In 9.2.3.2.4 Decoding process for binary decisions before termination, the sentence

When decoding end_of_sub_stream_one_bit, this last bit inserted in register codIOffset is interpreted as alignment_bit_equal_to_one.

should be added after

When decoding end_of_slice_segment_flag, this last bit inserted in register codIOffset is interpreted as rbsp_stop_one_bit.

The HM code (the latest version in the trunk) has an issue related to frame skipping with padding. When the input video size is not the multiple of 16 we enable the --PAD option. If at the same time we set the FrameSkip parameter to a non-zero value, it will cause a wrong framesize computation for skipping. The skip function uses m_iSourceWidth and m_iSourceHeight which are the values after padding. It should be changed to the original image width and height for skipping.

The bug located in file: TAppEncTop.cpp, line 226

in function Void TAppEncTop::xCreateLib()

current:

m_cTVideoIOYuvInputFile.skipFrames(m_FrameSkip, m_iSourceWidth, m_iSourceHeight);

should be:

m_cTVideoIOYuvInputFile.skipFrames(m_FrameSkip, m_iSourceWidth-m_aiPad[0], m_iSourceHeight-m_aiPad[1]);

When CU size is 8, m_aapucBS[iDir][0][uiBsAbsIdx] for some values of uiBsAbsIdx are accessed before being calculated.