[AArch64] Minor performance improvements to paired stores.

We can combine these two ST1 calls in to a single ST1 call.
Saves a few cycles in paired stores
This commit is contained in:
Ryan Houdek 2015-01-29 01:52:45 -06:00
parent 235f6c2445
commit 631c8a981e
1 changed files with 7 additions and 10 deletions

View File

@ -273,8 +273,7 @@ void JitArm64AsmRoutineManager::GenerateCommon()
float_emit.REV32(8, D0, D0);
MOVK(addr_reg, ((u64)Memory::base >> 32) & 0xFFFF, SHIFT_32);
float_emit.ST1(32, Q0, 0, addr_reg, SP);
float_emit.ST1(32, Q0, 1, addr_reg, SP);
float_emit.ST1(64, Q0, 0, addr_reg, SP);
RET(X30);
SetJumpTarget(argh);
@ -304,9 +303,9 @@ void JitArm64AsmRoutineManager::GenerateCommon()
TST(DecodeReg(addr_reg), 6, 1);
FixupBranch argh = B(CC_NEQ);
MOVK(addr_reg, ((u64)Memory::base >> 32) & 0xFFFF, SHIFT_32);
float_emit.ST1(8, Q0, 0, addr_reg, SP);
float_emit.ST1(8, Q0, 1, addr_reg, SP);
float_emit.ST1(16, Q0, 0, addr_reg, SP);
RET(X30);
SetJumpTarget(argh);
@ -335,9 +334,9 @@ void JitArm64AsmRoutineManager::GenerateCommon()
TST(DecodeReg(addr_reg), 6, 1);
FixupBranch argh = B(CC_NEQ);
MOVK(addr_reg, ((u64)Memory::base >> 32) & 0xFFFF, SHIFT_32);
float_emit.ST1(8, Q0, 0, addr_reg, SP);
float_emit.ST1(8, Q0, 1, addr_reg, SP);
float_emit.ST1(16, Q0, 0, addr_reg, SP);
RET(X30);
SetJumpTarget(argh);
@ -368,8 +367,7 @@ void JitArm64AsmRoutineManager::GenerateCommon()
TST(DecodeReg(addr_reg), 6, 1);
FixupBranch argh = B(CC_NEQ);
MOVK(addr_reg, ((u64)Memory::base >> 32) & 0xFFFF, SHIFT_32);
float_emit.ST1(16, Q0, 0, addr_reg, SP);
float_emit.ST1(16, Q0, 1, addr_reg, SP);
float_emit.ST1(32, Q0, 0, addr_reg, SP);
RET(X30);
SetJumpTarget(argh);
@ -399,8 +397,7 @@ void JitArm64AsmRoutineManager::GenerateCommon()
TST(DecodeReg(addr_reg), 6, 1);
FixupBranch argh = B(CC_NEQ);
MOVK(addr_reg, ((u64)Memory::base >> 32) & 0xFFFF, SHIFT_32);
float_emit.ST1(16, Q0, 0, addr_reg, SP);
float_emit.ST1(16, Q0, 1, addr_reg, SP);
float_emit.ST1(32, Q0, 0, addr_reg, SP);
RET(X30);
SetJumpTarget(argh);