view libmpcodecs/vf_2xsai.c @ 28992:947ef23ba798

Test if create_vdp_decoder() might succeed by calling it from config() with a small value for max_reference_frames. This does not make automatic recovery by using software decoder possible, but lets MPlayer fail more graciously on - actually existing - buggy hardware that does not support certain H264 widths when using hardware accelerated decoding (784, 864, 944, 1024, 1808, 1888 pixels on NVIDIA G98) and if the user tries to hardware-decode more samples at the same time than supported. Might break playback of H264 Intra-Only samples on hardware with very little video memory.

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <inttypes.h>

#include "config.h"
#include "mp_msg.h"

#include "img_format.h"
#include "mp_image.h"
#include "vf.h"

//===========================================================================//

/* FIXME: these all belong in the context, not as globals! */

static uint32_t colorMask = 0xF7DEF7DE;
static uint32_t lowPixelMask = 0x08210821;
static uint32_t qcolorMask = 0xE79CE79C;
static uint32_t qlowpixelMask = 0x18631863;
static uint32_t redblueMask = 0xF81F;
static uint32_t greenMask = 0x7E0;
static int PixelsPerMask = 2;

#define makecol(r,g,b) (r+(g<<8)+(b<<16))
#define makecol_depth(d,r,g,b) (r+(g<<8)+(b<<16))

int Init_2xSaI(int d)
{

	int minr = 0, ming = 0, minb = 0;
	int i;
	
//	if (d != 15 && d != 16 && d != 24 && d != 32)
//		return -1;

	/* Get lowest color bit */	
	for (i = 0; i < 255; i++) {
		if (!minr)
			minr = makecol(i, 0, 0);
		if (!ming)
			ming = makecol(0, i, 0);
		if (!minb)
			minb = makecol(0, 0, i);
	}

	colorMask = (makecol_depth(d, 255, 0, 0) - minr) | (makecol_depth(d, 0, 255, 0) - ming) | (makecol_depth(d, 0, 0, 255) - minb);
	lowPixelMask = minr | ming | minb;
	qcolorMask = (makecol_depth(d, 255, 0, 0) - 3 * minr) | (makecol_depth(d, 0, 255, 0) - 3 * ming) | (makecol_depth(d, 0, 0, 255) - 3 * minb);
	qlowpixelMask = (minr * 3) | (ming * 3) | (minb * 3);
	redblueMask = makecol_depth(d, 255, 0, 255);
	greenMask = makecol_depth(d, 0, 255, 0);

	PixelsPerMask = (d <= 16) ? 2 : 1;
	
	if (PixelsPerMask == 2) {
		colorMask |= (colorMask << 16);
		qcolorMask |= (qcolorMask << 16);
		lowPixelMask |= (lowPixelMask << 16);
		qlowpixelMask |= (qlowpixelMask << 16);
	}

//	TRACE("Color Mask:       0x%lX\n", colorMask);
//	TRACE("Low Pixel Mask:   0x%lX\n", lowPixelMask);
//	TRACE("QColor Mask:      0x%lX\n", qcolorMask);
//	TRACE("QLow Pixel Mask:  0x%lX\n", qlowpixelMask);
	
	return 0;
}


#define GET_RESULT(A, B, C, D) ((A != C || A != D) - (B != C || B != D))

#define INTERPOLATE(A, B) (((A & colorMask) >> 1) + ((B & colorMask) >> 1) + (A & B & lowPixelMask))

#define Q_INTERPOLATE(A, B, C, D) ((A & qcolorMask) >> 2) + ((B & qcolorMask) >> 2) + ((C & qcolorMask) >> 2) + ((D & qcolorMask) >> 2) \
	+ ((((A & qlowpixelMask) + (B & qlowpixelMask) + (C & qlowpixelMask) + (D & qlowpixelMask)) >> 2) & qlowpixelMask)


void Super2xSaI_ex(uint8_t *src, uint32_t src_pitch, 
		   uint8_t *dst, uint32_t dst_pitch,
		   uint32_t width, uint32_t height, int sbpp) {

	unsigned int x, y;
	uint32_t color[16];
	unsigned char *src_line[4];

	/* Point to the first 3 lines. */
	src_line[0] = src;
	src_line[1] = src;
	src_line[2] = src + src_pitch;
	src_line[3] = src + src_pitch * 2;
	
	x = 0, y = 0;
	
	if (PixelsPerMask == 2) {
		unsigned short *sbp;
		sbp = (unsigned short*)src_line[0];
		color[0] = *sbp;       color[1] = color[0];   color[2] = color[0];    color[3] = color[0];
		color[4] = color[0];   color[5] = color[0];   color[6] = *(sbp + 1);  color[7] = *(sbp + 2);
		sbp = (unsigned short*)src_line[2];
		color[8] = *sbp;     color[9] = color[8];     color[10] = *(sbp + 1); color[11] = *(sbp + 2);
		sbp = (unsigned short*)src_line[3];
		color[12] = *sbp;    color[13] = color[12];   color[14] = *(sbp + 1); color[15] = *(sbp + 2);
	}
	else {
		uint32_t *lbp;
		lbp = (uint32_t*)src_line[0];
		color[0] = *lbp;       color[1] = color[0];   color[2] = color[0];    color[3] = color[0];
		color[4] = color[0];   color[5] = color[0];   color[6] = *(lbp + 1);  color[7] = *(lbp + 2);
		lbp = (uint32_t*)src_line[2];
		color[8] = *lbp;     color[9] = color[8];     color[10] = *(lbp + 1); color[11] = *(lbp + 2);
		lbp = (uint32_t*)src_line[3];
		color[12] = *lbp;    color[13] = color[12];   color[14] = *(lbp + 1); color[15] = *(lbp + 2);
	}

	for (y = 0; y < height; y++) {
		unsigned char *dst_line[2];

		dst_line[0] = dst + dst_pitch*2*y;
		dst_line[1] = dst + dst_pitch*(2*y+1);
	
		/* Todo: x = width - 2, x = width - 1 */
		
		for (x = 0; x < width; x++) {
			uint32_t product1a, product1b, product2a, product2b;

//---------------------------------------  B0 B1 B2 B3    0  1  2  3
//                                         4  5* 6  S2 -> 4  5* 6  7
//                                         1  2  3  S1    8  9 10 11
//                                         A0 A1 A2 A3   12 13 14 15
//--------------------------------------
			if (color[9] == color[6] && color[5] != color[10]) {
				product2b = color[9];
				product1b = product2b;
			}
			else if (color[5] == color[10] && color[9] != color[6]) {
				product2b = color[5];
				product1b = product2b;
			}
			else if (color[5] == color[10] && color[9] == color[6]) {
				int r = 0;

				r += GET_RESULT(color[6], color[5], color[8], color[13]);
				r += GET_RESULT(color[6], color[5], color[4], color[1]);
				r += GET_RESULT(color[6], color[5], color[14], color[11]);
				r += GET_RESULT(color[6], color[5], color[2], color[7]);

				if (r > 0)
					product1b = color[6];
				else if (r < 0)
					product1b = color[5];
				else
					product1b = INTERPOLATE(color[5], color[6]);
					
				product2b = product1b;

			}
			else {
				if (color[6] == color[10] && color[10] == color[13] && color[9] != color[14] && color[10] != color[12])
					product2b = Q_INTERPOLATE(color[10], color[10], color[10], color[9]);
				else if (color[5] == color[9] && color[9] == color[14] && color[13] != color[10] && color[9] != color[15])
					product2b = Q_INTERPOLATE(color[9], color[9], color[9], color[10]);
				else
					product2b = INTERPOLATE(color[9], color[10]);

				if (color[6] == color[10] && color[6] == color[1] && color[5] != color[2] && color[6] != color[0])
					product1b = Q_INTERPOLATE(color[6], color[6], color[6], color[5]);
				else if (color[5] == color[9] && color[5] == color[2] && color[1] != color[6] && color[5] != color[3])
					product1b = Q_INTERPOLATE(color[6], color[5], color[5], color[5]);
				else
					product1b = INTERPOLATE(color[5], color[6]);
			}

			if (color[5] == color[10] && color[9] != color[6] && color[4] == color[5] && color[5] != color[14])
				product2a = INTERPOLATE(color[9], color[5]);
			else if (color[5] == color[8] && color[6] == color[5] && color[4] != color[9] && color[5] != color[12])
				product2a = INTERPOLATE(color[9], color[5]);
			else
				product2a = color[9];

			if (color[9] == color[6] && color[5] != color[10] && color[8] == color[9] && color[9] != color[2])
				product1a = INTERPOLATE(color[9], color[5]);
			else if (color[4] == color[9] && color[10] == color[9] && color[8] != color[5] && color[9] != color[0])
				product1a = INTERPOLATE(color[9], color[5]);
			else
				product1a = color[5];
	
			if (PixelsPerMask == 2) {
				*((uint32_t *) (&dst_line[0][x * 4])) = product1a | (product1b << 16);
				*((uint32_t *) (&dst_line[1][x * 4])) = product2a | (product2b << 16);
			}
			else {
				*((uint32_t *) (&dst_line[0][x * 8])) = product1a;
				*((uint32_t *) (&dst_line[0][x * 8 + 4])) = product1b;
				*((uint32_t *) (&dst_line[1][x * 8])) = product2a;
				*((uint32_t *) (&dst_line[1][x * 8 + 4])) = product2b;
			}
			
			/* Move color matrix forward */
			color[0] = color[1]; color[4] = color[5]; color[8] = color[9];   color[12] = color[13];
			color[1] = color[2]; color[5] = color[6]; color[9] = color[10];  color[13] = color[14];
			color[2] = color[3]; color[6] = color[7]; color[10] = color[11]; color[14] = color[15];
			
			if (x < width - 3) {
				x += 3;
				if (PixelsPerMask == 2) {
					color[3] = *(((unsigned short*)src_line[0]) + x);					
					color[7] = *(((unsigned short*)src_line[1]) + x);
					color[11] = *(((unsigned short*)src_line[2]) + x);
					color[15] = *(((unsigned short*)src_line[3]) + x);
				}
				else {
					color[3] = *(((uint32_t*)src_line[0]) + x);
					color[7] = *(((uint32_t*)src_line[1]) + x);
					color[11] = *(((uint32_t*)src_line[2]) + x);
					color[15] = *(((uint32_t*)src_line[3]) + x);
				}
				x -= 3;
			}
		}

		/* We're done with one line, so we shift the source lines up */
		src_line[0] = src_line[1];
		src_line[1] = src_line[2];
		src_line[2] = src_line[3];		

		/* Read next line */
		if (y + 3 >= height)
			src_line[3] = src_line[2];
		else
			src_line[3] = src_line[2] + src_pitch;
			
		/* Then shift the color matrix up */
		if (PixelsPerMask == 2) {
			unsigned short *sbp;
			sbp = (unsigned short*)src_line[0];
			color[0] = *sbp;     color[1] = color[0];    color[2] = *(sbp + 1);  color[3] = *(sbp + 2);
			sbp = (unsigned short*)src_line[1];
			color[4] = *sbp;     color[5] = color[4];    color[6] = *(sbp + 1);  color[7] = *(sbp + 2);
			sbp = (unsigned short*)src_line[2];
			color[8] = *sbp;     color[9] = color[9];    color[10] = *(sbp + 1); color[11] = *(sbp + 2);
			sbp = (unsigned short*)src_line[3];
			color[12] = *sbp;    color[13] = color[12];  color[14] = *(sbp + 1); color[15] = *(sbp + 2);
		}
		else {
			uint32_t *lbp;
			lbp = (uint32_t*)src_line[0];
			color[0] = *lbp;     color[1] = color[0];    color[2] = *(lbp + 1);  color[3] = *(lbp + 2);
			lbp = (uint32_t*)src_line[1];
			color[4] = *lbp;     color[5] = color[4];    color[6] = *(lbp + 1);  color[7] = *(lbp + 2);
			lbp = (uint32_t*)src_line[2];
			color[8] = *lbp;     color[9] = color[9];    color[10] = *(lbp + 1); color[11] = *(lbp + 2);
			lbp = (uint32_t*)src_line[3];
			color[12] = *lbp;    color[13] = color[12];  color[14] = *(lbp + 1); color[15] = *(lbp + 2);
		}
		
	} // y loop
	
}


//===========================================================================//

static int config(struct vf_instance_s* vf,
        int width, int height, int d_width, int d_height,
	unsigned int flags, unsigned int outfmt){

    Init_2xSaI(outfmt&255);

    return vf_next_config(vf,2*width,2*height,2*d_width,2*d_height,flags,outfmt);
}

static int put_image(struct vf_instance_s* vf, mp_image_t *mpi, double pts){
    mp_image_t *dmpi;

    // hope we'll get DR buffer:
    dmpi=vf_get_image(vf->next,mpi->imgfmt,
	MP_IMGTYPE_TEMP, MP_IMGFLAG_ACCEPT_STRIDE,
	2*mpi->w, 2*mpi->h);

    Super2xSaI_ex(mpi->planes[0], mpi->stride[0],
		  dmpi->planes[0], dmpi->stride[0],
		  mpi->w, mpi->h, mpi->bpp/8);
    
    return vf_next_put_image(vf,dmpi, pts);
}

//===========================================================================//

static int query_format(struct vf_instance_s* vf, unsigned int fmt){
    switch(fmt){
//    case IMGFMT_BGR15:
//    case IMGFMT_BGR16:
    case IMGFMT_BGR32:
	return vf_next_query_format(vf,fmt);
    }
    return 0;
}

static int open(vf_instance_t *vf, char* args){
    vf->config=config;
    vf->put_image=put_image;
    vf->query_format=query_format;
    return 1;
}

const vf_info_t vf_info_2xsai = {
    "2xSai BGR bitmap 2x scaler",
    "2xsai",
    "A'rpi",
    "http://elektron.its.tudelft.nl/~dalikifa/",
    open,
    NULL
};

//===========================================================================//