- Do SSE2 optimization for direct-color sprite renders.
- Make ARM9_LCD cache-aligned. Allows for SSE2 to perform aligned load/stores on certain operations, improving performance.
- Further templatize some methods.
- Do some misc. code cleanup.
- Do heavy code cleanup.
- Split the engine-specific functionality of the main and sub engines into the new GPUEngineA and GPUEngineB subclasses.
- Templatize some parameters. Greatly increases the generated code size, but restores (and possibly improves) performance from r5251.
- Be smarter about manually inlining functions. Greatly reduces the generated code size, and fixes making optimized builds on MSVC. (Regression from r5248.)
- This change may affect performance. This will need additional testing.
- Add support for handling combination native/custom rendering sizes.
- As a side-effect of supporting this feature, pixel scalers now work as intended when high-resolution rendering is enabled (but only if the incoming display framebuffer is at the native size).
- Finish support for combination native/custom rendering sizes. Can give a significant performance improvement when running the GPUs at a custom size, but only for frontends that support this feature.
- Cleanup and optimize OAM attributes handling. (Special thanks to Twinaphex from libretro for pointing this out to us.)
- Add SSE2 optimizations to display capture operations.
- Do a whole bunch more code cleanup.
- Don’t evict the texture cache in the middle of geometry rendering! Fixes app crashing with games like Advance Wars: Days of Ruin that actually need to evict the texture cache. (Regression from r5175.)
- Do some minor code cleanup.
- Revert r5176 until polygon IDs can be handled correctly in one go. Fixes missing polygon issues in certain games such as missing rings in Sonic Chronicles: The Dark Brotherhood and missing loop traces in the Pokemon Ranger: Shadows of Almia title screen. (Addresses one of the issues noted in bug #1253.)
- Change toon highlight blending to match SoftRasterizer. Fixes the “Shadows of Almia” logo in the Pokemon Ranger: Shadows of Almia title screen. (Addresses one of the issues noted in bug #1253.)
- Revert the SSE2 bit shift optimizations that were done in r5216. Fixes a regression related to fog, as well as a regression that caused a flickering problem in the title screen of Pokemon Ranger: Shadows of Almia. (Fixes bug #1487.)
- In addition to the UI controls in the Show Video Settings panel, also add the “Use Vertical Sync” and “Run Filters on GPU” options to the View menu.
- Disable UI controls for Depth Comparison Threshold, since the setting is now obsolete. (Will need to delete UI controls before release.)
- Also add HQ3x/HQ3xS filters to the Pixel Scaler menu in Display Preferences.
- Fix bug where the HQ3x/HQ3xS filters running on the GPU sometimes wouldn’t draw correctly.
- Add new malloc_alignedN() functions for easier dynamic allocation of aligned memory blocks.
- Rework buffer allocations using the new malloc_alignedN() functions.
- To enable SSSE3, also require ENABLE_SSE2 and ENABLE_SSE3.
- Add some more SSE2/SSSE3 optimizations.
- CACHE_ALIGN and malloc_alignedCacheLine() now set 64 byte alignment on 64-bit systems.
- Do a bunch more code cleanup.
rogerman