armsx/ARMSX2
0
0
Fork 0
mirror of https://github.com/ARMSX2/ARMSX2.git synced 2026-08-24 16:50:16 -07:00
Code Issues Packages Projects Releases Wiki Activity

EE: the multiply predicate is one instruction, not three

Browse Source 
Parking the mask in d10 removed the Mov/Fmov that materialised it. The
boundary term's instruction-count comparison, a few lines below, was
never updated to match.
This commit is contained in:
pstef
2026-08-09 11:21:25 +02:00
parent 9f449b6dd2
commit 0a0633e551
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
pcsx2/arm64/iFPUd-arm64.cpp
+1 -1
View File
@@ -567,7 +567,7 @@ static void recFPUOp(int info, int eeRecDst, int op /*0=add,1=sub*/, bool acc)
// one-directional (it can only miss a deficit, never invent one) and rare in
// general operand space, per the count above. The term needs a bitfield extract
// NEON has no equivalent for, so it has to go through GPRs and come back --
// sketched at ten instructions against this predicate's three.
// sketched at ten instructions against this predicate's one.
// The resulting interpreter divergence is pinned by
// EeRecFpuFull.MulDefectDropsTheBoundaryTermTheInterpreterModels.
//