1 //<MStar Software> 2 //****************************************************************************** 3 // MStar Software 4 // Copyright (c) 2010 - 2012 MStar Semiconductor, Inc. All rights reserved. 5 // All software, firmware and related documentation herein ("MStar Software") are 6 // intellectual property of MStar Semiconductor, Inc. ("MStar") and protected by 7 // law, including, but not limited to, copyright law and international treaties. 8 // Any use, modification, reproduction, retransmission, or republication of all 9 // or part of MStar Software is expressly prohibited, unless prior written 10 // permission has been granted by MStar. 11 // 12 // By accessing, browsing and/or using MStar Software, you acknowledge that you 13 // have read, understood, and agree, to be bound by below terms ("Terms") and to 14 // comply with all applicable laws and regulations: 15 // 16 // 1. MStar shall retain any and all right, ownership and interest to MStar 17 // Software and any modification/derivatives thereof. 18 // No right, ownership, or interest to MStar Software and any 19 // modification/derivatives thereof is transferred to you under Terms. 20 // 21 // 2. You understand that MStar Software might include, incorporate or be 22 // supplied together with third party`s software and the use of MStar 23 // Software may require additional licenses from third parties. 24 // Therefore, you hereby agree it is your sole responsibility to separately 25 // obtain any and all third party right and license necessary for your use of 26 // such third party`s software. 27 // 28 // 3. MStar Software and any modification/derivatives thereof shall be deemed as 29 // MStar`s confidential information and you agree to keep MStar`s 30 // confidential information in strictest confidence and not disclose to any 31 // third party. 32 // 33 // 4. MStar Software is provided on an "AS IS" basis without warranties of any 34 // kind. Any warranties are hereby expressly disclaimed by MStar, including 35 // without limitation, any warranties of merchantability, non-infringement of 36 // intellectual property rights, fitness for a particular purpose, error free 37 // and in conformity with any international standard. You agree to waive any 38 // claim against MStar for any loss, damage, cost or expense that you may 39 // incur related to your use of MStar Software. 40 // In no event shall MStar be liable for any direct, indirect, incidental or 41 // consequential damages, including without limitation, lost of profit or 42 // revenues, lost or damage of data, and unauthorized system use. 43 // You agree that this Section 4 shall still apply without being affected 44 // even if MStar Software has been modified by MStar in accordance with your 45 // request or instruction for your use, except otherwise agreed by both 46 // parties in writing. 47 // 48 // 5. If requested, MStar may from time to time provide technical supports or 49 // services in relation with MStar Software to you for your use of 50 // MStar Software in conjunction with your or your customer`s product 51 // ("Services"). 52 // You understand and agree that, except otherwise agreed by both parties in 53 // writing, Services are provided on an "AS IS" basis and the warranty 54 // disclaimer set forth in Section 4 above shall apply. 55 // 56 // 6. Nothing contained herein shall be construed as by implication, estoppels 57 // or otherwise: 58 // (a) conferring any license or right to use MStar name, trademark, service 59 // mark, symbol or any other identification; 60 // (b) obligating MStar or any of its affiliates to furnish any person, 61 // including without limitation, you and your customers, any assistance 62 // of any kind whatsoever, or any information; or 63 // (c) conferring any license or right under any intellectual property right. 64 // 65 // 7. These terms shall be governed by and construed in accordance with the laws 66 // of Taiwan, R.O.C., excluding its conflict of law rules. 67 // Any and all dispute arising out hereof or related hereto shall be finally 68 // settled by arbitration referred to the Chinese Arbitration Association, 69 // Taipei in accordance with the ROC Arbitration Law and the Arbitration 70 // Rules of the Association by three (3) arbitrators appointed in accordance 71 // with the said Rules. 72 // The place of arbitration shall be in Taipei, Taiwan and the language shall 73 // be English. 74 // The arbitration award shall be final and binding to both parties. 75 // 76 //****************************************************************************** 77 //<MStar Software> 78 /* 79 * jidctint.c 80 * 81 * Copyright (C) 1991-1998, Thomas G. Lane. 82 * This file is part of the Independent JPEG Group's software. 83 * For conditions of distribution and use, see the accompanying README file. 84 * 85 * This file contains a slow-but-accurate integer implementation of the 86 * inverse DCT (Discrete Cosine Transform). In the IJG code, this routine 87 * must also perform dequantization of the input coefficients. 88 * 89 * A 2-D IDCT can be done by 1-D IDCT on each column followed by 1-D IDCT 90 * on each row (or vice versa, but it's more convenient to emit a row at 91 * a time). Direct algorithms are also available, but they are much more 92 * complex and seem not to be any faster when reduced to code. 93 * 94 * This implementation is based on an algorithm described in 95 * C. Loeffler, A. Ligtenberg and G. Moschytz, "Practical Fast 1-D DCT 96 * Algorithms with 11 Multiplications", Proc. Int'l. Conf. on Acoustics, 97 * Speech, and Signal Processing 1989 (ICASSP '89), pp. 988-991. 98 * The primary algorithm described there uses 11 multiplies and 29 adds. 99 * We use their alternate method with 12 multiplies and 32 adds. 100 * The advantage of this method is that no data path contains more than one 101 * multiplication; this allows a very simple and accurate implementation in 102 * scaled fixed-point arithmetic, with a minimal number of shifts. 103 */ 104 105 #include "jpegmain.h" 106 #include "apiJPEG.h" 107 ///#define JPEG_INTERNALS 108 ///#include "jinclude.h" 109 ///#include "jpeglib.h" 110 ///#include "jdct.h" /* Private declarations for DCT subsystem */ 111 112 #if 1///def DCT_ISLOW_SUPPORTED 113 114 115 /* 116 * This module is specialized to the case DCTSIZE = 8. 117 */ 118 119 #define DCTSIZE 8 120 #define BITS_IN_JSAMPLE 8 121 122 123 #if DCTSIZE != 8 124 Sorry, this code only copes with 8x8 DCTs. /* deliberate syntax err */ 125 #endif 126 127 128 /* 129 * The poop on this scaling stuff is as follows: 130 * 131 * Each 1-D IDCT step produces outputs which are a factor of sqrt(N) 132 * larger than the true IDCT outputs. The final outputs are therefore 133 * a factor of N larger than desired; since N=8 this can be cured by 134 * a simple right shift at the end of the algorithm. The advantage of 135 * this arrangement is that we save two multiplications per 1-D IDCT, 136 * because the y0 and y4 inputs need not be divided by sqrt(N). 137 * 138 * We have to do addition and subtraction of the integer inputs, which 139 * is no problem, and multiplication by fractional constants, which is 140 * a problem to do in integer arithmetic. We multiply all the constants 141 * by CONST_SCALE and convert them to integer constants (thus retaining 142 * CONST_BITS bits of precision in the constants). After doing a 143 * multiplication we have to divide the product by CONST_SCALE, with proper 144 * rounding, to produce the correct output. This division can be done 145 * cheaply as a right shift of CONST_BITS bits. We postpone shifting 146 * as long as possible so that partial sums can be added together with 147 * full fractional precision. 148 * 149 * The outputs of the first pass are scaled up by PASS1_BITS bits so that 150 * they are represented to better-than-integral precision. These outputs 151 * require BITS_IN_JSAMPLE + PASS1_BITS + 3 bits; this fits in a 16-bit word 152 * with the recommended scaling. (To scale up 12-bit sample data further, an 153 * intermediate INT32 array would be needed.) 154 * 155 * To avoid overflow of the 32-bit intermediate results in pass 2, we must 156 * have BITS_IN_JSAMPLE + CONST_BITS + PASS1_BITS <= 26. Error analysis 157 * shows that the values given below are the most effective. 158 */ 159 160 #if BITS_IN_JSAMPLE == 8 161 #define CONST_BITS 13 162 #define PASS1_BITS 2 163 #else 164 #define CONST_BITS 13 165 #define PASS1_BITS 1 /* lose a little precision to avoid overflow */ 166 #endif 167 168 /* Some C compilers fail to reduce "FIX(constant)" at compile time, thus 169 * causing a lot of useless floating-point operations at run time. 170 * To get around this we use the following pre-calculated constants. 171 * If you change CONST_BITS you may want to add appropriate values. 172 * (With a reasonable C compiler, you can just rely on the FIX() macro...) 173 */ 174 175 #if CONST_BITS == 13 176 #define FIX_0_298631336 ((INT32) 2446) /* FIX(0.298631336) */ 177 #define FIX_0_390180644 ((INT32) 3196) /* FIX(0.390180644) */ 178 #define FIX_0_541196100 ((INT32) 4433) /* FIX(0.541196100) */ 179 #define FIX_0_765366865 ((INT32) 6270) /* FIX(0.765366865) */ 180 #define FIX_0_899976223 ((INT32) 7373) /* FIX(0.899976223) */ 181 #define FIX_1_175875602 ((INT32) 9633) /* FIX(1.175875602) */ 182 #define FIX_1_501321110 ((INT32) 12299) /* FIX(1.501321110) */ 183 #define FIX_1_847759065 ((INT32) 15137) /* FIX(1.847759065) */ 184 #define FIX_1_961570560 ((INT32) 16069) /* FIX(1.961570560) */ 185 #define FIX_2_053119869 ((INT32) 16819) /* FIX(2.053119869) */ 186 #define FIX_2_562915447 ((INT32) 20995) /* FIX(2.562915447) */ 187 #define FIX_3_072711026 ((INT32) 25172) /* FIX(3.072711026) */ 188 #else 189 #define FIX_0_298631336 FIX(0.298631336) 190 #define FIX_0_390180644 FIX(0.390180644) 191 #define FIX_0_541196100 FIX(0.541196100) 192 #define FIX_0_765366865 FIX(0.765366865) 193 #define FIX_0_899976223 FIX(0.899976223) 194 #define FIX_1_175875602 FIX(1.175875602) 195 #define FIX_1_501321110 FIX(1.501321110) 196 #define FIX_1_847759065 FIX(1.847759065) 197 #define FIX_1_961570560 FIX(1.961570560) 198 #define FIX_2_053119869 FIX(2.053119869) 199 #define FIX_2_562915447 FIX(2.562915447) 200 #define FIX_3_072711026 FIX(3.072711026) 201 #endif 202 203 204 /* Multiply an INT32 variable by an INT32 constant to yield an INT32 result. 205 * For 8-bit samples with the recommended scaling, all the variable 206 * and constant values involved are no more than 16 bits wide, so a 207 * 16x16->32 bit multiply can be used instead of a full 32x32 multiply. 208 * For 12-bit samples, a full 32-bit multiplication will be needed. 209 */ 210 211 /* 212 #if 0 ///BITS_IN_JSAMPLE == 8 213 #define MULTIPLY(var,const) MULTIPLY16C16(var,const) 214 #else 215 #define MULTIPLY(var,const) ((var) * (const)) 216 #endif 217 */ 218 #define MULTIPLY(var,cnst) ((var) * (cnst)) 219 220 /* Dequantize a coefficient by multiplying it by the multiplier-table 221 * entry; produce an int result. In this module, both inputs and result 222 * are 16 bits or less, so either int or short multiply will work. 223 */ 224 225 #define ISLOW_MULT_TYPE int 226 #define DEQUANTIZE(coef,quantval) (coef) //(((ISLOW_MULT_TYPE) (coef)) ) ///(((ISLOW_MULT_TYPE) (coef)) * (quantval)) 227 228 //#define DESCALE(x,n) ( ( (x) + (1 << ((n)-1)) ) >> n) 229 #define SCALEDONE ((int32) 1) 230 #define DESCALE(x,n) (((x) + (SCALEDONE << ((n)-1))) >> (n)) 231 232 /* 233 * Perform dequantization and inverse DCT on one block of coefficients. 234 */ 235 236 ///GLOBAL(void) 237 ///jpeg_idct_islow (j_decompress_ptr cinfo, jpeg_component_info * compptr, 238 /// JCOEFPTR coef_block, 239 /// JSAMPARRAY output_buf, JDIMENSION output_col) 240 #define clamp(i) if (i & 0xFF00) i = (((~i) >> 15) & 0xFF); 241 void jpeg_idct_islow( JPEG_BLOCK_TYPE *data, U8 *Pdst_ptr ) 242 { 243 #define INT32 S32 244 #define DCTSIZE2 64 245 #define DCTSIZE 8 246 247 INT32 tmp0, tmp1, tmp2, tmp3; 248 INT32 tmp10, tmp11, tmp12, tmp13; 249 INT32 z1, z2, z3, z4, z5; 250 ///JCOEFPTR inptr; 251 register JPEG_BLOCK_TYPE *inptr; 252 ///ISLOW_MULT_TYPE *quantptr; 253 U8 *outptr = Pdst_ptr; 254 ///JSAMPLE *range_limit = IDCT_range_limit(cinfo); 255 int ctr; 256 JPEG_BLOCK_TYPE workspace[DCTSIZE2]; /* buffers data between passes */ 257 JPEG_BLOCK_TYPE *wsptr; 258 ///SHIFT_TEMPS 259 S16 i; 260 //printf("Jidctint::jpeg_idct_islow\n"); 261 /* Pass 1: process columns from input, store into work array. */ 262 /* Note results are scaled up by sqrt(8) compared to a true IDCT; */ 263 /* furthermore, we scale the results by 2**PASS1_BITS. */ 264 265 inptr = data; 266 ///quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; 267 wsptr = workspace; 268 for ( ctr = DCTSIZE; ctr > 0; ctr-- ) 269 { 270 /* Due to quantization, we will usually find that many of the input 271 * coefficients are zero, especially the AC terms. We can exploit this 272 * by short-circuiting the IDCT calculation for any column in which all 273 * the AC terms are zero. In that case each output is equal to the 274 * DC coefficient (with scale factor as needed). 275 * With typical images and quantization tables, half or more of the 276 * column DCT calculations can be simplified this way. 277 */ 278 279 if ( ( inptr[DCTSIZE * 1] | inptr[DCTSIZE * 2] | inptr[DCTSIZE * 3] | inptr[DCTSIZE * 4] | inptr[DCTSIZE * 5] | inptr[DCTSIZE * 6] | inptr[DCTSIZE * 7] ) == 0 ) 280 { 281 /* AC terms all zero */ 282 int dcval = DEQUANTIZE( inptr[DCTSIZE*0], quantptr[DCTSIZE*0] ) << PASS1_BITS; 283 284 wsptr[DCTSIZE * 0] = dcval; 285 wsptr[DCTSIZE * 1] = dcval; 286 wsptr[DCTSIZE * 2] = dcval; 287 wsptr[DCTSIZE * 3] = dcval; 288 wsptr[DCTSIZE * 4] = dcval; 289 wsptr[DCTSIZE * 5] = dcval; 290 wsptr[DCTSIZE * 6] = dcval; 291 wsptr[DCTSIZE * 7] = dcval; 292 293 inptr++; /* advance pointers to next column */ 294 //quantptr++; 295 wsptr++; 296 continue; 297 } 298 299 /* Even part: reverse the even part of the forward DCT. */ 300 /* The rotator is sqrt(2)*c(-6). */ 301 302 z2 = DEQUANTIZE( inptr[DCTSIZE * 2], quantptr[DCTSIZE * 2] ); 303 z3 = DEQUANTIZE( inptr[DCTSIZE * 6], quantptr[DCTSIZE * 6] ); 304 305 z1 = MULTIPLY( z2 + z3, FIX_0_541196100 ); 306 tmp2 = z1 + MULTIPLY( z3, -FIX_1_847759065 ); 307 tmp3 = z1 + MULTIPLY( z2, FIX_0_765366865 ); 308 309 z2 = DEQUANTIZE( inptr[DCTSIZE * 0], quantptr[DCTSIZE * 0] ); 310 z3 = DEQUANTIZE( inptr[DCTSIZE * 4], quantptr[DCTSIZE * 4] ); 311 312 tmp0 = ( z2 + z3 ) << CONST_BITS; 313 tmp1 = ( z2 - z3 ) << CONST_BITS; 314 315 tmp10 = tmp0 + tmp3; 316 tmp13 = tmp0 - tmp3; 317 tmp11 = tmp1 + tmp2; 318 tmp12 = tmp1 - tmp2; 319 320 /* Odd part per figure 8; the matrix is unitary and hence its 321 * transpose is its inverse. i0..i3 are y7,y5,y3,y1 respectively. 322 */ 323 324 tmp0 = DEQUANTIZE( inptr[DCTSIZE * 7], quantptr[DCTSIZE * 7] ); 325 tmp1 = DEQUANTIZE( inptr[DCTSIZE * 5], quantptr[DCTSIZE * 5] ); 326 tmp2 = DEQUANTIZE( inptr[DCTSIZE * 3], quantptr[DCTSIZE * 3] ); 327 tmp3 = DEQUANTIZE( inptr[DCTSIZE * 1], quantptr[DCTSIZE * 1] ); 328 329 z1 = tmp0 + tmp3; 330 z2 = tmp1 + tmp2; 331 z3 = tmp0 + tmp2; 332 z4 = tmp1 + tmp3; 333 z5 = MULTIPLY( z3 + z4, FIX_1_175875602 ); /* sqrt(2) * c3 */ 334 335 tmp0 = MULTIPLY( tmp0, FIX_0_298631336 ); /* sqrt(2) * (-c1+c3+c5-c7) */ 336 tmp1 = MULTIPLY( tmp1, FIX_2_053119869 ); /* sqrt(2) * ( c1+c3-c5+c7) */ 337 tmp2 = MULTIPLY( tmp2, FIX_3_072711026 ); /* sqrt(2) * ( c1+c3+c5-c7) */ 338 tmp3 = MULTIPLY( tmp3, FIX_1_501321110 ); /* sqrt(2) * ( c1+c3-c5-c7) */ 339 z1 = MULTIPLY( z1, -FIX_0_899976223 ); /* sqrt(2) * (c7-c3) */ 340 z2 = MULTIPLY( z2, -FIX_2_562915447 ); /* sqrt(2) * (-c1-c3) */ 341 z3 = MULTIPLY( z3, -FIX_1_961570560 ); /* sqrt(2) * (-c3-c5) */ 342 z4 = MULTIPLY( z4, -FIX_0_390180644 ); /* sqrt(2) * (c5-c3) */ 343 344 z3 += z5; 345 z4 += z5; 346 347 tmp0 += z1 + z3; 348 tmp1 += z2 + z4; 349 tmp2 += z2 + z3; 350 tmp3 += z1 + z4; 351 352 /* Final output stage: inputs are tmp10..tmp13, tmp0..tmp3 */ 353 354 wsptr[DCTSIZE * 0] = ( int )DESCALE( tmp10 + tmp3, CONST_BITS - PASS1_BITS ); 355 wsptr[DCTSIZE * 7] = ( int )DESCALE( tmp10 - tmp3, CONST_BITS - PASS1_BITS ); 356 wsptr[DCTSIZE * 1] = ( int )DESCALE( tmp11 + tmp2, CONST_BITS - PASS1_BITS ); 357 wsptr[DCTSIZE * 6] = ( int )DESCALE( tmp11 - tmp2, CONST_BITS - PASS1_BITS ); 358 wsptr[DCTSIZE * 2] = ( int )DESCALE( tmp12 + tmp1, CONST_BITS - PASS1_BITS ); 359 wsptr[DCTSIZE * 5] = ( int )DESCALE( tmp12 - tmp1, CONST_BITS - PASS1_BITS ); 360 wsptr[DCTSIZE * 3] = ( int )DESCALE( tmp13 + tmp0, CONST_BITS - PASS1_BITS ); 361 wsptr[DCTSIZE * 4] = ( int )DESCALE( tmp13 - tmp0, CONST_BITS - PASS1_BITS ); 362 363 inptr++; /* advance pointers to next column */ 364 //quantptr++; 365 wsptr++; 366 } 367 368 /* Pass 2: process rows from work array, store into output array. */ 369 /* Note that we must descale the results by a factor of 8 == 2**3, */ 370 /* and also undo the PASS1_BITS scaling. */ 371 372 wsptr = workspace; 373 for ( ctr = 0; ctr < DCTSIZE; ctr++ ) 374 { 375 ///outptr = output_buf[ctr] + output_col; 376 /* Rows of zeroes can be exploited in the same way as we did with columns. 377 * However, the column calculation has created many nonzero AC terms, so 378 * the simplification applies less often (typically 5% to 10% of the time). 379 * On machines with very fast multiplication, it's possible that the 380 * test takes more time than it's worth. In that case this section 381 * may be commented out. 382 */ 383 384 #if 1///ndef NO_ZERO_ROW_TEST 385 if ( ( wsptr[1] | wsptr[2] | wsptr[3] | wsptr[4] | wsptr[5] | wsptr[6] | wsptr[7] ) == 0 ) 386 { 387 /* AC terms all zero */ 388 int dcval = ( int )DESCALE( ( INT32 )wsptr[DCTSIZE*0], PASS1_BITS + 3 ) + 128; ///range_limit[(int) DESCALE((INT32) wsptr[0], PASS1_BITS+3) & RANGE_MASK]; 389 clamp( dcval ) 390 outptr[0] = dcval; 391 outptr[1] = dcval; 392 outptr[2] = dcval; 393 outptr[3] = dcval; 394 outptr[4] = dcval; 395 outptr[5] = dcval; 396 outptr[6] = dcval; 397 outptr[7] = dcval; 398 399 wsptr += DCTSIZE; /* advance pointer to next row */ 400 outptr += DCTSIZE; 401 continue; 402 } 403 #endif 404 405 /* Even part: reverse the even part of the forward DCT. */ 406 /* The rotator is sqrt(2)*c(-6). */ 407 408 z2 = ( INT32 )wsptr[2]; 409 z3 = ( INT32 )wsptr[6]; 410 411 z1 = MULTIPLY( z2 + z3, FIX_0_541196100 ); 412 tmp2 = z1 + MULTIPLY( z3, -FIX_1_847759065 ); 413 tmp3 = z1 + MULTIPLY( z2, FIX_0_765366865 ); 414 415 tmp0 = ( ( INT32 )wsptr[0] + ( INT32 )wsptr[4] ) << CONST_BITS; 416 tmp1 = ( ( INT32 )wsptr[0] - ( INT32 )wsptr[4] ) << CONST_BITS; 417 418 tmp10 = tmp0 + tmp3; 419 tmp13 = tmp0 - tmp3; 420 tmp11 = tmp1 + tmp2; 421 tmp12 = tmp1 - tmp2; 422 423 /* Odd part per figure 8; the matrix is unitary and hence its 424 * transpose is its inverse. i0..i3 are y7,y5,y3,y1 respectively. 425 */ 426 427 tmp0 = ( INT32 )wsptr[7]; 428 tmp1 = ( INT32 )wsptr[5]; 429 tmp2 = ( INT32 )wsptr[3]; 430 tmp3 = ( INT32 )wsptr[1]; 431 432 z1 = tmp0 + tmp3; 433 z2 = tmp1 + tmp2; 434 z3 = tmp0 + tmp2; 435 z4 = tmp1 + tmp3; 436 z5 = MULTIPLY( z3 + z4, FIX_1_175875602 ); /* sqrt(2) * c3 */ 437 438 tmp0 = MULTIPLY( tmp0, FIX_0_298631336 ); /* sqrt(2) * (-c1+c3+c5-c7) */ 439 tmp1 = MULTIPLY( tmp1, FIX_2_053119869 ); /* sqrt(2) * ( c1+c3-c5+c7) */ 440 tmp2 = MULTIPLY( tmp2, FIX_3_072711026 ); /* sqrt(2) * ( c1+c3+c5-c7) */ 441 tmp3 = MULTIPLY( tmp3, FIX_1_501321110 ); /* sqrt(2) * ( c1+c3-c5-c7) */ 442 z1 = MULTIPLY( z1, -FIX_0_899976223 ); /* sqrt(2) * (c7-c3) */ 443 z2 = MULTIPLY( z2, -FIX_2_562915447 ); /* sqrt(2) * (-c1-c3) */ 444 z3 = MULTIPLY( z3, -FIX_1_961570560 ); /* sqrt(2) * (-c3-c5) */ 445 z4 = MULTIPLY( z4, -FIX_0_390180644 ); /* sqrt(2) * (c5-c3) */ 446 447 z3 += z5; 448 z4 += z5; 449 450 tmp0 += z1 + z3; 451 tmp1 += z2 + z4; 452 tmp2 += z2 + z3; 453 tmp3 += z1 + z4; 454 455 /* Final output stage: inputs are tmp10..tmp13, tmp0..tmp3 */ 456 i = ( int )DESCALE( tmp10 + tmp3, CONST_BITS + PASS1_BITS + 3 ) + 128; ///range_limit[(int) DESCALE(tmp10 + tmp3, CONST_BITS+PASS1_BITS+3) & RANGE_MASK]; 457 clamp( i ) 458 outptr[0] = ( U8 )i; 459 460 i = ( int )DESCALE( tmp10 - tmp3, CONST_BITS + PASS1_BITS + 3 ) + 128; ///range_limit[(int) DESCALE(tmp10 - tmp3, CONST_BITS+PASS1_BITS+3) & RANGE_MASK]; 461 clamp( i ) 462 outptr[7] = ( U8 )i; 463 i = ( int )DESCALE( tmp11 + tmp2, CONST_BITS + PASS1_BITS + 3 ) + 128; ///range_limit[(int) DESCALE(tmp11 + tmp2, CONST_BITS+PASS1_BITS+3) & RANGE_MASK]; 464 clamp( i ) 465 outptr[1] = ( U8 )i; 466 i = ( int )DESCALE( tmp11 - tmp2, CONST_BITS + PASS1_BITS + 3 ) + 128; ///range_limit[(int) DESCALE(tmp11 - tmp2, CONST_BITS+PASS1_BITS+3) & RANGE_MASK]; 467 clamp( i ) 468 outptr[6] = ( U8 )i; 469 i = ( int )DESCALE( tmp12 + tmp1, CONST_BITS + PASS1_BITS + 3 ) + 128; ///range_limit[(int) DESCALE(tmp12 + tmp1, CONST_BITS+PASS1_BITS+3) & RANGE_MASK]; 470 clamp( i ) 471 outptr[2] = ( U8 )i; 472 i = ( int )DESCALE( tmp12 - tmp1, CONST_BITS + PASS1_BITS + 3 ) + 128; ///range_limit[(int) DESCALE(tmp12 - tmp1, CONST_BITS+PASS1_BITS+3) & RANGE_MASK]; 473 clamp( i ) 474 outptr[5] = ( U8 )i; 475 i = ( int )DESCALE( tmp13 + tmp0, CONST_BITS + PASS1_BITS + 3 ) + 128; ///range_limit[(int) DESCALE(tmp13 + tmp0, CONST_BITS+PASS1_BITS+3) & RANGE_MASK]; 476 clamp( i ) 477 outptr[3] = ( U8 )i; 478 i = ( int )DESCALE( tmp13 - tmp0, CONST_BITS + PASS1_BITS + 3 ) + 128; ///range_limit[(int) DESCALE(tmp13 - tmp0, CONST_BITS+PASS1_BITS+3) & RANGE_MASK]; 479 clamp( i ) 480 outptr[4] = ( U8 )i; 481 482 wsptr += DCTSIZE; /* advance pointer to next row */ 483 outptr += DCTSIZE; 484 } 485 } 486 487 488 #endif /* DCT_ISLOW_SUPPORTED */ 489 490