Pscf::Prdc::Cuda::VecOp Namespace Reference

Functions that perform element-wise vector operations on the GPU. More...

Functions
void	eqV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector assignment, a[i] = b[i], kernel wrapper (cudaReal).
void	eqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector assignment, a[i] = b[i], kernel wrapper (cudaComplex).
void	eqS (DeviceArray< cudaReal > &a, cudaReal const b, const int beginIdA, const int n)
	Vector assignment, a[i] = b, kernel wrapper (cudaReal).
void	eqS (DeviceArray< cudaComplex > &a, cudaComplex const b, const int beginIdA, const int n)
	Vector assignment, a[i] = b, kernel wrapper (cudaComplex).
void	addVV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaReal > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector addition, a[i] = b[i] + c[i], kernel wrapper (cudaReal).
void	addVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaComplex > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector addition, a[i] = b[i] + c[i], kernel wrapper (cudaComplex).
void	addVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaComplex > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector addition, a[i] = b[i] + c[i], kernel wrapper (mixed, b = real).
void	addVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaReal > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector addition, a[i] = b[i] + c[i], kernel wrapper (mixed, c = real).
void	addVS (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, cudaReal const c, const int beginIdA, const int beginIdB, const int n)
	Vector addition, a[i] = b[i] + c, kernel wrapper (cudaReal).
void	addVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaComplex const c, const int beginIdA, const int beginIdB, const int n)
	Vector addition, a[i] = b[i] + c, kernel wrapper (cudaComplex).
void	addVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, cudaComplex const c, const int beginIdA, const int beginIdB, const int n)
	Vector addition, a[i] = b[i] + c, kernel wrapper (mixed, b = real).
void	addVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaReal const c, const int beginIdA, const int beginIdB, const int n)
	Vector addition, a[i] = b[i] + c, kernel wrapper (mixed, c = real).
void	subVV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaReal > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (cudaReal).
void	subVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaComplex > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (cudaComplex).
void	subVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaComplex > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (mixed, b = real).
void	subVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaReal > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (mixed, c = real).
void	subVS (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, cudaReal const c, const int beginIdA, const int beginIdB, const int n)
	Vector subtraction, a[i] = b[i] - c, kernel wrapper (cudaReal).
void	subVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaComplex const c, const int beginIdA, const int beginIdB, const int n)
	Vector subtraction, a[i] = b[i] - c, kernel wrapper (cudaComplex).
void	subVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, cudaComplex const c, const int beginIdA, const int beginIdB, const int n)
	Vector subtraction, a[i] = b[i] - c, kernel wrapper (mixed, b = real).
void	subVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaReal const c, const int beginIdA, const int beginIdB, const int n)
	Vector subtraction, a[i] = b[i] - c, kernel wrapper (mixed, c = real).
void	mulVV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaReal > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector multiplication, a[i] = b[i] * c[i], kernel wrapper (cudaReal).
void	mulVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaComplex > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector multiplication, a[i] = b[i] * c[i], kernel wrapper (cudaComplex).
void	mulVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaComplex > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector multiplication, a[i]=b[i]*c[i], kernel wrapper (mixed, b = real).
void	mulVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaReal > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector multiplication, a[i]=b[i]*c[i], kernel wrapper (mixed, c = real).
void	mulVS (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, cudaReal const c, const int beginIdA, const int beginIdB, const int n)
	Vector multiplication, a[i] = b[i] * c, kernel wrapper (cudaReal).
void	mulVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaComplex const c, const int beginIdA, const int beginIdB, const int n)
	Vector multiplication, a[i] = b[i] * c, kernel wrapper (cudaComplex).
void	mulVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, cudaComplex const c, const int beginIdA, const int beginIdB, const int n)
	Vector multiplication, a[i] = b[i] * c, kernel wrapper (mixed, b = real).
void	mulVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaReal const c, const int beginIdA, const int beginIdB, const int n)
	Vector multiplication, a[i] = b[i] * c, kernel wrapper (mixed, c = real).
void	divVV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaReal > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector division, a[i] = b[i] / c[i], kernel wrapper (cudaReal).
void	divVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaReal > const &c, const int beginIdA, const int beginIdB, const int beginIdC, const int n)
	Vector division, a[i] = b[i] / c[i], kernel wrapper (mixed, c = real).
void	divVS (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, cudaReal const c, const int beginIdA, const int beginIdB, const int n)
	Vector division, a[i] = b[i] / c, kernel wrapper (cudaReal).
void	divVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaReal const c, const int beginIdA, const int beginIdB, const int n)
	Vector division, a[i] = b[i] / c, kernel wrapper (mixed, c = real).
void	expV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector exponentiation, a[i] = exp(b[i]), kernel wrapper (cudaReal).
void	expV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector exponentiation, a[i] = exp(b[i]), kernel wrapper (cudaComplex).
void	addEqV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector addition in-place, a[i] += b[i], kernel wrapper (cudaReal).
void	addEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector addition in-place, a[i] += b[i], kernel wrapper (cudaComplex).
void	addEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector addition in-place, a[i] += b[i], kernel wrapper (mixed, b = real).
void	addEqS (DeviceArray< cudaReal > &a, cudaReal const b, const int beginIdA, const int n)
	Vector addition in-place, a[i] += b, kernel wrapper (cudaReal).
void	addEqS (DeviceArray< cudaComplex > &a, cudaComplex const b, const int beginIdA, const int n)
	Vector addition in-place, a[i] += b, kernel wrapper (cudaComplex).
void	addEqS (DeviceArray< cudaComplex > &a, cudaReal const b, const int beginIdA, const int n)
	Vector addition in-place, a[i] += b, kernel wrapper (mixed, b = real).
void	subEqV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector subtraction in-place, a[i] -= b[i], kernel wrapper (cudaReal).
void	subEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector subtraction in-place, a[i] -= b[i], kernel wrapper (cudaComplex).
void	subEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector subtraction in-place, a[i]-=b[i], kernel wrapper (mixed, b = real).
void	subEqS (DeviceArray< cudaReal > &a, cudaReal const b, const int beginIdA, const int n)
	Vector subtraction in-place, a[i] -= b, kernel wrapper (cudaReal).
void	subEqS (DeviceArray< cudaComplex > &a, cudaComplex const b, const int beginIdA, const int n)
	Vector subtraction in-place, a[i] -= b, kernel wrapper (cudaComplex).
void	subEqS (DeviceArray< cudaComplex > &a, cudaReal const b, const int beginIdA, const int n)
	Vector subtraction in-place, a[i] -= b, kernel wrapper (mixed, b = real).
void	mulEqV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector multiplication in-place, a[i] *= b[i], kernel wrapper (cudaReal).
void	mulEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector multiplication in-place, a[i]*=b[i], kernel wrapper (cudaComplex).
void	mulEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector multiplication in-place, a[i]*=b[i], kernel wrapper (mixed, b=real).
void	mulEqS (DeviceArray< cudaReal > &a, cudaReal const b, const int beginIdA, const int n)
	Vector multiplication in-place, a[i] *= b, kernel wrapper (cudaReal).
void	mulEqS (DeviceArray< cudaComplex > &a, cudaComplex const b, const int beginIdA, const int n)
	Vector multiplication in-place, a[i] *= b, kernel wrapper (cudaComplex).
void	mulEqS (DeviceArray< cudaComplex > &a, cudaReal const b, const int beginIdA, const int n)
	Vector multiplication in-place, a[i]*=b, kernel wrapper (mixed, b = real).
void	divEqV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector division in-place, a[i] /= b[i], kernel wrapper (cudaReal).
void	divEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, const int beginIdA, const int beginIdB, const int n)
	Vector division in-place, a[i] /= b[i], kernel wrapper (mixed, b = real).
void	divEqS (DeviceArray< cudaReal > &a, cudaReal const b, const int beginIdA, const int n)
	Vector division in-place, a[i] /= b, kernel wrapper (cudaReal).
void	divEqS (DeviceArray< cudaComplex > &a, cudaReal const b, const int beginIdA, const int n)
	Vector division in-place, a[i] /= b, kernel wrapper (mixed, b = real).
void	eqV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b)
	Vector assignment, a[i] = b[i], kernel wrapper (cudaReal).
void	eqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b)
	Vector assignment, a[i] = b[i], kernel wrapper (cudaComplex).
void	eqS (DeviceArray< cudaReal > &a, cudaReal const b)
	Vector assignment, a[i] = b, kernel wrapper (cudaReal).
void	eqS (DeviceArray< cudaComplex > &a, cudaComplex const b)
	Vector assignment, a[i] = b, kernel wrapper (cudaComplex).
void	addVV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaReal > const &c)
	Vector addition, a[i] = b[i] + c[i], kernel wrapper (cudaReal).
void	addVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaComplex > const &c)
	Vector addition, a[i] = b[i] + c[i], kernel wrapper (cudaComplex).
void	addVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaComplex > const &c)
	Vector addition, a[i] = b[i] + c[i], kernel wrapper (mixed, b = real).
void	addVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaReal > const &c)
	Vector addition, a[i] = b[i] + c[i], kernel wrapper (mixed, c = real).
void	addVS (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, cudaReal const c)
	Vector addition, a[i] = b[i] + c, kernel wrapper (cudaReal).
void	addVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaComplex const c)
	Vector addition, a[i] = b[i] + c, kernel wrapper (cudaComplex).
void	addVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, cudaComplex const c)
	Vector addition, a[i] = b[i] + c, kernel wrapper (mixed, b = real).
void	addVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaReal const c)
	Vector addition, a[i] = b[i] + c, kernel wrapper (mixed, c = real).
void	subVV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaReal > const &c)
	Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (cudaReal).
void	subVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaComplex > const &c)
	Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (cudaComplex).
void	subVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaComplex > const &c)
	Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (mixed, b = real).
void	subVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaReal > const &c)
	Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (mixed, c = real).
void	subVS (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, cudaReal const c)
	Vector subtraction, a[i] = b[i] - c, kernel wrapper (cudaReal).
void	subVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaComplex const c)
	Vector subtraction, a[i] = b[i] - c, kernel wrapper (cudaComplex).
void	subVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, cudaComplex const c)
	Vector subtraction, a[i] = b[i] - c, kernel wrapper (mixed, b = real).
void	subVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaReal const c)
	Vector subtraction, a[i] = b[i] - c, kernel wrapper (mixed, c = real).
void	mulVV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaReal > const &c)
	Vector multiplication, a[i] = b[i] * c[i], kernel wrapper (cudaReal).
void	mulVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaComplex > const &c)
	Vector multiplication, a[i] = b[i] * c[i], kernel wrapper (cudaComplex).
void	mulVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaComplex > const &c)
	Vector multiplication, a[i]=b[i]*c[i], kernel wrapper (mixed, b = real).
void	mulVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaReal > const &c)
	Vector multiplication, a[i]=b[i]*c[i], kernel wrapper (mixed, c = real).
void	mulVS (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, cudaReal const c)
	Vector multiplication, a[i] = b[i] * c, kernel wrapper (cudaReal).
void	mulVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaComplex const c)
	Vector multiplication, a[i] = b[i] * c, kernel wrapper (cudaComplex).
void	mulVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, cudaComplex const c)
	Vector multiplication, a[i] = b[i] * c, kernel wrapper (mixed, b = real).
void	mulVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaReal const c)
	Vector multiplication, a[i] = b[i] * c, kernel wrapper (mixed, c = real).
void	divVV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaReal > const &c)
	Vector division, a[i] = b[i] / c[i], kernel wrapper (cudaReal).
void	divVV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, DeviceArray< cudaReal > const &c)
	Vector division, a[i] = b[i] / c[i], kernel wrapper (mixed, c = real).
void	divVS (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, cudaReal const c)
	Vector division, a[i] = b[i] / c, kernel wrapper (cudaReal).
void	divVS (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b, cudaReal const c)
	Vector division, a[i] = b[i] / c, kernel wrapper (mixed, c = real).
void	expV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b)
	Vector exponentiation, a[i] = exp(b[i]), kernel wrapper (cudaReal).
void	expV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b)
	Vector exponentiation, a[i] = exp(b[i]), kernel wrapper (cudaComplex).
void	addEqV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b)
	Vector addition in-place, a[i] += b[i], kernel wrapper (cudaReal).
void	addEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b)
	Vector addition in-place, a[i] += b[i], kernel wrapper (cudaComplex).
void	addEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b)
	Vector addition in-place, a[i] += b[i], kernel wrapper (mixed, b = real).
void	addEqS (DeviceArray< cudaReal > &a, cudaReal const b)
	Vector addition in-place, a[i] += b, kernel wrapper (cudaReal).
void	addEqS (DeviceArray< cudaComplex > &a, cudaComplex const b)
	Vector addition in-place, a[i] += b, kernel wrapper (cudaComplex).
void	addEqS (DeviceArray< cudaComplex > &a, cudaReal const b)
	Vector addition in-place, a[i] += b, kernel wrapper (mixed, b = real).
void	subEqV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b)
	Vector subtraction in-place, a[i] -= b[i], kernel wrapper (cudaReal).
void	subEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b)
	Vector subtraction in-place, a[i] -= b[i], kernel wrapper (cudaComplex).
void	subEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b)
	Vector subtraction in-place, a[i]-=b[i], kernel wrapper (mixed, b = real).
void	subEqS (DeviceArray< cudaReal > &a, cudaReal const b)
	Vector subtraction in-place, a[i] -= b, kernel wrapper (cudaReal).
void	subEqS (DeviceArray< cudaComplex > &a, cudaComplex const b)
	Vector subtraction in-place, a[i] -= b, kernel wrapper (cudaComplex).
void	subEqS (DeviceArray< cudaComplex > &a, cudaReal const b)
	Vector subtraction in-place, a[i] -= b, kernel wrapper (mixed, b = real).
void	mulEqV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b)
	Vector multiplication in-place, a[i] *= b[i], kernel wrapper (cudaReal).
void	mulEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaComplex > const &b)
	Vector multiplication in-place, a[i]*=b[i], kernel wrapper (cudaComplex).
void	mulEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b)
	Vector multiplication in-place, a[i]*=b[i], kernel wrapper (mixed, b=real).
void	mulEqS (DeviceArray< cudaReal > &a, cudaReal const b)
	Vector multiplication in-place, a[i] *= b, kernel wrapper (cudaReal).
void	mulEqS (DeviceArray< cudaComplex > &a, cudaComplex const b)
	Vector multiplication in-place, a[i] *= b, kernel wrapper (cudaComplex).
void	mulEqS (DeviceArray< cudaComplex > &a, cudaReal const b)
	Vector multiplication in-place, a[i]*=b, kernel wrapper (mixed, b = real).
void	divEqV (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b)
	Vector division in-place, a[i] /= b[i], kernel wrapper (cudaReal).
void	divEqV (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b)
	Vector division in-place, a[i] /= b[i], kernel wrapper (mixed, b = real).
void	divEqS (DeviceArray< cudaReal > &a, cudaReal const b)
	Vector division in-place, a[i] /= b, kernel wrapper (cudaReal).
void	divEqS (DeviceArray< cudaComplex > &a, cudaReal const b)
	Vector division in-place, a[i] /= b, kernel wrapper (mixed, b = real).
void	addVcVc (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, cudaReal const c, DeviceArray< cudaReal > const &d, cudaReal const e)
	Vector addition w/ coefficient, a[i] = (b[i]*c) + (d[i]*e), kernel wrapper.
void	addVcVcVc (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, cudaReal const c, DeviceArray< cudaReal > const &d, cudaReal const e, DeviceArray< cudaReal > const &f, cudaReal const g)
	3-vec addition w coeff, a[i] = (b[i]*c) + (d[i]*e) + (f[i]*g), kernel wrapper.
void	addEqVc (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, cudaReal const c)
	Vector addition in-place w/ coefficient, a[i] += b[i] * c, kernel wrapper.
void	subVVS (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, DeviceArray< cudaReal > const &c, cudaReal const d)
	Vector subtraction, a[i] = b[i] - c[i] - d, kernel wrapper.
void	divEqVc (DeviceArray< cudaComplex > &a, DeviceArray< cudaReal > const &b, cudaReal const c)
	Vector division in-place w/ coeff., a[i] /= (b[i] * c), kernel wrapper.
void	expVc (DeviceArray< cudaReal > &a, DeviceArray< cudaReal > const &b, cudaReal const c)
	Vector exponentiation w/ coefficient, a[i] = exp(b[i]*c), kernel wrapper.
void	eqVPair (DeviceArray< cudaReal > &a1, DeviceArray< cudaReal > &a2, DeviceArray< cudaReal > const &s)
	Vector assignment in pairs, ax[i] = b[i], kernel wrapper.
void	mulVVPair (DeviceArray< cudaReal > &a1, DeviceArray< cudaReal > &a2, DeviceArray< cudaReal > const &b1, DeviceArray< cudaReal > const &b2, DeviceArray< cudaReal > const &s)
	Vector multiplication in pairs, ax[i] = bx[i] * s[i], kernel wrapper.
void	mulEqVPair (DeviceArray< cudaReal > &a1, DeviceArray< cudaReal > &a2, DeviceArray< cudaReal > const &s)
	In-place vector multiplication in pairs, ax[i] *= s[i], kernel wrapper.
void	addVMany (DeviceArray< cudaReal > &a, DArray< DeviceArray< cudaReal > > const &vecs)
	Add an undefined number of vectors pointwise, kernel wrapper.
void	addVMany (DeviceArray< cudaReal > &a, DArray< DeviceArray< cudaReal > const * > const &vecs)
	Add an undefined number of vectors pointwise, kernel wrapper.
void	mulVMany (DeviceArray< cudaReal > &a, DArray< DeviceArray< cudaReal > > const &vecs)
	Multiply an undefined number of vectors pointwise, kernel wrapper.
void	mulVMany (DeviceArray< cudaReal > &a, DArray< DeviceArray< cudaReal > const * > const &vecs)
	Multiply an undefined number of vectors pointwise, kernel wrapper.
void	sqNormV (DeviceArray< cudaReal > &a, DeviceArray< cudaComplex > const &b)
	Squared norm of complex number, a[i] = norm(b[i])^2, kernel wrapper.
void	sqSqNormV (DeviceArray< cudaReal > &a, DeviceArray< cudaComplex > const &b)
	Norm of complex number to the 4th power, a[i] = norm(b[i])^4, kernel wrapper.

Detailed Description

Functions that perform element-wise vector operations on the GPU.

The functions defined in this file are wrappers for CUDA kernels that perform the actual vector operations. The kernels themselves are only intended to be called through their wrappers, so they are defined in an anonymous namespace in VecOp.cu.

The wrapper functions operate on DeviceArray objects containing elements of type cudaReal or cudaComplex. The operations that are performed by these functions include addition, subtraction, multiplication, division, exponentiation, and assignment. The function names will, correspondingly, begin with "add", "sub", "mul", "div", "exp", or "eq" to indicate the operation being performed. Functions are also included to perform compound assignment operations, i.e. those that are performed using +=, -=, *=, and /= in C++. These functions have names that begin with "addEq", "subEq", "mulEq", and "divEq", respectively.

The functions are overloaded to perform their respective operations on any combination of cudaReal and cudaComplex input arrays, except those that would result in dividing by a complex number.

The output (the LHS of the vector operation) will always be the first parameter passed to the function. The input argument(s) (on the RHS of the vector operation) may be vectors or scalars. If an argument is a vector (scalar), the function name will contain a V (S). For example, addVV(A,B,C) implements vector-vector addition A[i] = B[i] + C[i], while addVS(A,B,c) implements vector-scalar addition A[i] = B[i] + c in which c is a scalar that is added to every element of B. In operations involving both vectors and scalars, the vectors will always be listed first. So, for example, addVS exists, but addSV does not.

Two wrapper functions are provided for each vector operation:

• The first accepts only the output array and the necessary input arrays / scalars. In these functions, each input array must be at least as long as the output array, and the element-wise operation will be performed for every element of the output array. All arrays will be indexed starting at element 0.
• The second allows for vector operations to be performed using only subsections (slices) of the input and output arrays. These functions require additional parameters: one index for each array involved in the operation (output and input), representing the element of each array at which to begin the slice, and an integer n, representing the size of the slices. Before calling the CUDA kernel, these functions check to ensure that the slices do not contain any indices that exceed the length of the corresponding arrays.

Additional functions that perform multiple operations within a single kernel are defined in VecOpMisc, which are not comprehensive and are written as-needed during the development of this software. VecOpMisc.h is included at the end of VecOp.h so that any code that includes VecOp.h will also include VecOpMisc.h.

Function Documentation

eqV() [1/4]

void Pscf::Prdc::Cuda::VecOp::eqV	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector assignment, a[i] = b[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1020 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::RealMove< D >::attemptMove(), Pscf::Rpg::FourthOrderParameter< D >::compute(), Pscf::Rpg::MaxOrderParameter< D >::compute(), eqV(), eqV(), Pscf::Rpg::ForceBiasMove< D >::move(), Pscf::Rpg::Simulator< D >::restoreState(), Pscf::Rpg::Simulator< D >::saveState(), Pscf::Rpg::WFieldContainer< D >::setRGrid(), Pscf::Rpg::WFieldContainer< D >::setRGrid(), Pscf::Rpg::AmCompressor< D >::setup(), Pscf::Rpg::LrAmCompressor< D >::setup(), Pscf::Rpg::LrAmPreCompressor< D >::setup(), Pscf::Rpg::Propagator< D >::solve(), Pscf::Rpg::ExplicitBdStep< D >::step(), Pscf::Rpg::LMBdStep< D >::step(), Pscf::Rpg::PredCorrBdStep< D >::step(), and Pscf::Rpg::MaskGenFilm< D >::stressTerm().

eqV() [2/4]

void Pscf::Prdc::Cuda::VecOp::eqV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector assignment, a[i] = b[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1037 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

eqS() [1/4]

void Pscf::Prdc::Cuda::VecOp::eqS	(	DeviceArray< cudaReal > &	a,
		cudaReal const	b,
		const int	beginIdA,
		const int	n )

Vector assignment, a[i] = b, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1054 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::Mixture< D >::compute(), Pscf::Rpg::Simulator< D >::computeCc(), Pscf::Rpg::Block< D >::computeConcentration(), Pscf::Rpg::Propagator< D >::computeHead(), Pscf::Rpg::Simulator< D >::computeWc(), Pscf::Rpg::Mixture< D >::createBlockCRGrid(), eqS(), eqS(), and Pscf::Rpg::FourthOrderParameter< D >::setup().

eqS() [2/4]

void Pscf::Prdc::Cuda::VecOp::eqS	(	DeviceArray< cudaComplex > &	a,
		cudaComplex const	b,
		const int	beginIdA,
		const int	n )

Vector assignment, a[i] = b, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1069 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

addVV() [1/8]

void Pscf::Prdc::Cuda::VecOp::addVV	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		DeviceArray< cudaReal > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector addition, a[i] = b[i] + c[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1084 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by addVMany(), addVMany(), addVV(), addVV(), addVV(), and addVV().

addVV() [2/8]

void Pscf::Prdc::Cuda::VecOp::addVV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		DeviceArray< cudaComplex > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector addition, a[i] = b[i] + c[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1103 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

addVV() [3/8]

void Pscf::Prdc::Cuda::VecOp::addVV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaReal > const &	b,
		DeviceArray< cudaComplex > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector addition, a[i] = b[i] + c[i], kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1122 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

addVV() [4/8]

void Pscf::Prdc::Cuda::VecOp::addVV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		DeviceArray< cudaReal > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector addition, a[i] = b[i] + c[i], kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1141 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

addVS() [1/8]

void Pscf::Prdc::Cuda::VecOp::addVS	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		cudaReal const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector addition, a[i] = b[i] + c, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1160 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by addVS(), addVS(), addVS(), and addVS().

addVS() [2/8]

void Pscf::Prdc::Cuda::VecOp::addVS	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		cudaComplex const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector addition, a[i] = b[i] + c, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1177 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

addVS() [3/8]

void Pscf::Prdc::Cuda::VecOp::addVS	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaReal > const &	b,
		cudaComplex const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector addition, a[i] = b[i] + c, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1194 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

addVS() [4/8]

void Pscf::Prdc::Cuda::VecOp::addVS	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		cudaReal const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector addition, a[i] = b[i] + c, kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1211 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

subVV() [1/8]

void Pscf::Prdc::Cuda::VecOp::subVV	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		DeviceArray< cudaReal > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1228 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::EinsteinCrystalPerturbation< D >::hamiltonian(), Pscf::Rpg::EinsteinCrystalPerturbation< D >::incrementDc(), Pscf::Rpg::PredCorrBdStep< D >::step(), Pscf::Rpg::MaskGenFilm< D >::stressTerm(), subVV(), subVV(), subVV(), and subVV().

subVV() [2/8]

void Pscf::Prdc::Cuda::VecOp::subVV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		DeviceArray< cudaComplex > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1247 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

subVV() [3/8]

void Pscf::Prdc::Cuda::VecOp::subVV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaReal > const &	b,
		DeviceArray< cudaComplex > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1266 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

subVV() [4/8]

void Pscf::Prdc::Cuda::VecOp::subVV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		DeviceArray< cudaReal > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1285 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

subVS() [1/8]

void Pscf::Prdc::Cuda::VecOp::subVS	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		cudaReal const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector subtraction, a[i] = b[i] - c, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1304 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::LrAmPreCompressor< D >::computeError(), Pscf::Rpg::Simulator< D >::computeHamiltonian(), subVS(), subVS(), subVS(), and subVS().

subVS() [2/8]

void Pscf::Prdc::Cuda::VecOp::subVS	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		cudaComplex const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector subtraction, a[i] = b[i] - c, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1322 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

subVS() [3/8]

void Pscf::Prdc::Cuda::VecOp::subVS	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaReal > const &	b,
		cudaComplex const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector subtraction, a[i] = b[i] - c, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1340 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

subVS() [4/8]

void Pscf::Prdc::Cuda::VecOp::subVS	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		cudaReal const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector subtraction, a[i] = b[i] - c, kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1358 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

mulVV() [1/8]

void Pscf::Prdc::Cuda::VecOp::mulVV	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		DeviceArray< cudaReal > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector multiplication, a[i] = b[i] * c[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1376 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by mulVMany(), mulVMany(), mulVV(), mulVV(), mulVV(), and mulVV().

mulVV() [2/8]

void Pscf::Prdc::Cuda::VecOp::mulVV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		DeviceArray< cudaComplex > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector multiplication, a[i] = b[i] * c[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1395 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

mulVV() [3/8]

void Pscf::Prdc::Cuda::VecOp::mulVV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaReal > const &	b,
		DeviceArray< cudaComplex > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector multiplication, a[i]=b[i]*c[i], kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1414 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

mulVV() [4/8]

void Pscf::Prdc::Cuda::VecOp::mulVV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		DeviceArray< cudaReal > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector multiplication, a[i]=b[i]*c[i], kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1433 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

mulVS() [1/8]

void Pscf::Prdc::Cuda::VecOp::mulVS	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		cudaReal const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector multiplication, a[i] = b[i] * c, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1452 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by mulVS(), mulVS(), mulVS(), and mulVS().

mulVS() [2/8]

void Pscf::Prdc::Cuda::VecOp::mulVS	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		cudaComplex const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector multiplication, a[i] = b[i] * c, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1470 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

mulVS() [3/8]

void Pscf::Prdc::Cuda::VecOp::mulVS	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaReal > const &	b,
		cudaComplex const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector multiplication, a[i] = b[i] * c, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1488 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

mulVS() [4/8]

void Pscf::Prdc::Cuda::VecOp::mulVS	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		cudaReal const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector multiplication, a[i] = b[i] * c, kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1506 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

divVV() [1/4]

void Pscf::Prdc::Cuda::VecOp::divVV	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		DeviceArray< cudaReal > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector division, a[i] = b[i] / c[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1524 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by divVV(), and divVV().

divVV() [2/4]

void Pscf::Prdc::Cuda::VecOp::divVV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		DeviceArray< cudaReal > const &	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	beginIdC,
		const int	n )

Vector division, a[i] = b[i] / c[i], kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
beginIdC	index of the first entry to evaluate in array c
n	the number of entries to evaluate

Definition at line 1543 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

divVS() [1/4]

void Pscf::Prdc::Cuda::VecOp::divVS	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		cudaReal const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector division, a[i] = b[i] / c, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1562 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by divVS(), and divVS().

divVS() [2/4]

void Pscf::Prdc::Cuda::VecOp::divVS	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		cudaReal const	c,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector division, a[i] = b[i] / c, kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1580 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

expV() [1/4]

void Pscf::Prdc::Cuda::VecOp::expV	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector exponentiation, a[i] = exp(b[i]), kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1598 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by expV(), and expV().

expV() [2/4]

void Pscf::Prdc::Cuda::VecOp::expV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector exponentiation, a[i] = exp(b[i]), kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1615 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

addEqV() [1/6]

void Pscf::Prdc::Cuda::VecOp::addEqV	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector addition in-place, a[i] += b[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1632 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by addEqV(), addEqV(), addEqV(), Pscf::Rpg::Mixture< D >::compute(), Pscf::Rpg::LrAmPreCompressor< D >::computeError(), and Pscf::Rpg::PredCorrBdStep< D >::step().

addEqV() [2/6]

void Pscf::Prdc::Cuda::VecOp::addEqV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector addition in-place, a[i] += b[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1649 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

addEqV() [3/6]

void Pscf::Prdc::Cuda::VecOp::addEqV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaReal > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector addition in-place, a[i] += b[i], kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1666 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

addEqS() [1/6]

void Pscf::Prdc::Cuda::VecOp::addEqS	(	DeviceArray< cudaReal > &	a,
		cudaReal const	b,
		const int	beginIdA,
		const int	n )

Vector addition in-place, a[i] += b, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1683 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by addEqS(), addEqS(), and addEqS().

addEqS() [2/6]

void Pscf::Prdc::Cuda::VecOp::addEqS	(	DeviceArray< cudaComplex > &	a,
		cudaComplex const	b,
		const int	beginIdA,
		const int	n )

Vector addition in-place, a[i] += b, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1698 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

addEqS() [3/6]

void Pscf::Prdc::Cuda::VecOp::addEqS	(	DeviceArray< cudaComplex > &	a,
		cudaReal const	b,
		const int	beginIdA,
		const int	n )

Vector addition in-place, a[i] += b, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1713 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

subEqV() [1/6]

void Pscf::Prdc::Cuda::VecOp::subEqV	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector subtraction in-place, a[i] -= b[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1728 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by subEqV(), subEqV(), and subEqV().

subEqV() [2/6]

void Pscf::Prdc::Cuda::VecOp::subEqV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector subtraction in-place, a[i] -= b[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1745 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

subEqV() [3/6]

void Pscf::Prdc::Cuda::VecOp::subEqV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaReal > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector subtraction in-place, a[i]-=b[i], kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1762 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

subEqS() [1/6]

void Pscf::Prdc::Cuda::VecOp::subEqS	(	DeviceArray< cudaReal > &	a,
		cudaReal const	b,
		const int	beginIdA,
		const int	n )

Vector subtraction in-place, a[i] -= b, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1779 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by subEqS(), subEqS(), and subEqS().

subEqS() [2/6]

void Pscf::Prdc::Cuda::VecOp::subEqS	(	DeviceArray< cudaComplex > &	a,
		cudaComplex const	b,
		const int	beginIdA,
		const int	n )

Vector subtraction in-place, a[i] -= b, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1794 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

subEqS() [3/6]

void Pscf::Prdc::Cuda::VecOp::subEqS	(	DeviceArray< cudaComplex > &	a,
		cudaReal const	b,
		const int	beginIdA,
		const int	n )

Vector subtraction in-place, a[i] -= b, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1809 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

mulEqV() [1/6]

void Pscf::Prdc::Cuda::VecOp::mulEqV	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector multiplication in-place, a[i] *= b[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1824 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::FourthOrderParameter< D >::compute(), mulEqV(), mulEqV(), mulEqV(), and Pscf::Rpg::Block< D >::step().

mulEqV() [2/6]

void Pscf::Prdc::Cuda::VecOp::mulEqV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaComplex > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector multiplication in-place, a[i]*=b[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1841 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

mulEqV() [3/6]

void Pscf::Prdc::Cuda::VecOp::mulEqV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaReal > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector multiplication in-place, a[i]*=b[i], kernel wrapper (mixed, b=real).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1858 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

mulEqS() [1/6]

void Pscf::Prdc::Cuda::VecOp::mulEqS	(	DeviceArray< cudaReal > &	a,
		cudaReal const	b,
		const int	beginIdA,
		const int	n )

Vector multiplication in-place, a[i] *= b, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1875 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::Solvent< D >::compute(), Pscf::Rpg::Block< D >::computeConcentration(), Pscf::Prdc::Cuda::FFTBatched< D >::forwardTransform(), Pscf::Rpg::EinsteinCrystalPerturbation< D >::incrementDc(), mulEqS(), mulEqS(), mulEqS(), Pscf::Rpg::FieldIo< D >::scaleFieldRGrid(), and Pscf::Rpg::PredCorrBdStep< D >::step().

mulEqS() [2/6]

void Pscf::Prdc::Cuda::VecOp::mulEqS	(	DeviceArray< cudaComplex > &	a,
		cudaComplex const	b,
		const int	beginIdA,
		const int	n )

Vector multiplication in-place, a[i] *= b, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1890 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

mulEqS() [3/6]

void Pscf::Prdc::Cuda::VecOp::mulEqS	(	DeviceArray< cudaComplex > &	a,
		cudaReal const	b,
		const int	beginIdA,
		const int	n )

Vector multiplication in-place, a[i]*=b, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1905 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

divEqV() [1/4]

void Pscf::Prdc::Cuda::VecOp::divEqV	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector division in-place, a[i] /= b[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1920 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by divEqV(), and divEqV().

divEqV() [2/4]

void Pscf::Prdc::Cuda::VecOp::divEqV	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaReal > const &	b,
		const int	beginIdA,
		const int	beginIdB,
		const int	n )

Vector division in-place, a[i] /= b[i], kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
beginIdA	index of the first entry to evaluate in array a
beginIdB	index of the first entry to evaluate in array b
n	the number of entries to evaluate

Definition at line 1937 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

divEqS() [1/4]

void Pscf::Prdc::Cuda::VecOp::divEqS	(	DeviceArray< cudaReal > &	a,
		cudaReal const	b,
		const int	beginIdA,
		const int	n )

Vector division in-place, a[i] /= b, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1954 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by divEqS(), and divEqS().

divEqS() [2/4]

void Pscf::Prdc::Cuda::VecOp::divEqS	(	DeviceArray< cudaComplex > &	a,
		cudaReal const	b,
		const int	beginIdA,
		const int	n )

Vector division in-place, a[i] /= b, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input scalar (RHS)
beginIdA	index of the first entry to evaluate in array a
n	the number of entries to evaluate

Definition at line 1969 of file VecOp.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

eqV() [3/4]

void Pscf::Prdc::Cuda::VecOp::eqV ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b )

inline

Vector assignment, a[i] = b[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 98 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and eqV().

eqV() [4/4]

void Pscf::Prdc::Cuda::VecOp::eqV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b )

inline

Vector assignment, a[i] = b[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 119 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and eqV().

eqS() [3/4]

void Pscf::Prdc::Cuda::VecOp::eqS ( DeviceArray< cudaReal > & a, cudaReal const b )

inline

Vector assignment, a[i] = b, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 140 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and eqS().

eqS() [4/4]

void Pscf::Prdc::Cuda::VecOp::eqS ( DeviceArray< cudaComplex > & a, cudaComplex const b )

inline

Vector assignment, a[i] = b, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 160 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and eqS().

addVV() [5/8]

void Pscf::Prdc::Cuda::VecOp::addVV ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b, DeviceArray< cudaReal > const & c )

inline

Vector addition, a[i] = b[i] + c[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 189 of file VecOp.h.

References addVV(), and Pscf::DeviceArray< Data >::capacity().

addVV() [6/8]

void Pscf::Prdc::Cuda::VecOp::addVV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, DeviceArray< cudaComplex > const & c )

inline

Vector addition, a[i] = b[i] + c[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 215 of file VecOp.h.

References addVV(), and Pscf::DeviceArray< Data >::capacity().

addVV() [7/8]

void Pscf::Prdc::Cuda::VecOp::addVV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaReal > const & b, DeviceArray< cudaComplex > const & c )

inline

Vector addition, a[i] = b[i] + c[i], kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 242 of file VecOp.h.

References addVV(), and Pscf::DeviceArray< Data >::capacity().

addVV() [8/8]

void Pscf::Prdc::Cuda::VecOp::addVV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, DeviceArray< cudaReal > const & c )

inline

Vector addition, a[i] = b[i] + c[i], kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 269 of file VecOp.h.

References addVV(), and Pscf::DeviceArray< Data >::capacity().

addVS() [5/8]

void Pscf::Prdc::Cuda::VecOp::addVS ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b, cudaReal const c )

inline

Vector addition, a[i] = b[i] + c, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 295 of file VecOp.h.

References addVS(), and Pscf::DeviceArray< Data >::capacity().

addVS() [6/8]

void Pscf::Prdc::Cuda::VecOp::addVS ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, cudaComplex const c )

inline

Vector addition, a[i] = b[i] + c, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 320 of file VecOp.h.

References addVS(), and Pscf::DeviceArray< Data >::capacity().

addVS() [7/8]

void Pscf::Prdc::Cuda::VecOp::addVS ( DeviceArray< cudaComplex > & a, DeviceArray< cudaReal > const & b, cudaComplex const c )

inline

Vector addition, a[i] = b[i] + c, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 345 of file VecOp.h.

References addVS(), and Pscf::DeviceArray< Data >::capacity().

addVS() [8/8]

void Pscf::Prdc::Cuda::VecOp::addVS ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, cudaReal const c )

inline

Vector addition, a[i] = b[i] + c, kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 370 of file VecOp.h.

References addVS(), and Pscf::DeviceArray< Data >::capacity().

subVV() [5/8]

void Pscf::Prdc::Cuda::VecOp::subVV ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b, DeviceArray< cudaReal > const & c )

inline

Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 400 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subVV().

subVV() [6/8]

void Pscf::Prdc::Cuda::VecOp::subVV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, DeviceArray< cudaComplex > const & c )

inline

Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 426 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subVV().

subVV() [7/8]

void Pscf::Prdc::Cuda::VecOp::subVV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaReal > const & b, DeviceArray< cudaComplex > const & c )

inline

Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 453 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subVV().

subVV() [8/8]

void Pscf::Prdc::Cuda::VecOp::subVV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, DeviceArray< cudaReal > const & c )

inline

Vector subtraction, a[i] = b[i] - c[i], kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 479 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subVV().

subVS() [5/8]

void Pscf::Prdc::Cuda::VecOp::subVS ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b, cudaReal const c )

inline

Vector subtraction, a[i] = b[i] - c, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 505 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subVS().

subVS() [6/8]

void Pscf::Prdc::Cuda::VecOp::subVS ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, cudaComplex const c )

inline

Vector subtraction, a[i] = b[i] - c, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 530 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subVS().

subVS() [7/8]

void Pscf::Prdc::Cuda::VecOp::subVS ( DeviceArray< cudaComplex > & a, DeviceArray< cudaReal > const & b, cudaComplex const c )

inline

Vector subtraction, a[i] = b[i] - c, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 555 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subVS().

subVS() [8/8]

void Pscf::Prdc::Cuda::VecOp::subVS ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, cudaReal const c )

inline

Vector subtraction, a[i] = b[i] - c, kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 581 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subVS().

mulVV() [5/8]

void Pscf::Prdc::Cuda::VecOp::mulVV ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b, DeviceArray< cudaReal > const & c )

inline

Vector multiplication, a[i] = b[i] * c[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 611 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulVV().

mulVV() [6/8]

void Pscf::Prdc::Cuda::VecOp::mulVV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, DeviceArray< cudaComplex > const & c )

inline

Vector multiplication, a[i] = b[i] * c[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 637 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulVV().

mulVV() [7/8]

void Pscf::Prdc::Cuda::VecOp::mulVV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaReal > const & b, DeviceArray< cudaComplex > const & c )

inline

Vector multiplication, a[i]=b[i]*c[i], kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 664 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulVV().

mulVV() [8/8]

void Pscf::Prdc::Cuda::VecOp::mulVV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, DeviceArray< cudaReal > const & c )

inline

Vector multiplication, a[i]=b[i]*c[i], kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 691 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulVV().

mulVS() [5/8]

void Pscf::Prdc::Cuda::VecOp::mulVS ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b, cudaReal const c )

inline

Vector multiplication, a[i] = b[i] * c, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 717 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulVS().

mulVS() [6/8]

void Pscf::Prdc::Cuda::VecOp::mulVS ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, cudaComplex const c )

inline

Vector multiplication, a[i] = b[i] * c, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 742 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulVS().

mulVS() [7/8]

void Pscf::Prdc::Cuda::VecOp::mulVS ( DeviceArray< cudaComplex > & a, DeviceArray< cudaReal > const & b, cudaComplex const c )

inline

Vector multiplication, a[i] = b[i] * c, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 767 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulVS().

mulVS() [8/8]

void Pscf::Prdc::Cuda::VecOp::mulVS ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, cudaReal const c )

inline

Vector multiplication, a[i] = b[i] * c, kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 792 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulVS().

divVV() [3/4]

void Pscf::Prdc::Cuda::VecOp::divVV ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b, DeviceArray< cudaReal > const & c )

inline

Vector division, a[i] = b[i] / c[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 822 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and divVV().

divVV() [4/4]

void Pscf::Prdc::Cuda::VecOp::divVV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, DeviceArray< cudaReal > const & c )

inline

Vector division, a[i] = b[i] / c[i], kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)

Definition at line 848 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and divVV().

divVS() [3/4]

void Pscf::Prdc::Cuda::VecOp::divVS ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b, cudaReal const c )

inline

Vector division, a[i] = b[i] / c, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 874 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and divVS().

divVS() [4/4]

void Pscf::Prdc::Cuda::VecOp::divVS ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b, cudaReal const c )

inline

Vector division, a[i] = b[i] / c, kernel wrapper (mixed, c = real).

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 899 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and divVS().

expV() [3/4]

void Pscf::Prdc::Cuda::VecOp::expV ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b )

inline

Vector exponentiation, a[i] = exp(b[i]), kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 925 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and expV().

expV() [4/4]

void Pscf::Prdc::Cuda::VecOp::expV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b )

inline

Vector exponentiation, a[i] = exp(b[i]), kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 946 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and expV().

addEqV() [4/6]

void Pscf::Prdc::Cuda::VecOp::addEqV ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b )

inline

Vector addition in-place, a[i] += b[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 972 of file VecOp.h.

References addEqV(), and Pscf::DeviceArray< Data >::capacity().

addEqV() [5/6]

void Pscf::Prdc::Cuda::VecOp::addEqV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b )

inline

Vector addition in-place, a[i] += b[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 993 of file VecOp.h.

References addEqV(), and Pscf::DeviceArray< Data >::capacity().

addEqV() [6/6]

void Pscf::Prdc::Cuda::VecOp::addEqV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaReal > const & b )

inline

Vector addition in-place, a[i] += b[i], kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 1015 of file VecOp.h.

References addEqV(), and Pscf::DeviceArray< Data >::capacity().

addEqS() [4/6]

void Pscf::Prdc::Cuda::VecOp::addEqS ( DeviceArray< cudaReal > & a, cudaReal const b )

inline

Vector addition in-place, a[i] += b, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 1036 of file VecOp.h.

References addEqS(), and Pscf::DeviceArray< Data >::capacity().

addEqS() [5/6]

void Pscf::Prdc::Cuda::VecOp::addEqS ( DeviceArray< cudaComplex > & a, cudaComplex const b )

inline

Vector addition in-place, a[i] += b, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 1056 of file VecOp.h.

References addEqS(), and Pscf::DeviceArray< Data >::capacity().

addEqS() [6/6]

void Pscf::Prdc::Cuda::VecOp::addEqS ( DeviceArray< cudaComplex > & a, cudaReal const b )

inline

Vector addition in-place, a[i] += b, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 1076 of file VecOp.h.

References addEqS(), and Pscf::DeviceArray< Data >::capacity().

subEqV() [4/6]

void Pscf::Prdc::Cuda::VecOp::subEqV ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b )

inline

Vector subtraction in-place, a[i] -= b[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 1101 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subEqV().

subEqV() [5/6]

void Pscf::Prdc::Cuda::VecOp::subEqV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b )

inline

Vector subtraction in-place, a[i] -= b[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 1122 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subEqV().

subEqV() [6/6]

void Pscf::Prdc::Cuda::VecOp::subEqV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaReal > const & b )

inline

Vector subtraction in-place, a[i]-=b[i], kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 1144 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subEqV().

subEqS() [4/6]

void Pscf::Prdc::Cuda::VecOp::subEqS ( DeviceArray< cudaReal > & a, cudaReal const b )

inline

Vector subtraction in-place, a[i] -= b, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 1165 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subEqS().

subEqS() [5/6]

void Pscf::Prdc::Cuda::VecOp::subEqS ( DeviceArray< cudaComplex > & a, cudaComplex const b )

inline

Vector subtraction in-place, a[i] -= b, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 1185 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subEqS().

subEqS() [6/6]

void Pscf::Prdc::Cuda::VecOp::subEqS ( DeviceArray< cudaComplex > & a, cudaReal const b )

inline

Vector subtraction in-place, a[i] -= b, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 1205 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and subEqS().

mulEqV() [4/6]

void Pscf::Prdc::Cuda::VecOp::mulEqV ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b )

inline

Vector multiplication in-place, a[i] *= b[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 1230 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulEqV().

mulEqV() [5/6]

void Pscf::Prdc::Cuda::VecOp::mulEqV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaComplex > const & b )

inline

Vector multiplication in-place, a[i]*=b[i], kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 1252 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulEqV().

mulEqV() [6/6]

void Pscf::Prdc::Cuda::VecOp::mulEqV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaReal > const & b )

inline

Vector multiplication in-place, a[i]*=b[i], kernel wrapper (mixed, b=real).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 1274 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulEqV().

mulEqS() [4/6]

void Pscf::Prdc::Cuda::VecOp::mulEqS ( DeviceArray< cudaReal > & a, cudaReal const b )

inline

Vector multiplication in-place, a[i] *= b, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 1295 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulEqS().

mulEqS() [5/6]

void Pscf::Prdc::Cuda::VecOp::mulEqS ( DeviceArray< cudaComplex > & a, cudaComplex const b )

inline

Vector multiplication in-place, a[i] *= b, kernel wrapper (cudaComplex).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 1315 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulEqS().

mulEqS() [6/6]

void Pscf::Prdc::Cuda::VecOp::mulEqS ( DeviceArray< cudaComplex > & a, cudaReal const b )

inline

Vector multiplication in-place, a[i]*=b, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 1335 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and mulEqS().

divEqV() [3/4]

void Pscf::Prdc::Cuda::VecOp::divEqV ( DeviceArray< cudaReal > & a, DeviceArray< cudaReal > const & b )

inline

Vector division in-place, a[i] /= b[i], kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 1360 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and divEqV().

divEqV() [4/4]

void Pscf::Prdc::Cuda::VecOp::divEqV ( DeviceArray< cudaComplex > & a, DeviceArray< cudaReal > const & b )

inline

Vector division in-place, a[i] /= b[i], kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 1381 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and divEqV().

divEqS() [3/4]

void Pscf::Prdc::Cuda::VecOp::divEqS ( DeviceArray< cudaReal > & a, cudaReal const b )

inline

Vector division in-place, a[i] /= b, kernel wrapper (cudaReal).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 1402 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and divEqS().

divEqS() [4/4]

void Pscf::Prdc::Cuda::VecOp::divEqS ( DeviceArray< cudaComplex > & a, cudaReal const b )

inline

Vector division in-place, a[i] /= b, kernel wrapper (mixed, b = real).

Parameters

a	output array (LHS)
b	input scalar (RHS)

Definition at line 1422 of file VecOp.h.

References Pscf::DeviceArray< Data >::capacity(), and divEqS().

addVcVc()

void Pscf::Prdc::Cuda::VecOp::addVcVc	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		cudaReal const	c,
		DeviceArray< cudaReal > const &	d,
		cudaReal const	e )

Vector addition w/ coefficient, a[i] = (b[i]*c) + (d[i]*e), kernel wrapper.

Parameters

a	output array (LHS)
b	input array 1 (RHS)
c	input scalar (RHS)
d	input array 2 (RHS)
e	input scalar 2 (RHS)

Definition at line 304 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::EinsteinCrystalPerturbation< D >::incrementDc(), Pscf::Rpg::ForceBiasMove< D >::move(), Pscf::Rpg::BinaryStructureFactorGrid< D >::sample(), Pscf::Rpg::ExplicitBdStep< D >::step(), and Pscf::Rpg::PredCorrBdStep< D >::step().

addVcVcVc()

void Pscf::Prdc::Cuda::VecOp::addVcVcVc	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		cudaReal const	c,
		DeviceArray< cudaReal > const &	d,
		cudaReal const	e,
		DeviceArray< cudaReal > const &	f,
		cudaReal const	g )

3-vec addition w coeff, a[i] = (b[i]*c) + (d[i]*e) + (f[i]*g), kernel wrapper.

Parameters

a	output array (LHS)
b	input array 1 (RHS)
c	input scalar 1 (RHS)
d	input array 2 (RHS)
e	input scalar 2 (RHS)
f	input array 3 (RHS)
g	input scalar 3 (RHS)

Definition at line 322 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::LMBdStep< D >::step(), and Pscf::Rpg::PredCorrBdStep< D >::step().

addEqVc()

void Pscf::Prdc::Cuda::VecOp::addEqVc	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		cudaReal const	c )

Vector addition in-place w/ coefficient, a[i] += b[i] * c, kernel wrapper.

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar

Definition at line 343 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::RealMove< D >::attemptMove(), Pscf::Rpg::Simulator< D >::computeCc(), Pscf::Rpg::Simulator< D >::computeWc(), Pscf::Rpg::ForceBiasMove< D >::move(), Pscf::Rpg::ExplicitBdStep< D >::step(), Pscf::Rpg::LMBdStep< D >::step(), Pscf::Rpg::PredCorrBdStep< D >::step(), and Pscf::Rpg::MaskGenFilm< D >::stressTerm().

subVVS()

void Pscf::Prdc::Cuda::VecOp::subVVS	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		DeviceArray< cudaReal > const &	c,
		cudaReal const	d )

Vector subtraction, a[i] = b[i] - c[i] - d, kernel wrapper.

Parameters

a	output array (LHS)
b	input array (RHS)
c	input array (RHS)
d	input scalar (RHS)

Definition at line 359 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

divEqVc()

void Pscf::Prdc::Cuda::VecOp::divEqVc	(	DeviceArray< cudaComplex > &	a,
		DeviceArray< cudaReal > const &	b,
		cudaReal const	c )

Vector division in-place w/ coeff., a[i] /= (b[i] * c), kernel wrapper.

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar (RHS)

Definition at line 377 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

expVc()

void Pscf::Prdc::Cuda::VecOp::expVc	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaReal > const &	b,
		cudaReal const	c )

Vector exponentiation w/ coefficient, a[i] = exp(b[i]*c), kernel wrapper.

Parameters

a	output array (LHS)
b	input array (RHS)
c	input scalar

Definition at line 393 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::Solvent< D >::compute(), and Pscf::Rpg::Block< D >::setupSolver().

eqVPair()

void Pscf::Prdc::Cuda::VecOp::eqVPair	(	DeviceArray< cudaReal > &	a1,
		DeviceArray< cudaReal > &	a2,
		DeviceArray< cudaReal > const &	s )

Vector assignment in pairs, ax[i] = b[i], kernel wrapper.

Parameters

a1	output array 1 (LHS)
a2	output array 2 (LHS)
s	shared input array to be assigned to both a1 and a2

Definition at line 409 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::PredCorrBdStep< D >::step().

mulVVPair()

void Pscf::Prdc::Cuda::VecOp::mulVVPair	(	DeviceArray< cudaReal > &	a1,
		DeviceArray< cudaReal > &	a2,
		DeviceArray< cudaReal > const &	b1,
		DeviceArray< cudaReal > const &	b2,
		DeviceArray< cudaReal > const &	s )

Vector multiplication in pairs, ax[i] = bx[i] * s[i], kernel wrapper.

Parameters

a1	output array 1 (LHS)
a2	output array 2 (LHS)
b1	input array 1 (RHS)
b2	input array 2 (RHS)
s	shared input array to be multiplied by both b1 and b2

Definition at line 426 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::Block< D >::step().

mulEqVPair()

void Pscf::Prdc::Cuda::VecOp::mulEqVPair	(	DeviceArray< cudaReal > &	a1,
		DeviceArray< cudaReal > &	a2,
		DeviceArray< cudaReal > const &	s )

In-place vector multiplication in pairs, ax[i] *= s[i], kernel wrapper.

Parameters

a1	output array 1 (LHS)
a2	output array 2 (LHS)
s	shared input array to be multiplied by both a1 and a2

Definition at line 448 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::Block< D >::step().

addVMany() [1/2]

void Pscf::Prdc::Cuda::VecOp::addVMany	(	DeviceArray< cudaReal > &	a,
		DArray< DeviceArray< cudaReal > > const &	vecs )

Add an undefined number of vectors pointwise, kernel wrapper.

The input array 'vecs' contains the arrays that will be added together. The size of vecs determines the number of vectors that will ultimately be added together by the GPU kernel.

Parameters

a	output array (LHS)
vecs	array of DeviceArrays to be added

Definition at line 465 of file VecOpMisc.cu.

References addVV(), Pscf::DeviceArray< Data >::cArray(), Util::Array< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

addVMany() [2/2]

void Pscf::Prdc::Cuda::VecOp::addVMany	(	DeviceArray< cudaReal > &	a,
		DArray< DeviceArray< cudaReal > const * > const &	vecs )

Add an undefined number of vectors pointwise, kernel wrapper.

The input array 'vecs' contains const pointers to each array that will be added together. The size of vecs determines the number of vectors that will ultimately be added together by the GPU kernel.

This version of addVMany is provided for cases in which one needs to add many arrays that are not already stored together in a DArray. The caller must simply assemble an array of pointers to all of the arrays that should be added, and then pass it to this method.

Parameters

a	output array (LHS)
vecs	array of pointers to DeviceArrays to be added

Definition at line 495 of file VecOpMisc.cu.

References addVV(), Pscf::DeviceArray< Data >::cArray(), Util::Array< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

mulVMany() [1/2]

void Pscf::Prdc::Cuda::VecOp::mulVMany	(	DeviceArray< cudaReal > &	a,
		DArray< DeviceArray< cudaReal > > const &	vecs )

Multiply an undefined number of vectors pointwise, kernel wrapper.

The input array 'vecs' contains the arrays that will be multiplied. The size of vecs determines the number of vectors that will ultimately be multiplied together by the GPU kernel.

Parameters

a	output array (LHS)
vecs	array of DeviceArrays to be multiplied

Definition at line 525 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::cArray(), Util::Array< Data >::cArray(), mulVV(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::Propagator< D >::computeHead().

mulVMany() [2/2]

void Pscf::Prdc::Cuda::VecOp::mulVMany	(	DeviceArray< cudaReal > &	a,
		DArray< DeviceArray< cudaReal > const * > const &	vecs )

Multiply an undefined number of vectors pointwise, kernel wrapper.

The input array 'vecs' contains const pointers to each array that will be multiplied together. The size of vecs determines the number of vectors that will ultimately be multiplied together by the GPU kernel.

This version of mulVMany is provided for cases in which one needs to multiply many arrays that are not already stored together in a DArray. The caller must simply assemble an array of pointers to all of the arrays that should be multiplied, and then pass it to this method.

Parameters

a	output array (LHS)
vecs	array of pointers to DeviceArrays to be multiplied

Definition at line 555 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::cArray(), Util::Array< Data >::cArray(), mulVV(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

sqNormV()

void Pscf::Prdc::Cuda::VecOp::sqNormV	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaComplex > const &	b )

Squared norm of complex number, a[i] = norm(b[i])^2, kernel wrapper.

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 585 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::MaxOrderParameter< D >::compute().

sqSqNormV()

void Pscf::Prdc::Cuda::VecOp::sqSqNormV	(	DeviceArray< cudaReal > &	a,
		DeviceArray< cudaComplex > const &	b )

Norm of complex number to the 4th power, a[i] = norm(b[i])^4, kernel wrapper.

Parameters

a	output array (LHS)
b	input array (RHS)

Definition at line 600 of file VecOpMisc.cu.

References Pscf::DeviceArray< Data >::capacity(), Pscf::DeviceArray< Data >::cArray(), Pscf::ThreadArray::setThreadsLogical(), and UTIL_CHECK.

Referenced by Pscf::Rpg::FourthOrderParameter< D >::compute().