void RSP_VNAND(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
rsp_vect_t result = rsp_vnand(LOAD_VS(), LOAD_VT());
write_acc_lo(acc, result);
STORE_RESULT();
}
Nd4jStatus LegacyScalarOp::validateAndExecute(Context &block) {
auto x = INPUT_VARIABLE(0);
int offset = 0;
auto z = OUTPUT_VARIABLE(0);
int opNum = block.opNum() < 0 ? this->_opNum : block.opNum();
if (block.width() > 1) {
auto y = INPUT_VARIABLE(1);
y->printIndexedBuffer("scalar");
// FIXME
NativeOpExcutioner::execScalar(opNum, x->getBuffer(), x->getShapeInfo(), z->getBuffer(), z->getShapeInfo(), y->buffer(), y->shapeInfo(), block.getTArguments()->data());
} else if (block.getTArguments()->size() > 0) {
auto y = NDArrayFactory::create(x->dataType(), T_ARG(0), block.getWorkspace());
offset++;
// FIXME
NativeOpExcutioner::execScalar(opNum, x->getBuffer(), x->getShapeInfo(), z->getBuffer(), z->getShapeInfo(), y.buffer(), y.shapeInfo(), block.getTArguments()->data() + offset);
} else {
// FIXME
NativeOpExcutioner::execScalar(opNum, x->getBuffer(), x->getShapeInfo(), z->getBuffer(), z->getShapeInfo(), _scalar.buffer(), _scalar.shapeInfo(), block.getTArguments()->data());
}
STORE_RESULT(*z);
return Status::OK();
}
//
// VMOV
//
void RSP_VMOV(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
unsigned de = vs & 0x7;
write_acc_lo(acc, LOAD_VT());
__m128i result = rsp_vmov(rsp, vt, e, vd, de);
STORE_RESULT();
}
static void rlf(void)
{
R.ALU = GET_REGFILE(ADDR);
R.ALU <<= 1;
if (R.STATUS & C_FLAG) R.ALU |= 1;
if (GET_REGFILE(ADDR) & 0x80) SET(C_FLAG);
else CLR(C_FLAG);
STORE_RESULT(ADDR, R.ALU);
}
static void rrf(void)
{
R.ALU = GET_REGFILE(ADDR);
R.ALU >>= 1;
if (R.STATUS & C_FLAG) R.ALU |= 0x80;
if (GET_REGFILE(ADDR) & 1) SET(C_FLAG);
else CLR(C_FLAG);
STORE_RESULT(ADDR, R.ALU);
}
static void subwf(void)
{
old_data = GET_REGFILE(ADDR);
R.ALU = old_data - R.W;
STORE_RESULT(ADDR, R.ALU);
CALCULATE_Z_FLAG();
CALCULATE_SUB_CARRY();
CALCULATE_SUB_DIGITCARRY();
}
void pic16c62x_device::subwf()
{
m_old_data = GET_REGFILE(ADDR);
m_ALU = m_old_data - m_W;
STORE_RESULT(ADDR, m_ALU);
CALCULATE_Z_FLAG();
CALCULATE_SUB_CARRY();
CALCULATE_SUB_DIGITCARRY();
}
void pic16c62x_device::rrf()
{
m_ALU = GET_REGFILE(ADDR);
m_ALU >>= 1;
if (STATUS & C_FLAG) m_ALU |= 0x80;
if (GET_REGFILE(ADDR) & 1) SET(STATUS, C_FLAG);
else CLR(STATUS, C_FLAG);
STORE_RESULT(ADDR, m_ALU);
}
static void addwf(void)
{
old_data = GET_REGFILE(ADDR);
R.ALU = old_data + R.W;
STORE_RESULT(ADDR, R.ALU);
CALCULATE_Z_FLAG();
CALCULATE_ADD_CARRY();
CALCULATE_ADD_DIGITCARRY();
}
void pic16c62x_device::incfsz()
{
m_ALU = GET_REGFILE(ADDR) + 1;
STORE_RESULT(ADDR, m_ALU);
if (m_ALU == 0)
{
m_PC++ ;
PCL = m_PC & 0xff;
m_inst_cycles += 1; /* Add NOP cycles */
}
}
//
// VADDC
//
void RSP_VADDC(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
rsp_vect_t sn;
rsp_vect_t result = rsp_vaddc(LOAD_VS(), LOAD_VT(), rsp_vzero(), &sn);
write_vco_hi(rsp->cp2.flags[RSP::RSP_VCO].e, rsp_vzero()); // TODO: Confirm.
write_vco_lo(rsp->cp2.flags[RSP::RSP_VCO].e, sn);
write_acc_lo(acc, result);
STORE_RESULT();
}
static void incfsz(void)
{
R.ALU = GET_REGFILE(ADDR) + 1;
STORE_RESULT(ADDR, R.ALU);
if (R.ALU == 0)
{
R.PC++ ;
R.PCL = R.PC & 0xff;
inst_cycles += cycles_000_other[0]; /* Add NOP cycles */
}
}
void RSP_VRSQ(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
unsigned de = vs & 0x7;
e &= 0x7;
write_acc_lo(acc, LOAD_VT());
rsp->cp2.dp_flag = 0;
rsp_vect_t result = rsp_vrcp_vrsq<true>(rsp, 0, vt, e, vd, de);
STORE_RESULT();
}
void RSP_VMULU(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
rsp_vect_t acc_lo, acc_md, acc_hi, result;
result = rsp_vmulf_vmulu<true>(LOAD_VS(), LOAD_VT(), rsp_vzero(), &acc_lo, &acc_md, &acc_hi);
write_acc_lo(acc, acc_lo);
write_acc_md(acc, acc_md);
write_acc_hi(acc, acc_hi);
STORE_RESULT();
}
//
// VMRG
//
void RSP_VMRG(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
rsp_vect_t le;
le = read_vcc_lo(rsp->cp2.flags[RSP::RSP_VCC].e);
rsp_vect_t result = rsp_vmrg(LOAD_VS(), LOAD_VT(), le);
write_vco_hi(rsp->cp2.flags[RSP::RSP_VCO].e, rsp_vzero());
write_vco_lo(rsp->cp2.flags[RSP::RSP_VCO].e, rsp_vzero());
write_acc_lo(acc, result);
STORE_RESULT();
}
//
// VSUBC
//
void RSP_VSUBC(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
rsp_vect_t eq, sn;
rsp_vect_t result = rsp_vsubc(LOAD_VS(), LOAD_VT(), rsp_vzero(), &eq, &sn);
write_vco_hi(rsp->cp2.flags[RSP::RSP_VCO].e, eq);
write_vco_lo(rsp->cp2.flags[RSP::RSP_VCO].e, sn);
write_acc_lo(acc, result);
STORE_RESULT();
}
//
// VADD
//
void RSP_VADD(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
rsp_vect_t carry, acc_lo;
carry = read_vco_lo(rsp->cp2.flags[RSP::RSP_VCO].e);
rsp_vect_t result = rsp_vadd(LOAD_VS(), LOAD_VT(), carry, &acc_lo);
write_vco_hi(rsp->cp2.flags[RSP::RSP_VCO].e, rsp_vzero());
write_vco_lo(rsp->cp2.flags[RSP::RSP_VCO].e, rsp_vzero());
write_acc_lo(acc, acc_lo);
STORE_RESULT();
}
void RSP_VRSQH(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
unsigned de = vs & 0x7;
e &= 0x7;
write_acc_lo(acc, LOAD_VT());
// Specify double-precision for VRCPL on the next pass.
rsp->cp2.dp_flag = 1;
rsp_vect_t result = rsp_vdivh(rsp, vt, e, vd, de);
STORE_RESULT();
}
//
// VSAR
//
void RSP_VSAR(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
rsp_vect_t result;
switch (e) {
case 8: result = read_acc_hi(acc); break;
case 9: result = read_acc_md(acc); break;
case 10: result = read_acc_lo(acc); break;
default: result = rsp_vzero(); break;
}
STORE_RESULT();
}
//
// VMACF
// VMACU
//
void RSP_VMACF(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
rsp_vect_t acc_lo, acc_md, acc_hi, result;
acc_lo = read_acc_lo(acc);
acc_md = read_acc_md(acc);
acc_hi = read_acc_hi(acc);
result = rsp_vmacf_vmacu<false>(LOAD_VS(), LOAD_VT(), rsp_vzero(), &acc_lo, &acc_md, &acc_hi);
write_acc_lo(acc, acc_lo);
write_acc_md(acc, acc_md);
write_acc_hi(acc, acc_hi);
STORE_RESULT();
}
//
// VCH
//
void RSP_VCH(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
rsp_vect_t ge, le, sign, eq, vce;
rsp_vect_t result = rsp_vch(LOAD_VS(), LOAD_VT(), rsp_vzero(), &ge, &le, &eq, &sign, &vce);
write_vcc_hi(rsp->cp2.flags[RSP::RSP_VCC].e, ge);
write_vcc_lo(rsp->cp2.flags[RSP::RSP_VCC].e, le);
write_vco_hi(rsp->cp2.flags[RSP::RSP_VCO].e, eq);
write_vco_lo(rsp->cp2.flags[RSP::RSP_VCO].e, sign);
write_vce (rsp->cp2.flags[RSP::RSP_VCE].e, vce);
write_acc_lo(acc, result);
STORE_RESULT();
}
Nd4jStatus LegacyPairwiseTransformOp::validateAndExecute(Context &block) {
auto x = INPUT_VARIABLE(0);
auto y = INPUT_VARIABLE(1);
auto z = OUTPUT_VARIABLE(0);
if (!x->isSameShape(y)) {
std::string sx = ShapeUtils::shapeAsString(x);
std::string sy = ShapeUtils::shapeAsString(y);
REQUIRE_TRUE(x->isSameShape(y) || y->isScalar(), 0, "Node_%i: For Pairwise transforms shapes of both operands should be equal but got %s vs %s", block.getNodeId(), sx.c_str(), sy.c_str());
}
int opNum = block.opNum() < 0 ? this->_opNum : block.opNum();
NativeOpExcutioner::execPairwiseTransform(opNum, x->getBuffer(), x->getShapeInfo(), y->getBuffer(), y->getShapeInfo(), z->getBuffer(), z->getShapeInfo(), block.getTArguments()->data());
STORE_RESULT(*z);
return ND4J_STATUS_OK;
}
//
// VCR
//
void RSP_VCR(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
rsp_vect_t ge, le;
rsp_vect_t result = rsp_vcr(LOAD_VS(), LOAD_VT(), rsp_vzero(), &ge, &le);
#ifdef INTENSE_DEBUG
for (unsigned i = 0; i < 8; i++)
fprintf(stderr, "VD[%d] = %d\n", i,
reinterpret_cast<int16_t*>(&result)[i]);
#endif
write_vcc_hi(rsp->cp2.flags[RSP::RSP_VCC].e, ge);
write_vcc_lo(rsp->cp2.flags[RSP::RSP_VCC].e, le);
write_vco_hi(rsp->cp2.flags[RSP::RSP_VCO].e, rsp_vzero());
write_vco_lo(rsp->cp2.flags[RSP::RSP_VCO].e, rsp_vzero());
write_vce (rsp->cp2.flags[RSP::RSP_VCE].e, rsp_vzero());
write_acc_lo(acc, result);
STORE_RESULT();
}
//
// VCL
//
void RSP_VCL(RSP::CPUState *rsp, unsigned vd, unsigned vs, unsigned vt, unsigned e)
{
uint16_t *acc = rsp->cp2.acc.e;
rsp_vect_t ge, le, eq, sign, vce;
ge = read_vcc_hi(rsp->cp2.flags[RSP::RSP_VCC].e);
le = read_vcc_lo(rsp->cp2.flags[RSP::RSP_VCC].e);
eq = read_vco_hi(rsp->cp2.flags[RSP::RSP_VCO].e);
sign = read_vco_lo(rsp->cp2.flags[RSP::RSP_VCO].e);
vce = read_vce(rsp->cp2.flags[RSP::RSP_VCE].e);
rsp_vect_t result = rsp_vcl(LOAD_VS(), LOAD_VT(), rsp_vzero(), &ge, &le, eq, sign, vce);
write_vcc_hi(rsp->cp2.flags[RSP::RSP_VCC].e, ge);
write_vcc_lo(rsp->cp2.flags[RSP::RSP_VCC].e, le);
write_vco_hi(rsp->cp2.flags[RSP::RSP_VCO].e, rsp_vzero());
write_vco_lo(rsp->cp2.flags[RSP::RSP_VCO].e, rsp_vzero());
write_vce (rsp->cp2.flags[RSP::RSP_VCE].e, rsp_vzero());
write_acc_lo(acc, result);
STORE_RESULT();
}
void pic16c62x_device::xorwf()
{
m_ALU = GET_REGFILE(ADDR) ^ m_W;
STORE_RESULT(ADDR, m_ALU);
CALCULATE_Z_FLAG();
}
void pic16c62x_device::swapf()
{
m_ALU = ((GET_REGFILE(ADDR) << 4) & 0xf0);
m_ALU |= ((GET_REGFILE(ADDR) >> 4) & 0x0f);
STORE_RESULT(ADDR, m_ALU);
}
static void swapf(void)
{
R.ALU = ((GET_REGFILE(ADDR) << 4) & 0xf0);
R.ALU |= ((GET_REGFILE(ADDR) >> 4) & 0x0f);
STORE_RESULT(ADDR, R.ALU);
}
static void xorwf(void)
{
R.ALU = GET_REGFILE(ADDR) ^ R.W;
STORE_RESULT(ADDR, R.ALU);
CALCULATE_Z_FLAG();
}
void pic16c62x_device::comf()
{
m_ALU = (UINT8)(~(GET_REGFILE(ADDR)));
STORE_RESULT(ADDR, m_ALU);
CALCULATE_Z_FLAG();
}
void pic16c62x_device::incf()
{
m_ALU = GET_REGFILE(ADDR) + 1;
STORE_RESULT(ADDR, m_ALU);
CALCULATE_Z_FLAG();
}
