Fixing overflow issue on extreme colors in color_distance().
Explicitly disabling unaligned addressing in miniz and transcoder when UBSAN is enabled
Disabling a needless optimization in transcoder causing ubsan issues (that could have easily been worked around in a different way, but it was an unnecessary optimization to begin with)
Rebuild wasm builds
- Minor fix to basisu_frontend::optimize_selector_codebook() so the new_selector_cluster_indices array is initialized correctly. (The backend doesn't use this array so no harm was being done.)
- Optimized ETC1S encoder (3-4.5x faster)
- Added optional SSE 4.1 support to encoder
- Switched from std::vector to a custom vector in the encoder and transcoder
- Added CppSPMD (SSE only for now)
- UASTC RDO is now more effective, but the command line parameter controlling qualiy has changed (to "lambda")
Some of our static analysis tools were complaining about unreachable code due to "#if ... return false; #endif code " blocks. This was changed to "#if ... return false #else ... #endif" which doesn't change the functionality but makes the static analyzers happy.
Also removed BASISU_NOTE_UNUSED on two variables that are no longer declared (resulting in a build error when BASISD_SUPPORT_UASTC was not defined)
Encoder is now a library in the "encoder" directory
JavaScript wrappers now expose the entire codec: encoder, transcoder, container independent transcoding, and .basis file information
(1 << num_bits) was converted to a signed integer that caused an overflow (as num_bits was usually set to 31). It worked fine since the value was later converted back to uint32_t but signed integer overflow is technically an undefined behavior and it was triggering errors in our automated tests.
With the default Xcode 11 settings, these non-functional changes remove warnings when building for iOS and Mac, notably:
Possible misuse of comma operator here
Implicit conversion loses integer precision: 'long' to 'int'
- Removing BC7 mode 6 support from transcoder (will alway use the much smaller mode 5 tables now)
- Adding BMP support
- Adding -bench mode, for development
- CMakeLists.txt file now compatible with Visual Studio 2019 (MSVC)
- Renaming -no_auto_global_sel_pal to -auto_global_sel_pal, disabling global selector palette usage by default to speed up compression on small images
- Renaming texture format enums
- Fixing uninitialized memory issue in front end - it wouldn't have affected the correctness of the output, or output quality
- Bumping command line tool up to v1.10
- For PVRTC1, we now memset() any extra blocks after the blocks we know will be transcoded into. This is done because in GL there is padding that occurs on textures smaller than 8x8.
- Removing all OpenMP usage and replacing it with a basis C++11 job system.
- Fixing TSVQ class so it can split very long "thin" clusters that before weren't being split. This improves endpoint quantization on videos and complex 2D textures.
- More frontend perf. optimizations.
Tested on OSX, Linux with gcc/clang, and Win using MSVC 2019.
This change makes it possible to exclude formats using compile time
options. BC7 is *never* exposed on the web and it has an absolutely huge
coding table which means we're just wasting space.
basis_transcoder.js 58707
basis_transcoder.wasm 224404
