_flow {
+ MAD_FLOW_CONTINUE = 0x0000, /* continue normally */
+ MAD_FLOW_STOP = 0x0010, /* stop decoding normally */
+ MAD_FLOW_BREAK = 0x0011, /* stop decoding and signal an error */
+ MAD_FLOW_IGNORE = 0x0020 /* ignore the current frame */
+ };
+
+ struct mad_decoder {
+ enum mad_decoder_mode mode;
+
+ int options;
+
+ struct {
+ long pid;
+ int in;
+ int out;
+ } async;
+
+ struct {
+ struct mad_stream stream;
+ struct mad_frame frame;
+ struct mad_synth synth;
+ } *sync;
+
+ void *cb_data;
+
+ enum mad_flow (*input_func)(void *, struct mad_stream *);
+ enum mad_flow (*header_func)(void *, struct mad_header const *);
+ enum mad_flow (*filter_func)(void *,
+ struct mad_stream const *, struct mad_frame *);
+ enum mad_flow (*output_func)(void *,
+ struct mad_header const *, struct mad_pcm *);
+ enum mad_flow (*error_func)(void *, struct mad_stream *, struct mad_frame *);
+ enum mad_flow (*message_func)(void *, void *, unsigned int *);
+ };
+
+ void mad_decoder_init(struct mad_decoder *, void *,
+ enum mad_flow (*)(void *, struct mad_stream *),
+ enum mad_flow (*)(void *, struct mad_header const *),
+ enum mad_flow (*)(void *,
+ struct mad_stream const *,
+ struct mad_frame *),
+ enum mad_flow (*)(void *,
+ struct mad_header const *,
+ struct mad_pcm *),
+ enum mad_flow (*)(void *,
+ struct mad_stream *,
+ struct mad_frame *),
+ enum mad_flow (*)(void *, void *, unsigned int *));
+ int mad_decoder_finish(struct mad_decoder *);
+
+ # define mad_decoder_options(decoder, opts) \
+ ((void) ((decoder)->options = (opts)))
+
+ int mad_decoder_run(struct mad_decoder *, enum mad_decoder_mode);
+ int mad_decoder_message(struct mad_decoder *, void *, unsigned int *);
+
+ # endif
+
+ # ifdef __cplusplus
+ }
+ # endif
Index: llvm/examples/SIMD/MADFilter/mad_filter.altivec.handwritten.c
diff -c /dev/null llvm/examples/SIMD/MADFilter/mad_filter.altivec.handwritten.c:1.1.2.1
*** /dev/null Sun Oct 23 17:50:18 2005
--- llvm/examples/SIMD/MADFilter/mad_filter.altivec.handwritten.c Sun Oct 23 17:49:40 2005
***************
*** 0 ****
--- 1,15 ----
+ void experimental_filter_vector(short *left_ch, short *right_ch, unsigned n) {
+ vector signed short *left_vp = (vector signed short*) left_ch;
+ vector signed short *right_vp = (vector signed short*) right_ch;
+ unsigned i;
+ vector unsigned short two = (vector unsigned short) (2);
+
+ for (i = 0; i < n/8; ++i) {
+ vector signed short left = left_vp[i];
+ vector signed short right = right_vp[i];
+ vector signed short left_sub = vec_sub(left, vec_sra(right, two));
+ vector signed short right_sub = vec_sub(right, vec_sra(left, two));
+ right_vp[i] = right_sub;
+ left_vp[i] = left_sub;
+ }
+ }
Index: llvm/examples/SIMD/MADFilter/mad_filter.sse.handwritten.c
diff -c /dev/null llvm/examples/SIMD/MADFilter/mad_filter.sse.handwritten.c:1.1.2.1
*** /dev/null Sun Oct 23 17:50:18 2005
--- llvm/examples/SIMD/MADFilter/mad_filter.sse.handwritten.c Sun Oct 23 17:49:40 2005
***************
*** 0 ****
--- 1,16 ----
+ #include "SSE.h"
+
+ void experimental_filter_vector(short *left_ch, short *right_ch, unsigned n) {
+ unsigned int i, j;
+ __m128i *left_vp = (__m128i*) left_ch;
+ __m128i *right_vp = (__m128i*) right_ch;
+
+ for (i = 0; i < n/8; ++i) {
+ __m128i left = left_vp[i];
+ __m128i right = right_vp[i];
+ __m128i left_sub = _mm_sub_epi16(left, _mm_srai_epi16(right, 2));
+ __m128i right_sub = _mm_sub_epi16(right, _mm_srai_epi16(left, 2));
+ right_vp[i] = right_sub;
+ left_vp[i] = left_sub;
+ }
+ }
Index: llvm/examples/SIMD/MADFilter/mad_filter.vectorc.c
diff -c /dev/null llvm/examples/SIMD/MADFilter/mad_filter.vectorc.c:1.1.2.1
*** /dev/null Sun Oct 23 17:50:18 2005
--- llvm/examples/SIMD/MADFilter/mad_filter.vectorc.c Sun Oct 23 17:49:40 2005
***************
*** 0 ****
--- 1,13 ----
+ #include "VectorC.h"
+
+ void experimental_filter_vector(short *left_ch, short *right_ch, unsigned n) {
+ unsigned i;
+ for (i = 0; i < n/8; ++i) {
+ short left = vllvm_load_short(left_ch, 8, i);
+ short right = vllvm_load_short(right_ch, 8, i);
+ short left_sub = left - (right >> 2);
+ short right_sub = right - (left >> 2);
+ vllvm_store_short(right_sub, right_ch, i);
+ vllvm_store_short(left_sub, left_ch, i);
+ }
+ }
Index: llvm/examples/SIMD/MADFilter/main.c
diff -c /dev/null llvm/examples/SIMD/MADFilter/main.c:1.1.2.1
*** /dev/null Sun Oct 23 17:50:18 2005
--- llvm/examples/SIMD/MADFilter/main.c Sun Oct 23 17:49:41 2005
***************
*** 0 ****
--- 1,129 ----
+ /*
+ * This program adapted from
+ * mad - MPEG audio decoder
+ * Copyright (C) 2000-2001 Robert Leslie
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program 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 General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * $Id: main.c,v 1.1.2.1 2005/10/23 22:49:41 bocchino Exp $
+ */
+
+ # include <stdlib.h>
+ # include <stdio.h>
+
+ # include "mad.h"
+ #include "../_malloc.h"
+ #include <sys/time.h>
+ #include <sys/times.h>
+
+ void experimental_filter_scalar(short*, short*, unsigned);
+ void experimental_filter_vector(short*, short*, unsigned);
+ void init_channel(short*, unsigned);
+ int compare_channels(short *ch1, short *ch2, unsigned);
+
+ short *scalar_left, *scalar_right;
+ short *vector_left, *vector_right;
+
+ #define CHANNEL_SIZE 1152
+ #define ARRAY_SIZE CHANNEL_SIZE*sizeof(short)
+
+ void run(long *scalar_time, long *vector_time) {
+ struct tms buf_s, buf_e;
+ unsigned i;
+
+ scalar_left = _malloc(ARRAY_SIZE);
+ scalar_right = _malloc(ARRAY_SIZE);
+ vector_left = _malloc(ARRAY_SIZE);
+ vector_right = _malloc(ARRAY_SIZE);
+
+ init_channel(scalar_left, CHANNEL_SIZE);
+ init_channel(scalar_right, CHANNEL_SIZE);
+
+ times(&buf_s);
+ for (i = 0; i < 1000000; ++i)
+ experimental_filter_scalar(scalar_left, scalar_right, CHANNEL_SIZE);
+ times(&buf_e);
+ *scalar_time = buf_e.tms_utime - buf_s.tms_utime;
+ printf("scalar time=%d, ", *scalar_time);
+
+ init_channel(vector_left, CHANNEL_SIZE);
+ init_channel(vector_right, CHANNEL_SIZE);
+
+ times(&buf_s);
+ for (i = 0; i < 1000000; ++i)
+ experimental_filter_vector(vector_left, vector_right, CHANNEL_SIZE);
+ times(&buf_e);
+ *vector_time = buf_e.tms_utime - buf_s.tms_utime;
+ printf("vector time=%d, ", *vector_time);
+
+ float speedup = (float) *scalar_time / *vector_time;
+ printf("speedup=%f\n", speedup);
+
+ if (!compare_channels(scalar_left, vector_left, CHANNEL_SIZE) ||
+ !compare_channels(scalar_right, vector_right, CHANNEL_SIZE)) {
+ printf("FAILED\n");
+ exit(1);
+ }
+
+ }
+
+ int
+ main (void) {
+ unsigned i;
+
+ long best_scalar = -1, best_vector = -1;
+ long scalar, vector;
+ for (i = 0; i < NRUNS; ++i) {
+ run (&scalar, &vector);
+ if (best_scalar < 0 || best_scalar > scalar)
+ best_scalar = scalar;
+ if (best_vector < 0 || best_vector > vector)
+ best_vector = vector;
+ }
+
+ printf("best scalar=%d, ", best_scalar);
+ printf("best vector=%d, ", best_vector);
+ printf("speedup=%f\n", ((float) best_scalar)/best_vector);
+ printf ("PASSED\n");
+ return 0;
+ }
+
+ void init_channel(short *channel, unsigned n) {
+ unsigned i, j;
+ for (i = 0; i < n; ++i)
+ channel[i] = i;
+ }
+
+ int compare_channels(short *ch1, short *ch2, unsigned n) {
+ unsigned i, j;
+ for (i = 0; i < n; ++i)
+ if(ch1[i] != ch2[i])
+ return 0;
+ return 1;
+ }
+
+ void experimental_filter_scalar(short *left_ch, short *right_ch, unsigned n) {
+ unsigned int i, j;
+
+ for (i = 0; i < n; ++i) {
+ short left, right;
+
+ left = left_ch[i];
+ right = right_ch[i];
+ right_ch[i] -= left >> 2;
+ left_ch[i] -= right >> 2;
+ }
+ }
+
More information about the llvm-commits
mailing list