infernap12/jdk
1
0
Fork 0
mirror of https://github.com/openjdk/jdk.git synced 2026-01-28 12:09:14 +00:00
Code Issues Packages Projects Releases Wiki Activity

Added stubroutine code

Browse Source
This commit is contained in:
Ben Perez 2025-10-22 21:38:05 -04:00
parent c0c1493026
commit 63a4317332
1 changed files with 14 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
src/hotspot/cpu/aarch64/stubGenerator_aarch64.cpp
View File

@ -7217,13 +7217,13 @@ class StubGenerator: public StubCodeGenerator {
__ mov(low_01, Assembler::D, 0, mul_tmp);
__ umulh(mul_tmp, a_i, b_j);
__ mov(high_01, Assembler::D, 0, mul_tmp);
//mul64ToVec(a_i, b_j, low_01, high_01, 0);
__ ldr(b_j, Address(b, 8));
__ mul(mul_tmp, a_i, b_j);
__ mov(low_01, Assembler::D, 1, mul_tmp);
__ umulh(mul_tmp, a_i, b_j);
__ mov(high_01, Assembler::D, 1, mul_tmp);
//mul64ToVec(a_i, b_j, low_01, high_01, 1);
__ shl(high_01, __ T2D, high_01, shift1);
__ ushr(tmp, __ T2D, low_01, shift2);
__ orr(high_01, __ T2D, high_01, tmp);
@ -7234,13 +7234,13 @@ class StubGenerator: public StubCodeGenerator {
__ mov(low_23, Assembler::D, 0, mul_tmp);
__ umulh(mul_tmp, a_i, b_j);
__ mov(high_23, Assembler::D, 0, mul_tmp);
//mul64ToVec(a_i, b_j, low_23, high_23, 0);
__ ldr(b_j, Address(b, 24));
__ mul(mul_tmp, a_i, b_j);
__ mov(low_23, Assembler::D, 1, mul_tmp);
__ umulh(mul_tmp, a_i, b_j);
__ mov(high_23, Assembler::D, 1, mul_tmp);
//mul64ToVec(a_i, b_j, low_23, high_23, 1);
__ shl(high_23, __ T2D, high_23, shift1);
__ ushr(tmp, __ T2D, low_23, shift2);
__ orr(high_23, __ T2D, high_23, tmp);
@ -7251,7 +7251,7 @@ class StubGenerator: public StubCodeGenerator {
__ mov(low_4x, Assembler::D, 0, mul_tmp);
__ umulh(mul_tmp, a_i, b_j);
__ mov(high_4x, Assembler::D, 0, mul_tmp);
//mul64ToVec(a_i, b_j, low_4x, high_4x, 0);
__ shl(high_4x, __ T2D, high_4x, shift1);
__ ushr(tmp, __ T2D, low_4x, shift2);
__ orr(high_4x, __ T2D, high_4x, tmp);
@ -7275,13 +7275,13 @@ class StubGenerator: public StubCodeGenerator {
__ mov(modmul_low, Assembler::D, 0, mul_tmp);
__ umulh(mul_tmp, n, mod_j);
__ mov(modmul_high, Assembler::D, 0, mul_tmp);
//mul64ToVec(n, mod_j, modmul_low, modmul_high, 0);
__ ldr(mod_j, Address(mod_ptr, 8));
__ mul(mul_tmp, n, mod_j);
__ mov(modmul_low, Assembler::D, 1, mul_tmp);
__ umulh(mul_tmp, n, mod_j);
__ mov(modmul_high, Assembler::D, 1, mul_tmp);
//mul64ToVec(n, mod_j, modmul_low, modmul_high, 1);
__ shl(modmul_high, __ T2D, modmul_high, shift1);
__ ushr(tmp, __ T2D, modmul_low, shift2);
__ orr(modmul_high, __ T2D, modmul_high, tmp);
@ -7294,13 +7294,13 @@ class StubGenerator: public StubCodeGenerator {
__ mov(modmul_low, Assembler::D, 0, mul_tmp);
__ umulh(mul_tmp, n, mod_j);
__ mov(modmul_high, Assembler::D, 0, mul_tmp);
//mul64ToVec(n, mod_j, modmul_low, modmul_high, 0);
__ ldr(mod_j, Address(mod_ptr, 24));
__ mul(mul_tmp, n, mod_j);
__ mov(modmul_low, Assembler::D, 1, mul_tmp);
__ umulh(mul_tmp, n, mod_j);
__ mov(modmul_high, Assembler::D, 1, mul_tmp);
//mul64ToVec(n, mod_j, modmul_low, modmul_high, 1);
__ shl(modmul_high, __ T2D, modmul_high, shift1);
__ ushr(tmp, __ T2D, modmul_low, shift2);
__ orr(modmul_high, __ T2D, modmul_high, tmp);
@ -7313,7 +7313,7 @@ class StubGenerator: public StubCodeGenerator {
__ mov(modmul_low, Assembler::D, 0, mul_tmp);
__ umulh(mul_tmp, n, mod_j);
__ mov(modmul_high, Assembler::D, 0, mul_tmp);
//mul64ToVec(n, mod_j, modmul_low, modmul_high, 0);
__ shl(modmul_high, __ T2D, modmul_high, shift1);
__ ushr(tmp, __ T2D, modmul_low, shift2);
__ orr(modmul_high, __ T2D, modmul_high, tmp);
@ -7485,17 +7485,6 @@ class StubGenerator: public StubCodeGenerator {
return start;
}
// Multiply both 64 bit lanes in b
void mul64ToVec(Register a, Register b, FloatRegister low, FloatRegister high, int lane) {
Register tmp = r14;
__ mul(tmp, a, b);
__ mov(low, Assembler::D, lane, tmp);
__ umulh(tmp, a, b);
__ mov(high, Assembler::D, lane, tmp);
}
void bcax5(Register a0, Register a1, Register a2, Register a3, Register a4,
Register tmp0, Register tmp1, Register tmp2) {
__ bic(tmp0, a2, a1); // for a0
@ -12236,6 +12225,10 @@ class StubGenerator: public StubCodeGenerator {
StubRoutines::_chacha20Block = generate_chacha20Block_blockpar();
}
if (UseIntPolyIntrinsics) {
StubRoutines::_intpoly_montgomeryMult_P256 = generate_intpoly_montgomeryMult_P256();
}
if (UseKyberIntrinsics) {
StubRoutines::_kyberNtt = generate_kyberNtt();
StubRoutines::_kyberInverseNtt = generate_kyberInverseNtt();