Home
last modified time | relevance | path

Searched refs:Phi (Results 1 – 4 of 4) sorted by relevance

/petsc/src/snes/impls/vi/ss/
H A Dviss.c26 static inline PetscScalar Phi(PetscScalar a, PetscScalar b) in Phi() function
74 phi_arr[i] = -Phi(u[i] - x_arr[i], -f_arr[i]); in SNESVIComputeFunction()
76 phi_arr[i] = Phi(x_arr[i] - l[i], f_arr[i]); in SNESVIComputeFunction()
80 phi_arr[i] = Phi(x_arr[i] - l[i], -Phi(u[i] - x_arr[i], -f_arr[i])); in SNESVIComputeFunction()
124 da1 = DPhi(x[i] - l[i], -Phi(u[i] - x[i], -f[i])); in SNESVIComputeBsubdifferentialVectors()
125 db1 = DPhi(-Phi(u[i] - x[i], -f[i]), x[i] - l[i]); in SNESVIComputeBsubdifferentialVectors()
/petsc/src/ksp/ksp/utils/lmvm/symbrdn/
H A Dsymbrdn.c61 …tscErrorCode MatLMVMSymBroydenGetConvexFactor(Mat B, SymBroydenProductsType Phi_t, LMProducts *Phi) in MatLMVMSymBroydenGetConvexFactor() argument
69 *Phi = lsb->products[Phi_t]; in MatLMVMSymBroydenGetConvexFactor()
74 start = PetscMax((*Phi)->k, oldest); in MatLMVMSymBroydenGetConvexFactor()
75 (*Phi)->k = start; in MatLMVMSymBroydenGetConvexFactor()
76 …for (PetscInt i = start; i < next; i++) PetscCall(LMProductsInsertNextDiagonalValue(*Phi, i, phi)); in MatLMVMSymBroydenGetConvexFactor()
96 LMProducts Phi, Psi = NULL; in SymBroydenRecursiveBasisUpdate() local
113 PetscCall(MatLMVMSymBroydenGetConvexFactor(B, Phi_t, &Phi)); in SymBroydenRecursiveBasisUpdate()
127 Phi->k = start; in SymBroydenRecursiveBasisUpdate()
165 PetscCall(LMProductsInsertNextDiagonalValue(Phi, j, phi_j)); in SymBroydenRecursiveBasisUpdate()
166 } else PetscCall(LMProductsGetDiagonalValue(Phi, j, &phi_j)); in SymBroydenRecursiveBasisUpdate()
[all …]
/petsc/doc/manual/
H A Dtao.md2703 of equations $\Phi(x) = 0$, where
2704 $\Phi:\mathbb R^n \to \mathbb R^n$. The reformulation is defined
2718 We note that $\Phi$ is not differentiable everywhere but satisfies
2722 $\Psi(x) := \frac{1}{2} \| \Phi(x) \|_2^2$, is continuously
2725 The two semismooth TAO solvers both solve the system $\Phi(x) = 0$
2728 $H^kd^k = -\Phi(x^k)$, where $H^k$ is an element of the
2730 $\Phi$ at $x^k$. If the direction calculated does not
H A Dperformance.md331 “Knights Landing” Intel Xeon Phi, for instance, L2 caches are shared