Lines Matching refs:idct
9 * This code selects a particular IDCT implementation to be used,
11 * is executed per IDCT step, only during output pass setup.
13 * Note that the IDCT routines are responsible for performing coefficient
14 * dequantization as well as the IDCT proper. This module sets up the
15 * dequantization multiplier table needed by the IDCT routine.
74 * to a multiplier table that will actually be used by the IDCT routine.
75 * The multiplier table contents are IDCT-method-dependent. To support
76 * application changes in IDCT method between scans, we can remake the
81 * to zeroes; the result of the IDCT will be a neutral gray level.
90 /* This array contains the IDCT method code that each multiplier table
114 /* The current scaled-IDCT routines require ISLOW-style multiplier tables,
128 * Here we select the proper IDCT routine for each component and build
135 my_idct_ptr idct = (my_idct_ptr) cinfo->idct;
144 /* Select the proper IDCT routine for this component's scaling */
237 idct->pub.inverse_DCT[ci] = method_ptr;
245 if (! compptr->component_needed || idct->cur_method[ci] == method)
250 idct->cur_method[ci] = method;
255 /* For LL&M IDCT method, multipliers are equal to raw quantization
268 /* For AA&N IDCT method, multipliers are equal to quantization
324 /* For float AA&N IDCT method, multipliers are equal to quantization
357 * Initialize IDCT manager.
363 my_idct_ptr idct;
367 idct = (my_idct_ptr)
370 cinfo->idct = (struct jpeg_inverse_dct *) idct;
371 idct->pub.start_pass = start_pass;
381 idct->cur_method[ci] = -1;
