- Try to use REP MOVSB when the size of the transfer is above a certain threshold
- This threshold is determined by the ERMS and FSRM cpuid flags
- The threshold values are (roughly) taken from GLIBC
- A threshold of 0xFFFFFFFF indicates that the cpu has neither flag
- This is just not part of spec, there is no enforcement for multiple of block size for width or height of s3tc compressed images.
- This restriction does indeed exist for ASTC and ETC but we're not using those formats.
macOS defines uint64 as an alias of unsigned long long. Drop the length
modifiers to resolve the warning on macOS:
rpcs3/rpcs3/Emu/GDB.cpp:107:37: warning: format specifies type 'unsigned long *' but the argument has type 'u64 *' (aka 'unsigned long long *') [-Wformat]
sscanf_s(val.c_str(), "%" HEX_U64, &result);
~~~~~~~~~~ ^~~~~~~
Also use std::from_chars to implement hex_to_{u8,u32,u64} instead of
sscanf.
Drop AVX2 path for now, since it usually operates on small data.
Rely on automatic SSE vectorization on recent compilers.
Side refactoring on JIT.h to workaround weird conflict issue.
- Calculates precise Z value as requested by the game
- Works properly if the underlying Z format matches the PS3 1:1 but may cause minor problems otherwise
Optimize CPU capability tests for arch-tuned builds.
Separate streaming and non-streaming utilities.
Rewritten copy_data_swap_u32(_cmp) with AVX2 path.
- Replace old format-based detection with proper aspect test.
Explicit image aspect has been available for a long time, but older
code was not updated.
- Reverse the order of the bytes in the selection masks. Previously it was assumed that byte 0 would determine the output of bit 0, but byte 7 determines the output of bit 0.
Implement ppu_attr::has_mfvscr (partially, module-wide search).
If this instruction isn't found, allow to drop setting SAT flag.
It's based on presumption that only MFVSCR can retrieve SAT flag.
