JVET_O0078_SINGLE_HMVPLUT: Single HMVP table for all CUs inside the shared merge list region for IBC

  - Previous fix broke dependent  quantization for the non-scaling list case.
  - Negative shift values are no longer set to zero, so the shift is applied repeatedly to the fixed invQScale value (g_invQuantScales) during each loop.

  - Use different SIMD operations as requested.

  - Initialize some weighted prediction and slice address variables in case they are not set during slice header parsing.

  - Multiplication of "d" and "hv" terms overlows signed 32-bit data type.

Remy Foray authored 5 years ago and Karsten Suehring committed 5 years ago

Fix RPL creation in case of IDR, or when getNumRPL0() return 1. Fix condition to parse ref_pic_list_sps_flag[1]

removes the last remains of HEVC related code

- also some minor white space alignment for JVET_O macros

JVET-O0055 integer-distance DMVR cost to disable BDOF and disable BDOF early-termination in all cases

JVET-O0047/O0058/O0064/O0067/O0290/O0301/O0430: use fixed-length codes (2 bits) for alf clipping indices, always signal alf clipping indices

Remy Foray authored 5 years ago and Frank Bossen committed 5 years ago

Add the missing condition on MaxNumTriangleMergeCand to derive MergeTriangleFlag

Extended the HAD calculation with SIMD to be performed in 32-bit space rather than 16-bit. This fixes #339