generic-library/vpx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Fix invalid memory access on 2x downscale.

Browse Source 
The issue was discovered on bitstream with 2x vertical downscale. For
zero MVs, y_pad is set to 1 only when vertical convolution is
required. The original code assumes that for y_step_q4 == 32 we don't
perform vertical convolution. But vp9_setup_scale_factors_for_frame()
sets convolve functions so that when x_step and y_step are both not
equal to 16, convolve in both directions is performed. And convolve()
unconditionally subtracts one stride from source pointer when calls
convolve_horiz(). This leads to invalid memory access.

Change-Id: I882dfa6081a58e172b5ffa55842bfcd6727f10bf
This commit is contained in:
Alexander Voronov 2014-09-09 14:56:00 +04:00
parent c731d6a4f1
commit 29071a418e
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
vp9/common/vp9_reconinter.c
View File

@ -376,13 +376,13 @@ static void dec_build_inter_predictors(MACROBLOCKD *xd, int plane, int block,
int y1 = ((y0_16 + (h - 1) * ys) >> SUBPEL_BITS) + 1;
int x_pad = 0, y_pad = 0;
if (subpel_x || (sf->x_step_q4 & SUBPEL_MASK)) {
if (subpel_x || (sf->x_step_q4 != 16)) {
x0 -= VP9_INTERP_EXTEND - 1;
x1 += VP9_INTERP_EXTEND;
x_pad = 1;
}
if (subpel_y || (sf->y_step_q4 & SUBPEL_MASK)) {
if (subpel_y || (sf->y_step_q4 != 16)) {
y0 -= VP9_INTERP_EXTEND - 1;
y1 += VP9_INTERP_EXTEND;
y_pad = 1;