jpeg_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr,
JCOEFPTR coef_block,
JSAMPARRAY output_buf, JDIMENSION output_col)
{
DCTELEM tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7;
DCTELEM tmp10, tmp11, tmp12, tmp13;
DCTELEM z5, z10, z11, z12, z13;
JCOEFPTR inptr;
IFAST_MULT_TYPE * quantptr;
int * wsptr;
JSAMPROW outptr;
JSAMPLE *range_limit = IDCT_range_limit(cinfo);
int ctr;
int workspace[DCTSIZE2]; /* buffers data between passes */
SHIFT_TEMPS /* for DESCALE */
ISHIFT_TEMPS /* for IRIGHT_SHIFT */
/* Pass 1: process columns from input, store into work array. */
inptr = coef_block;
quantptr = (IFAST_MULT_TYPE *) compptr->dct_table;
wsptr = workspace;
for (ctr = DCTSIZE; ctr > 0; ctr--) {
/* Due to quantization, we will usually find that many of the input
* coefficients are zero, especially the AC terms. We can exploit this
* by short-circuiting the IDCT calculation for any column in which all
* the AC terms are zero. In that case each output is equal to the
* DC coefficient (with scale factor as needed).
* With typical images and quantization tables, half or more of the
* column DCT calculations can be simplified this way.
*/
if (inptr[DCTSIZE*1] == 0 && inptr[DCTSIZE*2] == 0 &&
inptr[DCTSIZE*3] == 0 && inptr[DCTSIZE*4] == 0 &&
inptr[DCTSIZE*5] == 0 && inptr[DCTSIZE*6] == 0 &&
inptr[DCTSIZE*7] == 0) {
/* AC terms all zero */
int dcval = (int) DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]);
wsptr[DCTSIZE*0] = dcval;
wsptr[DCTSIZE*1] = dcval;
wsptr[DCTSIZE*2] = dcval;
wsptr[DCTSIZE*3] = dcval;
wsptr[DCTSIZE*4] = dcval;
wsptr[DCTSIZE*5] = dcval;
wsptr[DCTSIZE*6] = dcval;
wsptr[DCTSIZE*7] = dcval;
inptr++; /* advance pointers to next column */
quantptr++;
wsptr++;
continue;
}
/* Even part */
tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]);
tmp1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]);
tmp2 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]);
tmp3 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]);
tmp10 = tmp0 + tmp2; /* phase 3 */
tmp11 = tmp0 - tmp2;
tmp13 = tmp1 + tmp3; /* phases 5-3 */
tmp12 = MULTIPLY(tmp1 - tmp3, FIX_1_414213562) - tmp13; /* 2*c4 */
tmp0 = tmp10 + tmp13; /* phase 2 */
tmp3 = tmp10 - tmp13;
tmp1 = tmp11 + tmp12;
tmp2 = tmp11 - tmp12;
/* Odd part */
tmp4 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]);
tmp5 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]);
tmp6 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]);
tmp7 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]);
z13 = tmp6 + tmp5; /* phase 6 */
z10 = tmp6 - tmp5;
z11 = tmp4 + tmp7;
z12 = tmp4 - tmp7;
tmp7 = z11 + z13; /* phase 5 */
tmp11 = MULTIPLY(z11 - z13, FIX_1_414213562); /* 2*c4 */
z5 = MULTIPLY(z10 + z12, FIX_1_847759065); /* 2*c2 */
tmp10 = z5 - MULTIPLY(z12, FIX_1_082392200); /* 2*(c2-c6) */
tmp12 = z5 - MULTIPLY(z10, FIX_2_613125930); /* 2*(c2+c6) */
tmp6 = tmp12 - tmp7; /* phase 2 */
tmp5 = tmp11 - tmp6;
tmp4 = tmp10 - tmp5;
wsptr[DCTSIZE*0] = (int) (tmp0 + tmp7);
wsptr[DCTSIZE*7] = (int) (tmp0 - tmp7);
wsptr[DCTSIZE*1] = (int) (tmp1 + tmp6);
wsptr[DCTSIZE*6] = (int) (tmp1 - tmp6);
wsptr[DCTSIZE*2] = (int) (tmp2 + tmp5);
wsptr[DCTSIZE*5] = (int) (tmp2 - tmp5);
wsptr[DCTSIZE*3] = (int) (tmp3 + tmp4);
wsptr[DCTSIZE*4] = (int) (tmp3 - tmp4);
inptr++; /* advance pointers to next column */
quantptr++;
wsptr++;
}
/* Pass 2: process rows from work array, store into output array.
* Note that we must descale the results by a factor of 8 == 2**3,
* and also undo the PASS1_BITS scaling.
*/
wsptr = workspace;
for (ctr = 0; ctr < DCTSIZE; ctr++) {
outptr = output_buf[ctr] + output_col;
/* Add range center and fudge factor for final descale and range-limit. */
z5 = (DCTELEM) wsptr[0] +
((((DCTELEM) RANGE_CENTER) << (PASS1_BITS+3)) +
(1 << (PASS1_BITS+2)));
/* Rows of zeroes can be exploited in the same way as we did with columns.
* However, the column calculation has created many nonzero AC terms, so
* the simplification applies less often (typically 5% to 10% of the time).
* On machines with very fast multiplication, it's possible that the
* test takes more time than it's worth. In that case this section
* may be commented out.
*/
#ifndef NO_ZERO_ROW_TEST
if (wsptr[1] == 0 && wsptr[2] == 0 && wsptr[3] == 0 && wsptr[4] == 0 &&
wsptr[5] == 0 && wsptr[6] == 0 && wsptr[7] == 0) {
/* AC terms all zero */
JSAMPLE dcval = range_limit[(int) IRIGHT_SHIFT(z5, PASS1_BITS+3)
& RANGE_MASK];
outptr[0] = dcval;
outptr[1] = dcval;
outptr[2] = dcval;
outptr[3] = dcval;
outptr[4] = dcval;
outptr[5] = dcval;
outptr[6] = dcval;
outptr[7] = dcval;
wsptr += DCTSIZE; /* advance pointer to next row */
continue;
}
#endif
/* Even part */
tmp10 = z5 + (DCTELEM) wsptr[4];
tmp11 = z5 - (DCTELEM) wsptr[4];
tmp13 = (DCTELEM) wsptr[2] + (DCTELEM) wsptr[6];
tmp12 = MULTIPLY((DCTELEM) wsptr[2] - (DCTELEM) wsptr[6],
FIX_1_414213562) - tmp13; /* 2*c4 */
tmp0 = tmp10 + tmp13;
tmp3 = tmp10 - tmp13;
tmp1 = tmp11 + tmp12;
tmp2 = tmp11 - tmp12;
/* Odd part */
z13 = (DCTELEM) wsptr[5] + (DCTELEM) wsptr[3];
z10 = (DCTELEM) wsptr[5] - (DCTELEM) wsptr[3];
z11 = (DCTELEM) wsptr[1] + (DCTELEM) wsptr[7];
z12 = (DCTELEM) wsptr[1] - (DCTELEM) wsptr[7];
tmp7 = z11 + z13; /* phase 5 */
tmp11 = MULTIPLY(z11 - z13, FIX_1_414213562); /* 2*c4 */
z5 = MULTIPLY(z10 + z12, FIX_1_847759065); /* 2*c2 */
tmp10 = z5 - MULTIPLY(z12, FIX_1_082392200); /* 2*(c2-c6) */
tmp12 = z5 - MULTIPLY(z10, FIX_2_613125930); /* 2*(c2+c6) */
tmp6 = tmp12 - tmp7; /* phase 2 */
tmp5 = tmp11 - tmp6;
tmp4 = tmp10 - tmp5;
/* Final output stage: scale down by a factor of 8 and range-limit */
outptr[0] = range_limit[(int) IRIGHT_SHIFT(tmp0 + tmp7, PASS1_BITS+3)
& RANGE_MASK];
outptr[7] = range_limit[(int) IRIGHT_SHIFT(tmp0 - tmp7, PASS1_BITS+3)
& RANGE_MASK];
outptr[1] = range_limit[(int) IRIGHT_SHIFT(tmp1 + tmp6, PASS1_BITS+3)
& RANGE_MASK];
outptr[6] = range_limit[(int) IRIGHT_SHIFT(tmp1 - tmp6, PASS1_BITS+3)
& RANGE_MASK];
outptr[2] = range_limit[(int) IRIGHT_SHIFT(tmp2 + tmp5, PASS1_BITS+3)
& RANGE_MASK];
outptr[5] = range_limit[(int) IRIGHT_SHIFT(tmp2 - tmp5, PASS1_BITS+3)
& RANGE_MASK];
outptr[3] = range_limit[(int) IRIGHT_SHIFT(tmp3 + tmp4, PASS1_BITS+3)
& RANGE_MASK];
outptr[4] = range_limit[(int) IRIGHT_SHIFT(tmp3 - tmp4, PASS1_BITS+3)
& RANGE_MASK];
wsptr += DCTSIZE; /* advance pointer to next row */
}
}
encode_mcu_DC_first(j_compress_ptr cinfo, JBLOCKROW *MCU_data)
{
phuff_entropy_ptr entropy = (phuff_entropy_ptr)cinfo->entropy;
register int temp, temp2, temp3;
register int nbits;
int blkn, ci;
int Al = cinfo->Al;
JBLOCKROW block;
jpeg_component_info *compptr;
ISHIFT_TEMPS
entropy->next_output_byte = cinfo->dest->next_output_byte;
entropy->free_in_buffer = cinfo->dest->free_in_buffer;
/* Emit restart marker if needed */
if (cinfo->restart_interval)
if (entropy->restarts_to_go == 0)
emit_restart(entropy, entropy->next_restart_num);
/* Encode the MCU data blocks */
for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) {
block = MCU_data[blkn];
ci = cinfo->MCU_membership[blkn];
compptr = cinfo->cur_comp_info[ci];
/* Compute the DC value after the required point transform by Al.
* This is simply an arithmetic right shift.
*/
temp2 = IRIGHT_SHIFT((int)((*block)[0]), Al);
/* DC differences are figured on the point-transformed values. */
temp = temp2 - entropy->last_dc_val[ci];
entropy->last_dc_val[ci] = temp2;
/* Encode the DC coefficient difference per section G.1.2.1 */
/* This is a well-known technique for obtaining the absolute value without
* a branch. It is derived from an assembly language technique presented
* in "How to Optimize for the Pentium Processors", Copyright (c) 1996,
* 1997 by Agner Fog.
*/
temp3 = temp >> (CHAR_BIT * sizeof(int) - 1);
temp ^= temp3;
temp -= temp3; /* temp is abs value of input */
/* For a negative input, want temp2 = bitwise complement of abs(input) */
temp2 = temp ^ temp3;
/* Find the number of bits needed for the magnitude of the coefficient */
nbits = JPEG_NBITS(temp);
/* Check for out-of-range coefficient values.
* Since we're encoding a difference, the range limit is twice as much.
*/
if (nbits > MAX_COEF_BITS + 1)
ERREXIT(cinfo, JERR_BAD_DCT_COEF);
/* Count/emit the Huffman-coded symbol for the number of bits */
emit_symbol(entropy, compptr->dc_tbl_no, nbits);
/* Emit that number of bits of the value, if positive, */
/* or the complement of its magnitude, if negative. */
if (nbits) /* emit_bits rejects calls with size 0 */
emit_bits(entropy, (unsigned int)temp2, nbits);
}
cinfo->dest->next_output_byte = entropy->next_output_byte;
cinfo->dest->free_in_buffer = entropy->free_in_buffer;
/* Update restart-interval state too */
if (cinfo->restart_interval) {
if (entropy->restarts_to_go == 0) {
entropy->restarts_to_go = cinfo->restart_interval;
entropy->next_restart_num++;
entropy->next_restart_num &= 7;
}
entropy->restarts_to_go--;
}
return TRUE;
}
