## Summary
- fall back from database-applied `SPU Decoder: Recompiler (ASMJIT)` to
`Recompiler (LLVM)` on ARM64 builds during boot
- log the fallback in `fixup_settings()` instead of letting SPU
initialization fail later
## Problem
`Need for Speed: Most Wanted [BLES01659]` receives `SPU Decoder:
Recompiler (ASMJIT)` from the config database. On ARM64 builds that
decoder is unsupported, which caused RPCS3 to abort during SPU
initialization with `Unsupported spu decoder` while the UI remained on
the loading screen.
## Validation
- built RPCS3 locally on macOS Apple Silicon
- booted `BLES01659` through the database config path with the custom
per-game override temporarily removed
- verified the log now shows the ARM64 fallback warning and `SPU
Decoder: Recompiler (LLVM)` instead of the previous fatal error
- observed the title proceed past the old boot blocker into active RSX
program compilation
---------
Co-authored-by: Elad <18193363+elad335@users.noreply.github.com>
Prevent invoking more than utils::get_max_threads() active thrreads at a time.
Windows sees more than utils::get_max_threads() active threads as potentially malicious behavior and lowers the program's overall CPU time.
- LLVM can't handle the select pattern we use to avoid poison values for large shifts
- So let's just use the intrinsic to avoid poison values altogether for now
- https://github.com/llvm/llvm-project/issues/200698
Fully implements cellSysmodule and sets it to HLE by default.
There is an undocumented function that decrypts the OpenPSID. There
doesn't seem to be any information about this on the PS3 dev wiki or
anywhere else, so I thought I mention this separately.
Refactored `SHUFB` by moving the byteswap optimization prior to the target specific path. Also separated the shuffle and special index constants calculation into separate steps in order to reduce code duplication.
Added a number of TODOs as this refactor revealed some of optimizations opportunities. None of which has been added as this refactor intends to minimize new behavior.
The "Fill more block info" re-decode indexed result.data[(ia - lsa) / 41] instead of / 4 (its companion decode uses / 4), so it read the wrong word and corrupted the per-block reg_const / reg_val32 constant propagation.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Giga cleanup clears m_block_info for removed blocks but left their in-range edges in m_targets, so the register-state walk dereferenced a stale target and aborted ("Range check failed") - e.g. a brsl whose return is a stop-trap, hit in Skylanders Giants. Prune those edges at the m_targets cleanup; export out_target_list after the prune so callers see the cleaned map; keep the initiate_patterns and cond_next-walk guards as defense in depth.
Adds made-up-data regression tests (the exported-map test fails without the prune) and the rpcs3_test.vcxproj entry.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The `FREST` instruction uses a scalar lookup-table loop to calculate the
mantissa. Given that it's only 32 elements, we can load it into two
512-bit vectors and treat them as the sources to the `vperm2d` shuffle
instruction. This allows it to be calculated in parallel, improving its
performance.
The `FRSQEST` instruction uses a lookup-table to calculate both the
exponent and mantissa. The exponent follows a simple linear pattern and
thus can be calculated using integer arithmetic: `(exponent==0)? 0xFF :
190 - (exponent + 1) / 2`. This patch replaces the LUT with this method,
vectorizing its calculating and saving on 1kB.
The mantissa still relies on a LUT, which bottlenecks this instruction's
emulation despite the improvement.
