libjpeg-turbo/jdct.h

233 lines
9.4 KiB
C
Raw Normal View History

2013-12-09 20:29:04 +00:00
/*
* jdct.h
*
* Copyright (C) 1994-1996, Thomas G. Lane.
* This file is part of the Independent JPEG Group's software.
* For conditions of distribution and use, see the accompanying README file.
*
* This include file contains common declarations for the forward and
* inverse DCT modules. These declarations are private to the DCT managers
* (jcdctmgr.c, jddctmgr.c) and the individual DCT algorithms.
* The individual DCT algorithms are kept in separate files to ease
* machine-dependent tuning (e.g., assembly coding).
*/
/*
* A forward DCT routine is given a pointer to a work area of type DCTELEM[];
* the DCT is to be performed in-place in that buffer. Type DCTELEM is int
* for 8-bit samples, INT32 for 12-bit samples. (NOTE: Floating-point DCT
* implementations use an array of type FAST_FLOAT, instead.)
* The DCT inputs are expected to be signed (range +-CENTERJSAMPLE).
* The DCT outputs are returned scaled up by a factor of 8; they therefore
* have a range of +-8K for 8-bit data, +-128K for 12-bit data. This
* convention improves accuracy in integer implementations and saves some
* work in floating-point ones.
* Quantization of the output coefficients is done by jcdctmgr.c. This
* step requires an unsigned type and also one with twice the bits.
*/
#if BITS_IN_JSAMPLE == 8
#ifndef WITH_SIMD
typedef int DCTELEM; /* 16 or 32 bits is fine */
typedef unsigned int UDCTELEM;
typedef unsigned long long UDCTELEM2;
#else
typedef short DCTELEM; /* prefer 16 bit with SIMD for parellelism */
typedef unsigned short UDCTELEM;
typedef unsigned int UDCTELEM2;
#endif
#else
typedef INT32 DCTELEM; /* must have 32 bits */
typedef UINT32 UDCTELEM;
typedef unsigned long long UDCTELEM2;
#endif
/*
* An inverse DCT routine is given a pointer to the input JBLOCK and a pointer
* to an output sample array. The routine must dequantize the input data as
* well as perform the IDCT; for dequantization, it uses the multiplier table
* pointed to by compptr->dct_table. The output data is to be placed into the
* sample array starting at a specified column. (Any row offset needed will
* be applied to the array pointer before it is passed to the IDCT code.)
* Note that the number of samples emitted by the IDCT routine is
* DCT_scaled_size * DCT_scaled_size.
*/
/* typedef inverse_DCT_method_ptr is declared in jpegint.h */
/*
* Each IDCT routine has its own ideas about the best dct_table element type.
*/
typedef MULTIPLIER ISLOW_MULT_TYPE; /* short or int, whichever is faster */
#if BITS_IN_JSAMPLE == 8
typedef MULTIPLIER IFAST_MULT_TYPE; /* 16 bits is OK, use short if faster */
#define IFAST_SCALE_BITS 2 /* fractional bits in scale factors */
#else
typedef INT32 IFAST_MULT_TYPE; /* need 32 bits for scaled quantizers */
#define IFAST_SCALE_BITS 13 /* fractional bits in scale factors */
#endif
typedef FAST_FLOAT FLOAT_MULT_TYPE; /* preferred floating type */
/*
* Each IDCT routine is responsible for range-limiting its results and
* converting them to unsigned form (0..MAXJSAMPLE). The raw outputs could
* be quite far out of range if the input data is corrupt, so a bulletproof
* range-limiting step is required. We use a mask-and-table-lookup method
* to do the combined operations quickly. See the comments with
* prepare_range_limit_table (in jdmaster.c) for more info.
*/
#define IDCT_range_limit(cinfo) ((cinfo)->sample_range_limit + CENTERJSAMPLE)
#define RANGE_MASK (MAXJSAMPLE * 4 + 3) /* 2 bits wider than legal samples */
/* Short forms of external names for systems with brain-damaged linkers. */
#ifdef NEED_SHORT_EXTERNAL_NAMES
#define jpeg_fdct_islow jFDislow
#define jpeg_fdct_ifast jFDifast
#define jpeg_fdct_float jFDfloat
#define jpeg_idct_islow jRDislow
#define jpeg_idct_ifast jRDifast
#define jpeg_idct_float jRDfloat
#define jpeg_idct_7x7 jRD7x7
#define jpeg_idct_6x6 jRD6x6
#define jpeg_idct_5x5 jRD5x5
#define jpeg_idct_4x4 jRD4x4
#define jpeg_idct_3x3 jRD3x3
#define jpeg_idct_2x2 jRD2x2
#define jpeg_idct_1x1 jRD1x1
#define jpeg_idct_9x9 jRD9x9
#define jpeg_idct_10x10 jRD10x10
#define jpeg_idct_11x11 jRD11x11
#define jpeg_idct_12x12 jRD12x12
#define jpeg_idct_13x13 jRD13x13
#define jpeg_idct_14x14 jRD14x14
#define jpeg_idct_15x15 jRD15x15
#define jpeg_idct_16x16 jRD16x16
#endif /* NEED_SHORT_EXTERNAL_NAMES */
/* Extern declarations for the forward and inverse DCT routines. */
EXTERN(void) jpeg_fdct_islow JPP((DCTELEM * data));
EXTERN(void) jpeg_fdct_ifast JPP((DCTELEM * data));
EXTERN(void) jpeg_fdct_float JPP((FAST_FLOAT * data));
EXTERN(void) jpeg_idct_islow
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_ifast
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_float
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_7x7
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_6x6
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_5x5
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_4x4
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_3x3
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_2x2
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_1x1
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_9x9
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_10x10
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_11x11
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_12x12
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_13x13
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_14x14
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_15x15
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
EXTERN(void) jpeg_idct_16x16
JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col));
/*
* Macros for handling fixed-point arithmetic; these are used by many
* but not all of the DCT/IDCT modules.
*
* All values are expected to be of type INT32.
* Fractional constants are scaled left by CONST_BITS bits.
* CONST_BITS is defined within each module using these macros,
* and may differ from one module to the next.
*/
#define ONE ((INT32) 1)
#define CONST_SCALE (ONE << CONST_BITS)
/* Convert a positive real constant to an integer scaled by CONST_SCALE.
* Caution: some C compilers fail to reduce "FIX(constant)" at compile time,
* thus causing a lot of useless floating-point operations at run time.
*/
#define FIX(x) ((INT32) ((x) * CONST_SCALE + 0.5))
/* Descale and correctly round an INT32 value that's scaled by N bits.
* We assume RIGHT_SHIFT rounds towards minus infinity, so adding
* the fudge factor is correct for either sign of X.
*/
#define DESCALE(x,n) RIGHT_SHIFT((x) + (ONE << ((n)-1)), n)
/* Multiply an INT32 variable by an INT32 constant to yield an INT32 result.
* This macro is used only when the two inputs will actually be no more than
* 16 bits wide, so that a 16x16->32 bit multiply can be used instead of a
* full 32x32 multiply. This provides a useful speedup on many machines.
* Unfortunately there is no way to specify a 16x16->32 multiply portably
* in C, but some C compilers will do the right thing if you provide the
* correct combination of casts.
*/
#ifdef SHORTxSHORT_32 /* may work if 'int' is 32 bits */
#define MULTIPLY16C16(var,const) (((INT16) (var)) * ((INT16) (const)))
#endif
#ifdef SHORTxLCONST_32 /* known to work with Microsoft C 6.0 */
#define MULTIPLY16C16(var,const) (((INT16) (var)) * ((INT32) (const)))
#endif
#ifndef MULTIPLY16C16 /* default definition */
#define MULTIPLY16C16(var,const) ((var) * (const))
#endif
/* Same except both inputs are variables. */
#ifdef SHORTxSHORT_32 /* may work if 'int' is 32 bits */
#define MULTIPLY16V16(var1,var2) (((INT16) (var1)) * ((INT16) (var2)))
#endif
#ifndef MULTIPLY16V16 /* default definition */
#define MULTIPLY16V16(var1,var2) ((var1) * (var2))
#endif