Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

docs improvements/typos (MATLAB all kernels; Fortran just Stokes) #37

Merged
merged 15 commits into from
Aug 8, 2023
Merged

docs improvements/typos (MATLAB all kernels; Fortran just Stokes) #37

Show file tree
Hide file tree
Changes from 12 commits
Commits
Show all changes
15 commits
Select commit Hold shift + click to select a range
6c1b443
removed obsolete srcinfo.dipoles from stokes matlab example
ahbarnett Aug 4, 2023
ec11126
stfmm3d_simpletest.m
ahbarnett Aug 4, 2023
b426f43
clarify stfmm3d.f docs and fix typos in kernel defs
ahbarnett Aug 4, 2023
5408c55
tidy up Stokes docs in Fortran and MATLAB
ahbarnett Aug 7, 2023
6410b79
make stokes and em docs in matlab more matlab-like and clear
ahbarnett Aug 7, 2023
440ea38
lap and helm matlab docs more matlab-like, tidy up
ahbarnett Aug 7, 2023
4a8a03d
add Contents.m for MATLAB/Octave docs
ahbarnett Aug 7, 2023
8ba2bd1
Merge branch 'flatironinstitute:master' into master
ahbarnett Aug 7, 2023
1aebdcb
make h3ddir and l3ddir consistent with st3ddir, cut down repeated
askhamwhat Aug 8, 2023
d84c574
add some clarification about evaluation points for sums
askhamwhat Aug 8, 2023
d81dfaa
small clean
askhamwhat Aug 8, 2023
281893d
add math formulae for emfmm3d
askhamwhat Aug 8, 2023
fcebcbd
updating test file names to be consistent and fixing exit in legacy t…
mrachh Aug 8, 2023
9eb1ff4
updating contents
mrachh Aug 8, 2023
eef9634
updating docs via mw
mrachh Aug 8, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
35 changes: 35 additions & 0 deletions matlab/Contents.m
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,35 @@
% FMM3D: MATLAB/Octave wrappers to 3D fast multipole methods.
% Center for Computational Mathematics, Flatiron Institute.
%
% Functions:
% lfmm3d - FMM in 3D for Laplace (electrostatic) kernels.
% l3ddir - Direct (slow) 3D Laplace kernel sums (reference for LFMM3D).
% hfmm3d - FMM in 3D for Helmholtz (acoustic frequency-domain) kernel.
% h3ddir - Direct (slow) 3D Helmholtz kernel sums (reference for HFMM3D).
% emfmm3d - FMM in 3D for Maxwell (frequency-domain electromagnetic) kernels.
% em3ddir - Direct (slow) 3D Maxwell kernel sums (reference for EMFMM3D).
% stfmm3d - FMM in 3D for Stokes (viscous fluid hydrodynamic) kernels.
% st3ddir - Direct (slow) 3D Stokes kernel sums (reference for STFMM3D).
%
% For tester driver scripts see:
% test_lfmm3d
% test_hfmm3d
% test_emfmm3d
% test_stfmm3d
%
% For examples of use see:
% lfmm3d_example
% lfmm3d_big_example
% hfmm3d_example
% hfmm3d_big_example
% emfmm3d_example
% stfmm3d_example
% stfmm3d_simpletest
%
% Legacy codes/interfaces (from the Gimbutas-Greengard CMCL 2012 library):
% lfmm3dpart - Laplace particle targ FMM in R^3.
% l3dpartdirect - Laplace interactions in R^3, direct (slow) evaluation.
% hfmm3dpart - Helmholtz particle targ FMM in R^3.
% h3dpartdirect - Helmholtz interactions in R^3, direct (slow) evaluation.
% test_lfmm3dpart_direct - Test Laplace particle FMMs in R^3
% test_hfmm3dpart_direct - Test Helmholtz particle FMMs in R^3
54 changes: 10 additions & 44 deletions matlab/em3ddir.m
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,51 +1,17 @@
function [U] = em3ddir(zk,srcinfo,targ,ifE,ifcurlE,ifdivE)
% EM3DDIR Slow direct Maxwell kernel sums (reference for EMFMM3D).
%
% U = em3ddir(zk,srcinfo,targ,ifE,ifcurlE,ifdivE)
%
% This subroutine computes
% E = curl S_{k}[h_current] + S_{k}[e_current] + grad S_{k}[e_charge] -- (1)
% using the vector Helmholtz fmm.
% The subroutine also computes divE, curlE
% with appropriate flags
% Remark: the subroutine uses a stabilized representation
% for computing the divergence by using integration by parts
% wherever possible. If the divergence is not requested, then the
% helmholtz fmm is called with 3*nd densities, while if the divergence
% is requested, then the helmholtz fmm is calld with 4*nd densities
%
% Args:
% Maxwell direct evaluation in R^3: evaluate all pairwise particle
% interactions with targs. This is the slow O(N^2) direct code used
% as a reference for testing the fast code emfmm3d.
%
% - zk: complex
% Helmholtz parameter, k
% - srcinfo: structure
% structure containing sourceinfo
%
% * srcinfo.sources: double(3,n)
% source locations, $x_{j}$
% * srcinfo.nd: integer
% number of charge/dipole vectors (optional,
% default - nd = 1)
% * srcinfo.h_current: complex(nd,3,n)
% a vector source (optional,
% default - term corresponding to h_current dropped)
% * srcinfo.e_current: complex(nd,3,n)
% b vector source (optional,
% default - term corresponding to e_current dropped)
% * srcinfo.e_charge: complex(nd,n)
% e_charge source (optional,
% default - term corresponding to e_charge dropped)
% - targ: double(3,nt)
% target locations, $t_{i}$
% - ifE: integer
% E is returned at the target locations if ifE = 1
% - ifcurlE: integer
% curl E is returned at the target locations if ifcurlE = 1
% - ifdivE: integer
% div E is returned at the target locations if ifdivE = 1
% Returns:
%
% - U.E: E field defined in (1) above at target locations if requested
% - U.curlE: curl of E field at target locations if requested
% - U.divE: divergence of E at target locations if requested
% Kernel definitions, input and outputs arguments are identical to
% emfmm3d (see that function for all definitions), except that the first
% argument (eps) is absent.
%
% See also: EMFMM3D

if(nargin<4)
return;
Expand Down
47 changes: 36 additions & 11 deletions matlab/emfmm3d.m
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,16 +1,39 @@
function [U] = emfmm3d(eps,zk,srcinfo,targ,ifE,ifcurlE,ifdivE)
% EMFMM3D FMM in 3D for Maxwell (frequency-domain electromagnetic) kernels.
%
% U = emfmm3d(eps,zk,srcinfo,targ,ifE,ifcurlE,ifdivE)
%
% Frequency-domain Maxwell FMM in R^3: evaluate all pair-wise
% particle interactions with targets using the fast multipole
% method with precision eps.
%
% Specifically, this subroutine computes a sum for the electric field
%
% E(x) = sum_m curl G_k(x,y^{(m)}) h_current_m
% + G_k(x,y^{(m)}) e_current_m
% + grad G_k(x,y^{(m)}) e_charge_m
%
% for each requested evaluation point x, where h_current and e_current
% are 3-vector densities and e_charge is a scalar density supplied
% at each source point y^{(m)}. G_k is the Helmholtz Green function
% without the 1/(4pi) scaling:
%
% G_k(x,y) = e^(ik|x-y|)/|x-y|.
%
% In contrast with other FMM routines in the library, this routine
% has only 1 option for the evaluation points: they are specified
% as targets. If a target x coincides with a source point y^{(m)}
% that term in the sum is omitted.
%
% The electric field is, naturally, a 3-vector at each point x.
% With appropriate input flags, the subroutine also computes divE,
% curlE.
%
% This subroutine computes
% E = curl S_{k}[h_current] + S_{k}[e_current] + grad S_{k}[e_charge] -- (1)
% using the vector Helmholtz fmm.
% The subroutine also computes divE, curlE
% with appropriate flags
% Remark: the subroutine uses a stabilized representation
% for computing the divergence by using integration by parts
% wherever possible. If the divergence is not requested, then the
% helmholtz fmm is called with 3*nd densities, while if the divergence
% is requested, then the helmholtz fmm is calld with 4*nd densities
% Helmholtz FMM is called with 3*nd densities, while if the divergence
% is requested, then the Helmholtz FMM is called with 4*nd densities
%
% Args:
%
Expand All @@ -19,8 +42,7 @@
% - zk: complex
% Helmholtz parameter, k
% - srcinfo: structure
% structure containing sourceinfo
%
% structure containing the following info about the sources:
% * srcinfo.sources: double(3,n)
% source locations, $x_{j}$
% * srcinfo.nd: integer
Expand All @@ -43,11 +65,14 @@
% curl E is returned at the target locations if ifcurlE = 1
% - ifdivE: integer
% div E is returned at the target locations if ifdivE = 1
%
% Returns:
%
% - U.E: E field defined in (1) above at target locations if requested
% - U.E: E field defined in above equation at targets if requested
% - U.curlE: curl of E field at target locations if requested
% - U.divE: divergence of E at target locations if requested
% - U.divE: divergence of E at target locations if requested
%
% See also: HFMM3D, EM3DDIR

if(nargin<5)
return;
Expand Down
Loading