gh-mirror/mikepaul-LuaJIT
1
0
Fork 0
mirror of https://github.com/LuaJIT/LuaJIT.git synced 2025-02-07 15:14:08 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

ARM64: Fix LDP/STP fusion (again).

Browse Source 
Reported and analyzed by Zhongwei Yao. Fix by Peter Cawley. #1075
This commit is contained in:
Mike Pall 2023-09-09 18:01:37 +02:00
parent 0705ef6ce4
commit b8c6ccd50c
1 changed files with 13 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
src/lj_emit_arm64.h
View File

@ -113,6 +113,17 @@ static int emit_checkofs(A64Ins ai, int64_t ofs)
} }
} }
static LJ_AINLINE uint32_t emit_lso_pair_candidate(A64Ins ai, int ofs, int sc)
{
if (ofs >= 0) {
return ai | A64F_U12(ofs>>sc); /* Subsequent lj_ror checks ofs. */
} else if (ofs >= -256) {
return (ai^A64I_LS_U) | A64F_S9(ofs & 0x1ff);
} else {
return A64F_D(31); /* Will mismatch prev. */
}
}
static void emit_lso(ASMState *as, A64Ins ai, Reg rd, Reg rn, int64_t ofs) static void emit_lso(ASMState *as, A64Ins ai, Reg rd, Reg rn, int64_t ofs)
{ {
int ot = emit_checkofs(ai, ofs), sc = (ai >> 30) & 3; int ot = emit_checkofs(ai, ofs), sc = (ai >> 30) & 3;
@ -124,11 +135,9 @@ static void emit_lso(ASMState *as, A64Ins ai, Reg rd, Reg rn, int64_t ofs)
uint32_t prev = *as->mcp & ~A64F_D(31); uint32_t prev = *as->mcp & ~A64F_D(31);
int ofsm = ofs - (1<<sc), ofsp = ofs + (1<<sc); int ofsm = ofs - (1<<sc), ofsp = ofs + (1<<sc);
A64Ins aip; A64Ins aip;
if (prev == (ai | A64F_N(rn) | A64F_U12(ofsm>>sc)) || if (prev == emit_lso_pair_candidate(ai | A64F_N(rn), ofsm, sc)) {
prev == ((ai^A64I_LS_U) | A64F_N(rn) | A64F_S9(ofsm&0x1ff))) {
aip = (A64F_A(rd) | A64F_D(*as->mcp & 31)); aip = (A64F_A(rd) | A64F_D(*as->mcp & 31));
} else if (prev == (ai | A64F_N(rn) | A64F_U12(ofsp>>sc)) || } else if (prev == emit_lso_pair_candidate(ai | A64F_N(rn), ofsp, sc)) {
prev == ((ai^A64I_LS_U) | A64F_N(rn) | A64F_S9(ofsp&0x1ff))) {
aip = (A64F_D(rd) | A64F_A(*as->mcp & 31)); aip = (A64F_D(rd) | A64F_A(*as->mcp & 31));
ofsm = ofs; ofsm = ofs;
} else { } else {