YUV software decode path stride fixes.

media/base/video_frame.cc

Index: media/base/video_frame.cc
diff --git a/media/base/video_frame.cc b/media/base/video_frame.cc
index 3d764103f1965397fd1fff8c26d24f63df827609..fbd3f046f2adcec6e7e73518a481e51300737bf6 100644
--- a/media/base/video_frame.cc
+++ b/media/base/video_frame.cc
@@ -124,6 +124,7 @@ static const int kFrameSizeAlignment = 16;
 static const int kFramePadBytes = 15;
 
 void VideoFrame::AllocateRGB(size_t bytes_per_pixel) {
+  coded_size_.set_width(RoundUp(coded_size_.width(), 4));

[sheu, 2012/11/16 03:03:45]

This is kinda hacky.  I'm doing this basically because I don't want to have to
touch glStorePixeli(GL_UNPACK_ALIGNMENT), as if the coded size can be
arbitrarily aligned, then we have to set the unpack alignment to 1 when we do
the texture upload.

GL_EXT_unpack_subimage still gets bit by the unpack alignment.

   // Round up to align at least at a 16-byte boundary for each row.
   // This is sufficient for MMX and SSE2 reads (movq/movdqa).
   size_t bytes_per_row = RoundUp(coded_size_.width(),
@@ -144,19 +145,22 @@ void VideoFrame::AllocateRGB(size_t bytes_per_pixel) {
 
 void VideoFrame::AllocateYUV() {
   DCHECK(format_ == VideoFrame::YV12 || format_ == VideoFrame::YV16);
-  // Align Y rows at least at 16 byte boundaries.  The stride for both
-  // YV12 and YV16 is 1/2 of the stride of Y.  For YV12, every row of bytes for
-  // U and V applies to two rows of Y (one byte of UV for 4 bytes of Y), so in
-  // the case of YV12 the strides are identical for the same width surface, but
-  // the number of bytes allocated for YV12 is 1/2 the amount for U & V as
-  // YV16. We also round the height of the surface allocated to be an even
-  // number to avoid any potential of faulting by code that attempts to access
-  // the Y values of the final row, but assumes that the last row of U & V
-  // applies to a full two rows of Y.
+  coded_size_.set_width(RoundUp(coded_size_.width(), 4));
+  // Align Y rows at least at 32 byte boundaries, so the stride for both YV12
+  // and YV16 at 1/2 of the stride of Y is aligned to 16. For YV12, every row of
+  // bytes for U and V applies to two rows of Y (one byte of UV for 4 bytes of
+  // Y), so in the case of YV12 the strides are identical for the same width
+  // surface, but the number of bytes allocated for YV12 is 1/2 the amount for
+  // U & V as YV16. We also round the height of the surface allocated to be an
+  // even number to avoid any potential of faulting by code that attempts to
+  // access the Y values of the final row, but assumes that the last row of U &
+  // V applies to a full two rows of Y.
   size_t y_stride = RoundUp(row_bytes(VideoFrame::kYPlane),
-                            kFrameSizeAlignment);
-  size_t uv_stride = RoundUp(row_bytes(VideoFrame::kUPlane),
-                             kFrameSizeAlignment);
+                            kFrameSizeAlignment * 2);
+  // We keep the UV stride (and height below) proportional to the Y stride
+  // (and height), since we will be assuming an identical scale factor for
+  // stride (and height) for the Y and UV planes.
+  size_t uv_stride = y_stride / 2;
   // The *2 here is because some formats (e.g. h264) allow interlaced coding,
   // and then the size needs to be a multiple of two macroblocks (vertically).
   // See libavcodec/utils.c:avcodec_align_dimensions2().