static inline void
vtx(volatile union ixgbe_adv_tx_desc *txdp,
struct rte_mbuf **pkt, uint16_t nb_pkts, uint64_t flags)
{
int i;
for (i = 0; i < nb_pkts; ++i, ++txdp, ++pkt)
vtx1(txdp, *pkt, flags);
}
uint16_t
ixgbe_xmit_pkts_vec(void *tx_queue, struct rte_mbuf **tx_pkts,
uint16_t nb_pkts)
{
struct ixgbe_tx_queue *txq = (struct ixgbe_tx_queue *)tx_queue;
volatile union ixgbe_adv_tx_desc *txdp;
struct ixgbe_tx_entry_v *txep;
uint16_t n, nb_commit, tx_id;
uint64_t flags = DCMD_DTYP_FLAGS;
uint64_t rs = IXGBE_ADVTXD_DCMD_RS|DCMD_DTYP_FLAGS;
int i;
/* cross rx_thresh boundary is not allowed */
nb_pkts = RTE_MIN(nb_pkts, txq->tx_rs_thresh);
if (txq->nb_tx_free < txq->tx_free_thresh)
ixgbe_tx_free_bufs(txq);
nb_commit = nb_pkts = (uint16_t)RTE_MIN(txq->nb_tx_free, nb_pkts);
if (unlikely(nb_pkts == 0))
return 0;
tx_id = txq->tx_tail;
txdp = &txq->tx_ring[tx_id];
txep = &txq->sw_ring_v[tx_id];
txq->nb_tx_free = (uint16_t)(txq->nb_tx_free - nb_pkts);
n = (uint16_t)(txq->nb_tx_desc - tx_id);
if (nb_commit >= n) {
tx_backlog_entry(txep, tx_pkts, n);
for (i = 0; i < n - 1; ++i, ++tx_pkts, ++txdp)
vtx1(txdp, *tx_pkts, flags);
vtx1(txdp, *tx_pkts++, rs);
nb_commit = (uint16_t)(nb_commit - n);
tx_id = 0;
txq->tx_next_rs = (uint16_t)(txq->tx_rs_thresh - 1);
/* avoid reach the end of ring */
txdp = &(txq->tx_ring[tx_id]);
txep = &txq->sw_ring_v[tx_id];
}
tx_backlog_entry(txep, tx_pkts, nb_commit);
vtx(txdp, tx_pkts, nb_commit, flags);
tx_id = (uint16_t)(tx_id + nb_commit);
if (tx_id > txq->tx_next_rs) {
txq->tx_ring[txq->tx_next_rs].read.cmd_type_len |=
rte_cpu_to_le_32(IXGBE_ADVTXD_DCMD_RS);
txq->tx_next_rs = (uint16_t)(txq->tx_next_rs +
txq->tx_rs_thresh);
}
txq->tx_tail = tx_id;
IXGBE_PCI_REG_WRITE(txq->tdt_reg_addr, txq->tx_tail);
return nb_pkts;
}
uint16_t
i40e_xmit_fixed_burst_vec(void *tx_queue, struct rte_mbuf **tx_pkts,
uint16_t nb_pkts)
{
struct i40e_tx_queue *txq = (struct i40e_tx_queue *)tx_queue;
volatile struct i40e_tx_desc *txdp;
struct i40e_tx_entry *txep;
uint16_t n, nb_commit, tx_id;
uint64_t flags = I40E_TD_CMD;
uint64_t rs = I40E_TX_DESC_CMD_RS | I40E_TD_CMD;
int i;
/* cross rx_thresh boundary is not allowed */
nb_pkts = RTE_MIN(nb_pkts, txq->tx_rs_thresh);
if (txq->nb_tx_free < txq->tx_free_thresh)
i40e_tx_free_bufs(txq);
nb_commit = nb_pkts = (uint16_t)RTE_MIN(txq->nb_tx_free, nb_pkts);
if (unlikely(nb_pkts == 0))
return 0;
tx_id = txq->tx_tail;
txdp = &txq->tx_ring[tx_id];
txep = &txq->sw_ring[tx_id];
txq->nb_tx_free = (uint16_t)(txq->nb_tx_free - nb_pkts);
n = (uint16_t)(txq->nb_tx_desc - tx_id);
if (nb_commit >= n) {
tx_backlog_entry(txep, tx_pkts, n);
for (i = 0; i < n - 1; ++i, ++tx_pkts, ++txdp)
vtx1(txdp, *tx_pkts, flags);
vtx1(txdp, *tx_pkts++, rs);
nb_commit = (uint16_t)(nb_commit - n);
tx_id = 0;
txq->tx_next_rs = (uint16_t)(txq->tx_rs_thresh - 1);
/* avoid reach the end of ring */
txdp = &txq->tx_ring[tx_id];
txep = &txq->sw_ring[tx_id];
}
tx_backlog_entry(txep, tx_pkts, nb_commit);
vtx(txdp, tx_pkts, nb_commit, flags);
tx_id = (uint16_t)(tx_id + nb_commit);
if (tx_id > txq->tx_next_rs) {
txq->tx_ring[txq->tx_next_rs].cmd_type_offset_bsz |=
rte_cpu_to_le_64(((uint64_t)I40E_TX_DESC_CMD_RS) <<
I40E_TXD_QW1_CMD_SHIFT);
txq->tx_next_rs =
(uint16_t)(txq->tx_next_rs + txq->tx_rs_thresh);
}
txq->tx_tail = tx_id;
I40E_PCI_REG_WRITE(txq->qtx_tail, txq->tx_tail);
return nb_pkts;
}
uint16_t
fm10k_xmit_pkts_vec(void *tx_queue, struct rte_mbuf **tx_pkts,
uint16_t nb_pkts)
{
struct fm10k_tx_queue *txq = (struct fm10k_tx_queue *)tx_queue;
volatile struct fm10k_tx_desc *txdp;
struct rte_mbuf **txep;
uint16_t n, nb_commit, tx_id;
uint64_t flags = FM10K_TXD_FLAG_LAST;
uint64_t rs = FM10K_TXD_FLAG_RS | FM10K_TXD_FLAG_LAST;
int i;
/* cross rx_thresh boundary is not allowed */
nb_pkts = RTE_MIN(nb_pkts, txq->rs_thresh);
if (txq->nb_free < txq->free_thresh)
fm10k_tx_free_bufs(txq);
nb_commit = nb_pkts = (uint16_t)RTE_MIN(txq->nb_free, nb_pkts);
if (unlikely(nb_pkts == 0))
return 0;
tx_id = txq->next_free;
txdp = &txq->hw_ring[tx_id];
txep = &txq->sw_ring[tx_id];
txq->nb_free = (uint16_t)(txq->nb_free - nb_pkts);
n = (uint16_t)(txq->nb_desc - tx_id);
if (nb_commit >= n) {
tx_backlog_entry(txep, tx_pkts, n);
for (i = 0; i < n - 1; ++i, ++tx_pkts, ++txdp)
vtx1(txdp, *tx_pkts, flags);
vtx1(txdp, *tx_pkts++, rs);
nb_commit = (uint16_t)(nb_commit - n);
tx_id = 0;
txq->next_rs = (uint16_t)(txq->rs_thresh - 1);
/* avoid reach the end of ring */
txdp = &(txq->hw_ring[tx_id]);
txep = &txq->sw_ring[tx_id];
}
tx_backlog_entry(txep, tx_pkts, nb_commit);
vtx(txdp, tx_pkts, nb_commit, flags);
tx_id = (uint16_t)(tx_id + nb_commit);
if (tx_id > txq->next_rs) {
txq->hw_ring[txq->next_rs].flags |= FM10K_TXD_FLAG_RS;
txq->next_rs = (uint16_t)(txq->next_rs + txq->rs_thresh);
}
txq->next_free = tx_id;
FM10K_PCI_REG_WRITE(txq->tail_ptr, txq->next_free);
return nb_pkts;
}
void CaculateNorm::run()
{
const IndexType k = 36; //36;
int size = SampleSet::get_instance().size();
for ( int ii = 0 ; ii<size; ++ii){
Sample& smp = SampleSet::get_instance()[ii];
std::cout<< "caculate:"<< ii<<std::endl;
if( !smp.num_triangles()) //only have points
{
for ( IndexType i=0; i < smp.num_vertices(); i++ )
{
MatrixX3 k_nearest_verts(k, 3);
IndexType neighbours[k];
ScalarType dist[k];
smp.neighbours( i, k, neighbours, dist );
for ( IndexType j=0; j<k; j++ )
{
IndexType neighbour_idx = neighbours[j];
k_nearest_verts.row(j) << smp[neighbour_idx].x(), smp[neighbour_idx].y(), smp[neighbour_idx].z();
}
MatrixX3 vert_mean = k_nearest_verts.colwise().mean();
MatrixX3 Q(k, 3);
for ( IndexType j=0; j<k;j++)
{
Q.row(j) = k_nearest_verts.row(j) - vert_mean;
}
Matrix33 sigma = Q.transpose() * Q;
Eigen::EigenSolver<Matrix33> eigen_solver(sigma, Eigen::ComputeFullU | Eigen::ComputeFullV);
auto ev = eigen_solver.eigenvectors();
auto eval = eigen_solver.eigenvalues();
ScalarType tmp[3] = { eval(0).real(),eval(1).real(),eval(2).real() };
IndexType min_idx = std::min_element(tmp,tmp+3) - tmp;
NormalType nv;
nv(0) = (ev.col(min_idx))(0).real();
nv(1) = (ev.col(min_idx))(1).real();
nv(2) = (ev.col(min_idx))(2).real();
nv.normalize();
if ( (baseline_).dot(nv) < 0 )
{
nv = - nv;
}
smp[i].set_normal( nv );
}
}else
{ //has face
//auto& m_triangles = smp.triangle_array;
for ( IndexType i=0; i < smp.num_triangles(); ++i )
{
IndexType i_vetex1 = smp.getTriangle(i).get_i_vertex(0);
IndexType i_vetex2 = smp.getTriangle(i).get_i_vertex(1);
IndexType i_vetex3 = smp.getTriangle(i).get_i_vertex(2);
PointType vtx1(smp[i_vetex1].x(),smp[i_vetex1].y(),smp[i_vetex1].z());
PointType vtx2(smp[i_vetex2].x(),smp[i_vetex2].y(),smp[i_vetex2].z());
PointType vtx3(smp[i_vetex3].x(),smp[i_vetex3].y(),smp[i_vetex3].z());
PointType vector1 = vtx2 - vtx1;
PointType vector2 = vtx3 - vtx1;
vector1.normalize();
vector2.normalize();
PointType vector3 = vector1.cross(vector2); //get the normal of the triangle
vector3.normalize();
//Logger<<"vector1: "<<vector1(0)<<" "<<vector1(1)<<" "<<vector1(2)<<std::endl;
//Logger<<"vector2: "<<vector2(0)<<" "<<vector2(1)<<" "<<vector2(2)<<std::endl;
//Logger<<"vector3: "<<vector3(0)<<" "<<vector3(1)<<" "<<vector3(2)<<std::endl;
//assign the normal to all the vertex of the triangle
for(int x = 0 ; x<3;++x)
{
IndexType i_normal = smp.getTriangle(i).get_i_normal(x);
//Logger<<"norm: "<<smp[i_normal].nx()<<" "<<smp[i_normal].ny()<<" "<<smp[i_normal].nz()<<std::endl;
smp[i_normal].set_normal(
NormalType( smp[i_normal].nx() + vector3(0),smp[i_normal].ny() +vector3(1),smp[i_normal].nz()+vector3(2) ));
}
}
for ( IndexType i=0; i < smp.num_vertices(); i++ )
{
NormalType norm(smp[i].nx(),smp[i].ny(),smp[i].nz());
norm.normalize();
// Logger<<"norm: "<<norm(0)<<" "<<norm(1)<<" "<<norm(2)<<std::endl;
smp[i].set_normal(norm);
}
}
}
}