ryan/mozjpeg
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
19c791cdac6df84418c66eb9d67cfa9703bc2461
mozjpeg/jfdctint.c
DRC 19c791cdac Improve code formatting consistency 
With rare exceptions ...
- Always separate line continuation characters by one space from
  preceding code.
- Always use two-space indentation.  Never use tabs.
- Always use K&R-style conditional blocks.
- Always surround operators with spaces, except in raw assembly code.
- Always put a space after, but not before, a comma.
- Never put a space between type casts and variables/function calls.
- Never put a space between the function name and the argument list in
  function declarations and prototypes.
- Always surround braces ('{' and '}') with spaces.
- Always surround statements (if, for, else, catch, while, do, switch)
  with spaces.
- Always attach pointer symbols ('*' and '**') to the variable or
  function name.
- Always precede pointer symbols ('*' and '**') by a space in type
  casts.
- Use the MIN() macro from jpegint.h within the libjpeg and TurboJPEG
  API libraries (using min() from tjutil.h is still necessary for
  TJBench.)
- Where it makes sense (particularly in the TurboJPEG code), put a blank
  line after variable declaration blocks.
- Always separate statements in one-liners by two spaces.

The purpose of this was to ease maintenance on my part and also to make
it easier for contributors to figure out how to format patch
submissions.  This was admittedly confusing (even to me sometimes) when
we had 3 or 4 different style conventions in the same source tree.  The
new convention is more consistent with the formatting of other OSS code
bases.

This commit corrects deviations from the chosen formatting style in the
libjpeg API code and reformats the TurboJPEG API code such that it
conforms to the same standard.

NOTES:
- Although it is no longer necessary for the function name in function
  declarations to begin in Column 1 (this was historically necessary
  because of the ansi2knr utility, which allowed libjpeg to be built
  with non-ANSI compilers), we retain that formatting for the libjpeg
  code because it improves readability when using libjpeg's function
  attribute macros (GLOBAL(), etc.)
- This reformatting project was accomplished with the help of AStyle and
  Uncrustify, although neither was completely up to the task, and thus
  a great deal of manual tweaking was required.  Note to developers of
  code formatting utilities:  the libjpeg-turbo code base is an
  excellent test bed, because AFAICT, it breaks every single one of the
  utilities that are currently available.
- The legacy (MMX, SSE, 3DNow!) assembly code for i386 has been
  formatted to match the SSE2 code (refer to
  ff5685d5344273df321eb63a005eaae19d2496e3.)  I hadn't intended to
  bother with this, but the Loongson MMI implementation demonstrated
  that there is still academic value to the MMX implementation, as an
  algorithmic model for other 64-bit vector implementations.  Thus, it
  is desirable to improve its readability in the same manner as that of
  the SSE2 implementation.
2018-03-16 02:14:34 -05:00

289 lines
11 KiB
C
Raw Blame History

/*
* jfdctint.c
*
* This file was part of the Independent JPEG Group's software.
* Copyright (C) 1991-1996, Thomas G. Lane.
* libjpeg-turbo Modifications:
* Copyright (C) 2015, D. R. Commander.
* For conditions of distribution and use, see the accompanying README.ijg
* file.
*
* This file contains a slow-but-accurate integer implementation of the
* forward DCT (Discrete Cosine Transform).
*
* A 2-D DCT can be done by 1-D DCT on each row followed by 1-D DCT
* on each column. Direct algorithms are also available, but they are
* much more complex and seem not to be any faster when reduced to code.
*
* This implementation is based on an algorithm described in
* C. Loeffler, A. Ligtenberg and G. Moschytz, "Practical Fast 1-D DCT
* Algorithms with 11 Multiplications", Proc. Int'l. Conf. on Acoustics,
* Speech, and Signal Processing 1989 (ICASSP '89), pp. 988-991.
* The primary algorithm described there uses 11 multiplies and 29 adds.
* We use their alternate method with 12 multiplies and 32 adds.
* The advantage of this method is that no data path contains more than one
* multiplication; this allows a very simple and accurate implementation in
* scaled fixed-point arithmetic, with a minimal number of shifts.
*/
#define JPEG_INTERNALS
#include "jinclude.h"
#include "jpeglib.h"
#include "jdct.h" /* Private declarations for DCT subsystem */
#ifdef DCT_ISLOW_SUPPORTED
/*
* This module is specialized to the case DCTSIZE = 8.
*/
#if DCTSIZE != 8
Sorry, this code only copes with 8x8 DCTs. /* deliberate syntax err */
#endif
/*
* The poop on this scaling stuff is as follows:
*
* Each 1-D DCT step produces outputs which are a factor of sqrt(N)
* larger than the true DCT outputs. The final outputs are therefore
* a factor of N larger than desired; since N=8 this can be cured by
* a simple right shift at the end of the algorithm. The advantage of
* this arrangement is that we save two multiplications per 1-D DCT,
* because the y0 and y4 outputs need not be divided by sqrt(N).
* In the IJG code, this factor of 8 is removed by the quantization step
* (in jcdctmgr.c), NOT in this module.
*
* We have to do addition and subtraction of the integer inputs, which
* is no problem, and multiplication by fractional constants, which is
* a problem to do in integer arithmetic. We multiply all the constants
* by CONST_SCALE and convert them to integer constants (thus retaining
* CONST_BITS bits of precision in the constants). After doing a
* multiplication we have to divide the product by CONST_SCALE, with proper
* rounding, to produce the correct output. This division can be done
* cheaply as a right shift of CONST_BITS bits. We postpone shifting
* as long as possible so that partial sums can be added together with
* full fractional precision.
*
* The outputs of the first pass are scaled up by PASS1_BITS bits so that
* they are represented to better-than-integral precision. These outputs
* require BITS_IN_JSAMPLE + PASS1_BITS + 3 bits; this fits in a 16-bit word
* with the recommended scaling. (For 12-bit sample data, the intermediate
* array is JLONG anyway.)
*
* To avoid overflow of the 32-bit intermediate results in pass 2, we must
* have BITS_IN_JSAMPLE + CONST_BITS + PASS1_BITS <= 26. Error analysis
* shows that the values given below are the most effective.
*/
#if BITS_IN_JSAMPLE == 8
#define CONST_BITS 13
#define PASS1_BITS 2
#else
#define CONST_BITS 13
#define PASS1_BITS 1 /* lose a little precision to avoid overflow */
#endif
/* Some C compilers fail to reduce "FIX(constant)" at compile time, thus
* causing a lot of useless floating-point operations at run time.
* To get around this we use the following pre-calculated constants.
* If you change CONST_BITS you may want to add appropriate values.
* (With a reasonable C compiler, you can just rely on the FIX() macro...)
*/
#if CONST_BITS == 13
#define FIX_0_298631336 ((JLONG)2446) /* FIX(0.298631336) */
#define FIX_0_390180644 ((JLONG)3196) /* FIX(0.390180644) */
#define FIX_0_541196100 ((JLONG)4433) /* FIX(0.541196100) */
#define FIX_0_765366865 ((JLONG)6270) /* FIX(0.765366865) */
#define FIX_0_899976223 ((JLONG)7373) /* FIX(0.899976223) */
#define FIX_1_175875602 ((JLONG)9633) /* FIX(1.175875602) */
#define FIX_1_501321110 ((JLONG)12299) /* FIX(1.501321110) */
#define FIX_1_847759065 ((JLONG)15137) /* FIX(1.847759065) */
#define FIX_1_961570560 ((JLONG)16069) /* FIX(1.961570560) */
#define FIX_2_053119869 ((JLONG)16819) /* FIX(2.053119869) */
#define FIX_2_562915447 ((JLONG)20995) /* FIX(2.562915447) */
#define FIX_3_072711026 ((JLONG)25172) /* FIX(3.072711026) */
#else
#define FIX_0_298631336 FIX(0.298631336)
#define FIX_0_390180644 FIX(0.390180644)
#define FIX_0_541196100 FIX(0.541196100)
#define FIX_0_765366865 FIX(0.765366865)
#define FIX_0_899976223 FIX(0.899976223)
#define FIX_1_175875602 FIX(1.175875602)
#define FIX_1_501321110 FIX(1.501321110)
#define FIX_1_847759065 FIX(1.847759065)
#define FIX_1_961570560 FIX(1.961570560)
#define FIX_2_053119869 FIX(2.053119869)
#define FIX_2_562915447 FIX(2.562915447)
#define FIX_3_072711026 FIX(3.072711026)
#endif
/* Multiply an JLONG variable by an JLONG constant to yield an JLONG result.
* For 8-bit samples with the recommended scaling, all the variable
* and constant values involved are no more than 16 bits wide, so a
* 16x16->32 bit multiply can be used instead of a full 32x32 multiply.
* For 12-bit samples, a full 32-bit multiplication will be needed.
*/
#if BITS_IN_JSAMPLE == 8
#define MULTIPLY(var, const) MULTIPLY16C16(var, const)
#else
#define MULTIPLY(var, const) ((var) * (const))
#endif
/*
* Perform the forward DCT on one block of samples.
*/
GLOBAL(void)
jpeg_fdct_islow(DCTELEM *data)
{
JLONG tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7;
JLONG tmp10, tmp11, tmp12, tmp13;
JLONG z1, z2, z3, z4, z5;
DCTELEM *dataptr;
int ctr;
SHIFT_TEMPS
/* Pass 1: process rows. */
/* Note results are scaled up by sqrt(8) compared to a true DCT; */
/* furthermore, we scale the results by 2**PASS1_BITS. */
dataptr = data;
for (ctr = DCTSIZE - 1; ctr >= 0; ctr--) {
tmp0 = dataptr[0] + dataptr[7];
tmp7 = dataptr[0] - dataptr[7];
tmp1 = dataptr[1] + dataptr[6];
tmp6 = dataptr[1] - dataptr[6];
tmp2 = dataptr[2] + dataptr[5];
tmp5 = dataptr[2] - dataptr[5];
tmp3 = dataptr[3] + dataptr[4];
tmp4 = dataptr[3] - dataptr[4];
/* Even part per LL&M figure 1 --- note that published figure is faulty;
* rotator "sqrt(2)*c1" should be "sqrt(2)*c6".
*/
tmp10 = tmp0 + tmp3;
tmp13 = tmp0 - tmp3;
tmp11 = tmp1 + tmp2;
tmp12 = tmp1 - tmp2;
dataptr[0] = (DCTELEM)LEFT_SHIFT(tmp10 + tmp11, PASS1_BITS);
dataptr[4] = (DCTELEM)LEFT_SHIFT(tmp10 - tmp11, PASS1_BITS);
z1 = MULTIPLY(tmp12 + tmp13, FIX_0_541196100);
dataptr[2] = (DCTELEM)DESCALE(z1 + MULTIPLY(tmp13, FIX_0_765366865),
CONST_BITS - PASS1_BITS);
dataptr[6] = (DCTELEM)DESCALE(z1 + MULTIPLY(tmp12, -FIX_1_847759065),
CONST_BITS - PASS1_BITS);
/* Odd part per figure 8 --- note paper omits factor of sqrt(2).
* cK represents cos(K*pi/16).
* i0..i3 in the paper are tmp4..tmp7 here.
*/
z1 = tmp4 + tmp7;
z2 = tmp5 + tmp6;
z3 = tmp4 + tmp6;
z4 = tmp5 + tmp7;
z5 = MULTIPLY(z3 + z4, FIX_1_175875602); /* sqrt(2) * c3 */
tmp4 = MULTIPLY(tmp4, FIX_0_298631336); /* sqrt(2) * (-c1+c3+c5-c7) */
tmp5 = MULTIPLY(tmp5, FIX_2_053119869); /* sqrt(2) * ( c1+c3-c5+c7) */
tmp6 = MULTIPLY(tmp6, FIX_3_072711026); /* sqrt(2) * ( c1+c3+c5-c7) */
tmp7 = MULTIPLY(tmp7, FIX_1_501321110); /* sqrt(2) * ( c1+c3-c5-c7) */
z1 = MULTIPLY(z1, -FIX_0_899976223); /* sqrt(2) * ( c7-c3) */
z2 = MULTIPLY(z2, -FIX_2_562915447); /* sqrt(2) * (-c1-c3) */
z3 = MULTIPLY(z3, -FIX_1_961570560); /* sqrt(2) * (-c3-c5) */
z4 = MULTIPLY(z4, -FIX_0_390180644); /* sqrt(2) * ( c5-c3) */
z3 += z5;
z4 += z5;
dataptr[7] = (DCTELEM)DESCALE(tmp4 + z1 + z3, CONST_BITS - PASS1_BITS);
dataptr[5] = (DCTELEM)DESCALE(tmp5 + z2 + z4, CONST_BITS - PASS1_BITS);
dataptr[3] = (DCTELEM)DESCALE(tmp6 + z2 + z3, CONST_BITS - PASS1_BITS);
dataptr[1] = (DCTELEM)DESCALE(tmp7 + z1 + z4, CONST_BITS - PASS1_BITS);
dataptr += DCTSIZE; /* advance pointer to next row */
}
/* Pass 2: process columns.
* We remove the PASS1_BITS scaling, but leave the results scaled up
* by an overall factor of 8.
*/
dataptr = data;
for (ctr = DCTSIZE - 1; ctr >= 0; ctr--) {
tmp0 = dataptr[DCTSIZE * 0] + dataptr[DCTSIZE * 7];
tmp7 = dataptr[DCTSIZE * 0] - dataptr[DCTSIZE * 7];
tmp1 = dataptr[DCTSIZE * 1] + dataptr[DCTSIZE * 6];
tmp6 = dataptr[DCTSIZE * 1] - dataptr[DCTSIZE * 6];
tmp2 = dataptr[DCTSIZE * 2] + dataptr[DCTSIZE * 5];
tmp5 = dataptr[DCTSIZE * 2] - dataptr[DCTSIZE * 5];
tmp3 = dataptr[DCTSIZE * 3] + dataptr[DCTSIZE * 4];
tmp4 = dataptr[DCTSIZE * 3] - dataptr[DCTSIZE * 4];
/* Even part per LL&M figure 1 --- note that published figure is faulty;
* rotator "sqrt(2)*c1" should be "sqrt(2)*c6".
*/
tmp10 = tmp0 + tmp3;
tmp13 = tmp0 - tmp3;
tmp11 = tmp1 + tmp2;
tmp12 = tmp1 - tmp2;
dataptr[DCTSIZE * 0] = (DCTELEM)DESCALE(tmp10 + tmp11, PASS1_BITS);
dataptr[DCTSIZE * 4] = (DCTELEM)DESCALE(tmp10 - tmp11, PASS1_BITS);
z1 = MULTIPLY(tmp12 + tmp13, FIX_0_541196100);
dataptr[DCTSIZE * 2] =
(DCTELEM)DESCALE(z1 + MULTIPLY(tmp13, FIX_0_765366865),
CONST_BITS + PASS1_BITS);
dataptr[DCTSIZE * 6] =
(DCTELEM)DESCALE(z1 + MULTIPLY(tmp12, -FIX_1_847759065),
CONST_BITS + PASS1_BITS);
/* Odd part per figure 8 --- note paper omits factor of sqrt(2).
* cK represents cos(K*pi/16).
* i0..i3 in the paper are tmp4..tmp7 here.
*/
z1 = tmp4 + tmp7;
z2 = tmp5 + tmp6;
z3 = tmp4 + tmp6;
z4 = tmp5 + tmp7;
z5 = MULTIPLY(z3 + z4, FIX_1_175875602); /* sqrt(2) * c3 */
tmp4 = MULTIPLY(tmp4, FIX_0_298631336); /* sqrt(2) * (-c1+c3+c5-c7) */
tmp5 = MULTIPLY(tmp5, FIX_2_053119869); /* sqrt(2) * ( c1+c3-c5+c7) */
tmp6 = MULTIPLY(tmp6, FIX_3_072711026); /* sqrt(2) * ( c1+c3+c5-c7) */
tmp7 = MULTIPLY(tmp7, FIX_1_501321110); /* sqrt(2) * ( c1+c3-c5-c7) */
z1 = MULTIPLY(z1, -FIX_0_899976223); /* sqrt(2) * ( c7-c3) */
z2 = MULTIPLY(z2, -FIX_2_562915447); /* sqrt(2) * (-c1-c3) */
z3 = MULTIPLY(z3, -FIX_1_961570560); /* sqrt(2) * (-c3-c5) */
z4 = MULTIPLY(z4, -FIX_0_390180644); /* sqrt(2) * ( c5-c3) */
z3 += z5;
z4 += z5;
dataptr[DCTSIZE * 7] = (DCTELEM)DESCALE(tmp4 + z1 + z3,
CONST_BITS + PASS1_BITS);
dataptr[DCTSIZE * 5] = (DCTELEM)DESCALE(tmp5 + z2 + z4,
CONST_BITS + PASS1_BITS);
dataptr[DCTSIZE * 3] = (DCTELEM)DESCALE(tmp6 + z2 + z3,
CONST_BITS + PASS1_BITS);
dataptr[DCTSIZE * 1] = (DCTELEM)DESCALE(tmp7 + z1 + z4,
CONST_BITS + PASS1_BITS);
dataptr++; /* advance pointer to next column */
}
}
#endif /* DCT_ISLOW_SUPPORTED */
Reference in New Issue View Git Blame Copy Permalink