indeo3: use unaligned reads on reference blocks.

They are not guaranteed to be aligned.
Fixes Bug 503.

CC:libav-stable@libav.org
This commit is contained in:
Anton Khirnov 2013-05-04 09:51:17 +02:00
parent 70f9661542
commit a97d8cc16e
1 changed files with 20 additions and 20 deletions

View File

@ -276,10 +276,10 @@ static int copy_cell(Indeo3DecodeContext *ctx, Plane *plane, Cell *cell)
/* Average 4/8 pixels at once without rounding using SWAR */ /* Average 4/8 pixels at once without rounding using SWAR */
#define AVG_32(dst, src, ref) \ #define AVG_32(dst, src, ref) \
AV_WN32A(dst, ((AV_RN32A(src) + AV_RN32A(ref)) >> 1) & 0x7F7F7F7FUL) AV_WN32A(dst, ((AV_RN32(src) + AV_RN32(ref)) >> 1) & 0x7F7F7F7FUL)
#define AVG_64(dst, src, ref) \ #define AVG_64(dst, src, ref) \
AV_WN64A(dst, ((AV_RN64A(src) + AV_RN64A(ref)) >> 1) & 0x7F7F7F7F7F7F7F7FULL) AV_WN64A(dst, ((AV_RN64(src) + AV_RN64(ref)) >> 1) & 0x7F7F7F7F7F7F7F7FULL)
/* /*
@ -338,7 +338,7 @@ if (*data_ptr >= last_ptr) \
ctx->hdsp.put_pixels_tab[2][0](dst, ref, row_offset, 4 << v_zoom) ctx->hdsp.put_pixels_tab[2][0](dst, ref, row_offset, 4 << v_zoom)
#define RLE_BLOCK_COPY_8 \ #define RLE_BLOCK_COPY_8 \
pix64 = AV_RN64A(ref);\ pix64 = AV_RN64(ref);\
if (is_first_row) {/* special prediction case: top line of a cell */\ if (is_first_row) {/* special prediction case: top line of a cell */\
pix64 = replicate64(pix64);\ pix64 = replicate64(pix64);\
fill_64(dst + row_offset, pix64, 7, row_offset);\ fill_64(dst + row_offset, pix64, 7, row_offset);\
@ -350,7 +350,7 @@ if (*data_ptr >= last_ptr) \
ctx->hdsp.put_pixels_tab[2][0](dst, ref, row_offset, num_lines << v_zoom) ctx->hdsp.put_pixels_tab[2][0](dst, ref, row_offset, num_lines << v_zoom)
#define RLE_LINES_COPY_M10 \ #define RLE_LINES_COPY_M10 \
pix64 = AV_RN64A(ref);\ pix64 = AV_RN64(ref);\
if (is_top_of_cell) {\ if (is_top_of_cell) {\
pix64 = replicate64(pix64);\ pix64 = replicate64(pix64);\
fill_64(dst + row_offset, pix64, (num_lines << 1) - 1, row_offset);\ fill_64(dst + row_offset, pix64, (num_lines << 1) - 1, row_offset);\
@ -360,12 +360,12 @@ if (*data_ptr >= last_ptr) \
#define APPLY_DELTA_4 \ #define APPLY_DELTA_4 \
AV_WN16A(dst + line_offset ,\ AV_WN16A(dst + line_offset ,\
(AV_RN16A(ref ) + delta_tab->deltas[dyad1]) & 0x7F7F);\ (AV_RN16(ref ) + delta_tab->deltas[dyad1]) & 0x7F7F);\
AV_WN16A(dst + line_offset + 2,\ AV_WN16A(dst + line_offset + 2,\
(AV_RN16A(ref + 2) + delta_tab->deltas[dyad2]) & 0x7F7F);\ (AV_RN16(ref + 2) + delta_tab->deltas[dyad2]) & 0x7F7F);\
if (mode >= 3) {\ if (mode >= 3) {\
if (is_top_of_cell && !cell->ypos) {\ if (is_top_of_cell && !cell->ypos) {\
AV_COPY32(dst, dst + row_offset);\ AV_COPY32U(dst, dst + row_offset);\
} else {\ } else {\
AVG_32(dst, ref, dst + row_offset);\ AVG_32(dst, ref, dst + row_offset);\
}\ }\
@ -375,20 +375,20 @@ if (*data_ptr >= last_ptr) \
/* apply two 32-bit VQ deltas to next even line */\ /* apply two 32-bit VQ deltas to next even line */\
if (is_top_of_cell) { \ if (is_top_of_cell) { \
AV_WN32A(dst + row_offset , \ AV_WN32A(dst + row_offset , \
(replicate32(AV_RN32A(ref )) + delta_tab->deltas_m10[dyad1]) & 0x7F7F7F7F);\ (replicate32(AV_RN32(ref )) + delta_tab->deltas_m10[dyad1]) & 0x7F7F7F7F);\
AV_WN32A(dst + row_offset + 4, \ AV_WN32A(dst + row_offset + 4, \
(replicate32(AV_RN32A(ref + 4)) + delta_tab->deltas_m10[dyad2]) & 0x7F7F7F7F);\ (replicate32(AV_RN32(ref + 4)) + delta_tab->deltas_m10[dyad2]) & 0x7F7F7F7F);\
} else { \ } else { \
AV_WN32A(dst + row_offset , \ AV_WN32A(dst + row_offset , \
(AV_RN32A(ref ) + delta_tab->deltas_m10[dyad1]) & 0x7F7F7F7F);\ (AV_RN32(ref ) + delta_tab->deltas_m10[dyad1]) & 0x7F7F7F7F);\
AV_WN32A(dst + row_offset + 4, \ AV_WN32A(dst + row_offset + 4, \
(AV_RN32A(ref + 4) + delta_tab->deltas_m10[dyad2]) & 0x7F7F7F7F);\ (AV_RN32(ref + 4) + delta_tab->deltas_m10[dyad2]) & 0x7F7F7F7F);\
} \ } \
/* odd lines are not coded but rather interpolated/replicated */\ /* odd lines are not coded but rather interpolated/replicated */\
/* first line of the cell on the top of image? - replicate */\ /* first line of the cell on the top of image? - replicate */\
/* otherwise - interpolate */\ /* otherwise - interpolate */\
if (is_top_of_cell && !cell->ypos) {\ if (is_top_of_cell && !cell->ypos) {\
AV_COPY64(dst, dst + row_offset);\ AV_COPY64U(dst, dst + row_offset);\
} else \ } else \
AVG_64(dst, ref, dst + row_offset); AVG_64(dst, ref, dst + row_offset);
@ -396,22 +396,22 @@ if (*data_ptr >= last_ptr) \
#define APPLY_DELTA_1011_INTER \ #define APPLY_DELTA_1011_INTER \
if (mode == 10) { \ if (mode == 10) { \
AV_WN32A(dst , \ AV_WN32A(dst , \
(AV_RN32A(dst ) + delta_tab->deltas_m10[dyad1]) & 0x7F7F7F7F);\ (AV_RN32(dst ) + delta_tab->deltas_m10[dyad1]) & 0x7F7F7F7F);\
AV_WN32A(dst + 4 , \ AV_WN32A(dst + 4 , \
(AV_RN32A(dst + 4 ) + delta_tab->deltas_m10[dyad2]) & 0x7F7F7F7F);\ (AV_RN32(dst + 4 ) + delta_tab->deltas_m10[dyad2]) & 0x7F7F7F7F);\
AV_WN32A(dst + row_offset , \ AV_WN32A(dst + row_offset , \
(AV_RN32A(dst + row_offset ) + delta_tab->deltas_m10[dyad1]) & 0x7F7F7F7F);\ (AV_RN32(dst + row_offset ) + delta_tab->deltas_m10[dyad1]) & 0x7F7F7F7F);\
AV_WN32A(dst + row_offset + 4, \ AV_WN32A(dst + row_offset + 4, \
(AV_RN32A(dst + row_offset + 4) + delta_tab->deltas_m10[dyad2]) & 0x7F7F7F7F);\ (AV_RN32(dst + row_offset + 4) + delta_tab->deltas_m10[dyad2]) & 0x7F7F7F7F);\
} else { \ } else { \
AV_WN16A(dst , \ AV_WN16A(dst , \
(AV_RN16A(dst ) + delta_tab->deltas[dyad1]) & 0x7F7F);\ (AV_RN16(dst ) + delta_tab->deltas[dyad1]) & 0x7F7F);\
AV_WN16A(dst + 2 , \ AV_WN16A(dst + 2 , \
(AV_RN16A(dst + 2 ) + delta_tab->deltas[dyad2]) & 0x7F7F);\ (AV_RN16(dst + 2 ) + delta_tab->deltas[dyad2]) & 0x7F7F);\
AV_WN16A(dst + row_offset , \ AV_WN16A(dst + row_offset , \
(AV_RN16A(dst + row_offset ) + delta_tab->deltas[dyad1]) & 0x7F7F);\ (AV_RN16(dst + row_offset ) + delta_tab->deltas[dyad1]) & 0x7F7F);\
AV_WN16A(dst + row_offset + 2, \ AV_WN16A(dst + row_offset + 2, \
(AV_RN16A(dst + row_offset + 2) + delta_tab->deltas[dyad2]) & 0x7F7F);\ (AV_RN16(dst + row_offset + 2) + delta_tab->deltas[dyad2]) & 0x7F7F);\
} }