aboutsummaryrefslogtreecommitdiff
path: root/drivers/video/c2p.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/video/c2p.c')
-rw-r--r--drivers/video/c2p.c232
1 files changed, 0 insertions, 232 deletions
diff --git a/drivers/video/c2p.c b/drivers/video/c2p.c
deleted file mode 100644
index 376bc07ff95..00000000000
--- a/drivers/video/c2p.c
+++ /dev/null
@@ -1,232 +0,0 @@
-/*
- * Fast C2P (Chunky-to-Planar) Conversion
- *
- * Copyright (C) 2003 Geert Uytterhoeven
- *
- * NOTES:
- * - This code was inspired by Scout's C2P tutorial
- * - It assumes to run on a big endian system
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License. See the file COPYING in the main directory of this archive
- * for more details.
- */
-
-#include <linux/module.h>
-#include <linux/string.h>
-#include "c2p.h"
-
-
- /*
- * Basic transpose step
- */
-
-#define _transp(d, i1, i2, shift, mask) \
- do { \
- u32 t = (d[i1] ^ (d[i2] >> shift)) & mask; \
- d[i1] ^= t; \
- d[i2] ^= t << shift; \
- } while (0)
-
-static inline u32 get_mask(int n)
-{
- switch (n) {
- case 1:
- return 0x55555555;
- break;
-
- case 2:
- return 0x33333333;
- break;
-
- case 4:
- return 0x0f0f0f0f;
- break;
-
- case 8:
- return 0x00ff00ff;
- break;
-
- case 16:
- return 0x0000ffff;
- break;
- }
- return 0;
-}
-
-#define transp_nx1(d, n) \
- do { \
- u32 mask = get_mask(n); \
- /* First block */ \
- _transp(d, 0, 1, n, mask); \
- /* Second block */ \
- _transp(d, 2, 3, n, mask); \
- /* Third block */ \
- _transp(d, 4, 5, n, mask); \
- /* Fourth block */ \
- _transp(d, 6, 7, n, mask); \
- } while (0)
-
-#define transp_nx2(d, n) \
- do { \
- u32 mask = get_mask(n); \
- /* First block */ \
- _transp(d, 0, 2, n, mask); \
- _transp(d, 1, 3, n, mask); \
- /* Second block */ \
- _transp(d, 4, 6, n, mask); \
- _transp(d, 5, 7, n, mask); \
- } while (0)
-
-#define transp_nx4(d, n) \
- do { \
- u32 mask = get_mask(n); \
- _transp(d, 0, 4, n, mask); \
- _transp(d, 1, 5, n, mask); \
- _transp(d, 2, 6, n, mask); \
- _transp(d, 3, 7, n, mask); \
- } while (0)
-
-#define transp(d, n, m) transp_nx ## m(d, n)
-
-
- /*
- * Perform a full C2P step on 32 8-bit pixels, stored in 8 32-bit words
- * containing
- * - 32 8-bit chunky pixels on input
- * - permuted planar data on output
- */
-
-static void c2p_8bpp(u32 d[8])
-{
- transp(d, 16, 4);
- transp(d, 8, 2);
- transp(d, 4, 1);
- transp(d, 2, 4);
- transp(d, 1, 2);
-}
-
-
- /*
- * Array containing the permution indices of the planar data after c2p
- */
-
-static const int perm_c2p_8bpp[8] = { 7, 5, 3, 1, 6, 4, 2, 0 };
-
-
- /*
- * Compose two values, using a bitmask as decision value
- * This is equivalent to (a & mask) | (b & ~mask)
- */
-
-static inline unsigned long comp(unsigned long a, unsigned long b,
- unsigned long mask)
-{
- return ((a ^ b) & mask) ^ b;
-}
-
-
- /*
- * Store a full block of planar data after c2p conversion
- */
-
-static inline void store_planar(char *dst, u32 dst_inc, u32 bpp, u32 d[8])
-{
- int i;
-
- for (i = 0; i < bpp; i++, dst += dst_inc)
- *(u32 *)dst = d[perm_c2p_8bpp[i]];
-}
-
-
- /*
- * Store a partial block of planar data after c2p conversion
- */
-
-static inline void store_planar_masked(char *dst, u32 dst_inc, u32 bpp,
- u32 d[8], u32 mask)
-{
- int i;
-
- for (i = 0; i < bpp; i++, dst += dst_inc)
- *(u32 *)dst = comp(d[perm_c2p_8bpp[i]], *(u32 *)dst, mask);
-}
-
-
- /*
- * c2p - Copy 8-bit chunky image data to a planar frame buffer
- * @dst: Starting address of the planar frame buffer
- * @dx: Horizontal destination offset (in pixels)
- * @dy: Vertical destination offset (in pixels)
- * @width: Image width (in pixels)
- * @height: Image height (in pixels)
- * @dst_nextline: Frame buffer offset to the next line (in bytes)
- * @dst_nextplane: Frame buffer offset to the next plane (in bytes)
- * @src_nextline: Image offset to the next line (in bytes)
- * @bpp: Bits per pixel of the planar frame buffer (1-8)
- */
-
-void c2p(u8 *dst, const u8 *src, u32 dx, u32 dy, u32 width, u32 height,
- u32 dst_nextline, u32 dst_nextplane, u32 src_nextline, u32 bpp)
-{
- int dst_idx;
- u32 d[8], first, last, w;
- const u8 *c;
- u8 *p;
-
- dst += dy*dst_nextline+(dx & ~31);
- dst_idx = dx % 32;
- first = ~0UL >> dst_idx;
- last = ~(~0UL >> ((dst_idx+width) % 32));
- while (height--) {
- c = src;
- p = dst;
- w = width;
- if (dst_idx+width <= 32) {
- /* Single destination word */
- first &= last;
- memset(d, 0, sizeof(d));
- memcpy((u8 *)d+dst_idx, c, width);
- c += width;
- c2p_8bpp(d);
- store_planar_masked(p, dst_nextplane, bpp, d, first);
- p += 4;
- } else {
- /* Multiple destination words */
- w = width;
- /* Leading bits */
- if (dst_idx) {
- w = 32 - dst_idx;
- memset(d, 0, dst_idx);
- memcpy((u8 *)d+dst_idx, c, w);
- c += w;
- c2p_8bpp(d);
- store_planar_masked(p, dst_nextplane, bpp, d, first);
- p += 4;
- w = width-w;
- }
- /* Main chunk */
- while (w >= 32) {
- memcpy(d, c, 32);
- c += 32;
- c2p_8bpp(d);
- store_planar(p, dst_nextplane, bpp, d);
- p += 4;
- w -= 32;
- }
- /* Trailing bits */
- w %= 32;
- if (w > 0) {
- memcpy(d, c, w);
- memset((u8 *)d+w, 0, 32-w);
- c2p_8bpp(d);
- store_planar_masked(p, dst_nextplane, bpp, d, last);
- }
- }
- src += src_nextline;
- dst += dst_nextline;
- }
-}
-EXPORT_SYMBOL_GPL(c2p);
-
-MODULE_LICENSE("GPL");