CPU Impl of AssemblePointBlockDiagonal (#503)

* Operator - add AssemblePointBlockDiagonal for CPU backends, with test

* CUDA - add point block diagonal not supported message

* make style

*

CPU Impl of AssemblePointBlockDiagonal (#503)

* Operator - add AssemblePointBlockDiagonal for CPU backends, with test

* CUDA - add point block diagonal not supported message

* make style

* Operator - improve point block description

* OCCA - explicitly remove OCCA fallback to CPU assembly functions, will update after new OCCA backend

* Op - remove gap removal in point block diagonal

* Op - update diagonal assembly documentation

* Update backends/ref/ceed-ref-operator.c

Co-authored-by: Jed Brown <jed@jedbrown.org>

* style - fix extra space in * with nopad

Co-authored-by: Jed Brown <jed@jedbrown.org>

show more ...

CEED_STRIDES_BACKEND optimization for cuda-ref operator apply (#528)

* add check for backend stride status for input vectors

* add backend strides check for output vectors

* replace output cop

CEED_STRIDES_BACKEND optimization for cuda-ref operator apply (#528)

* add check for backend stride status for input vectors

* add backend strides check for output vectors

* replace output copy with elem restriction for none emode

* move input skip_restrict check to setup and never allocate E-vec if not needed

* add boolean variable for E/Q vector layout for
further optimization of output and add wrapper function in magma backend
to create a cuda-ref operator and change this state variable

* Add missing CeedChks

* style changes to better match cuda backends

* missed style change for evec check

* add CeedChk from PR #525 (merge conflict)

* make style changes

* adjust size of nqpts for non-tensor basis

show more ...

CUDA - support empty restrictions

CUDA/MAGMA - add tidy fixes

GPU - add ElemRestrictionGetOffsets

More comments in CUDA backends (#518)

* CUDA - adding comments as I work to understand these backends

* PETSc - remove extra include, breaks single source

* make style

CUDA - use new component stride over interlace mode

Solid Mechanics Example (#474)

libCEED + PETSc Solid Mechanics example

Co-authored-by: Jeremy L Thompson <jeremy.thompson@colorado.edu>
Co-authored-by: Arash Mehraban <arashm81@gmail.com>
Co-au

Solid Mechanics Example (#474)

libCEED + PETSc Solid Mechanics example

Co-authored-by: Jeremy L Thompson <jeremy.thompson@colorado.edu>
Co-authored-by: Arash Mehraban <arashm81@gmail.com>
Co-authored-by: Jed Brown <jed@jedbrown.org>
Co-authored-by: Valeria Barra <valeriabarra21@gmail.com>


* Solids - initial commit of elasticity example

Co-authored-by: Jeremy L Thompson <jeremy.thompson@colorado.edu>
Co-authored-by: Arash Mehraban <arashm81@gmail.com>

* Solids - add VLA macro

* Solids - fix test for mesh file output string

* Solids - update documentation

Co-authored-by: Jeremy L Thompson <jeremy.thompson@colorado.edu>
Co-authored-by: Arash Mehraban <arashm81@gmail.com>

* Solids - update README

* Solids - update ctx struct and casting for CUDA

* Tests - clean up test conditions for GPU backends

* GPU - fix falbacks identified

* CUDA - whitelist FDM Inverse

* make style

* doc: solids editing/correctness

* Solids - separate FE boilerplate into its own function

* Solids - clean up double includes

* doc: add more complete explanation of hyperelasticity and formulation

* Solids - Use ISGetSize for global number of face sets (had no effect on PETSc bug in MR 2615)

* edited docs for hyperFS

* Improve doc of solids example with some constitent notation and fonts

* HyperSS docs modified

* edited typo

* linearization added

* fixed a typo

* Improved doc

* doc: more writing on hyperelasticity and implementation tradeoffs

* Solids - fix typo

* Solids - compute and store C, Cinv, and S as symmetric

* Solids - use new Newton linearization, but its not converging

* Solids - update BC handling

* Solids - update README with new BC options

* removed old notation

* Solids - whitespace changes only in documentation

* doc: adjust differential notation and derive dP(dF)

* Solids - small documentation flow edits

* Solids - simplify setting BCs

* Solids - Symmetritize e in SS

* Solids - set more reasonable default number of steps for nonlinear

* Solids - fix documentation typo

* Solids - clean up computation of deltaS

* Solids - fix sample runs

* Solids - fix hyperSS jacobian

* Solids - fix hyperSS documentation

* Solids - clean up code for hyperSS

* Solids - add file and function docstrings

* doc: commutative diagram of linearization and stable stress expression

* Solids - update hyperSS documentation with linearization

* doc: further tips on stable computation

* Solids - fix typo with trace in hyperSS

* Solids - intro and hyperSS documentation updates

* Solids - update hyperFS for numerical stability with small E and nearly incompressible

* examples/solids: simpler C = I + 2E

* Solids - more compact/readable E computation

* Solids - minor style consistency change

* Solids - replace log1p with series approximation, log1p is not vectorized in libc

* Solids - refactor common components between hyperFSF and hyperFSdF

* examples/solids: use PETSc dependency flags for incremental rebuild

* examples/solids: extend range of log1p_series via shift/scale

* Solids - add bc_clamp_max option to specify maximum displacement for clamped boundary

* make style

* Solids - update with sample run informantion and spellchecking

* Doc - typo fixes

* Solids - homogonize presentation of command line options

* examples/solids: fix -Wmisleading-indentation

* examples/fluids: PETSc 3.13 compatibility

* examples/solids: use log1p for numerical stability in hyperSS

* doc: copy editing in examples/solids

* Solids - update documentation on running the code

* Solids - convert README to rst and include in documentation

* Solids - clean dependency file

* Solids - use log1p approximation in hyperSS as well

* Solids - update README to explain mms

* Solids - switch range on hyperSS log1p series approximation

* Solids - update README to clarify mandatory options

* Solids - add link to sample meshes

* Solids - simplify listing of mesh options

* Travis - use newer PETSc

* Solids - update sample run to list [.exo file] instead of name of sample mesh

* Solids - update make clean to remove src/*.d

* examples/solids: makefile cleanup

* make style

* doc examples/solids: clean up and explain nondimensionalization

* examples/solids: document algebraic solver

* examples/solids: use CG for Chebyshev eig estimation

* doc: fix references to examples/fluids

* Solids - remove mesh folder

* Solids - drop expicit ExodusII requirement, PETSc supports many types of meshes

* doc: mention PETSc --download-exodusii for Exodus

* doc: fix latexpdf

* Fix \longrightarrow rendering in PDF

* Solids - fix toctree link to README for pdflatex

* Examples and BPs Docs: Homogenize math with dS and dV and other tiny esthetics

* Solids - fix unshifted series for hyperSS log1p

* Improve doc of bpssphere: Add differential for consistency with all other examples and explain domain for clarity

* Solids - include sample mesh image

* Move mesh figure after note and slightly rename inclusion marker

Co-authored-by: Arash Mehraban <arashm81@gmail.com>
Co-authored-by: Jed Brown <jed@jedbrown.org>
Co-authored-by: valeriabarra <valeriabarra21@gmail.com>

show more ...

Add VLA support for CUDA backends (#481)

* CUDA - add VLA support for CUDA backends

* Navier-Stokes - guard #include math.h for CUDA

* Navier-Stokes - fix struct definitions for code generatio

Add VLA support for CUDA backends (#481)

* CUDA - add VLA support for CUDA backends

* Navier-Stokes - guard #include math.h for CUDA

* Navier-Stokes - fix struct definitions for code generation

* Navier-Stokes - fix file location of mass QFunction

* Navier-Stokes - minor style consistency

* Navier-Stokes - more context sturct fixing

* Navier-Stokes - cast context to stop CUDA errors

* Navier-Stokes - change helper functions to inline to silence CUDA errors

* CUDA - clean up CEED_Q_VLA definition for cuda/gen

* Navier-Stokes - populate TODOs for undocumenced functions

* Doc - document libCEED macros

* CUDA - use -default-device for NVRTC

* Tests - add t507 to test VLA support

* CUDA - fix my bug in dim, Q1d for cuda/gen

* Tests - update tap/junit to reflect current CUDA capabilities

* make style

* CUDA - check output bases for dim, Q1d as well (this fixes multigrid example support)

* CUDA - fix readSliceQuads, not all arguments were defined

* Navier-Stokes - document all functions in Navier-Stokes QFunction files

* make style

show more ...

Merge pull request #421 from SanderA/sanderarens/fix_ceed_cuda_subclasses

Add Ceed_Cuda struct to Ceed_Cuda_ref/shared/gen.

CUDA - initial impl of strided restrictions in cuda/gen

CUDA - update ref/reg restrictions to use CEED_STRIDES_BACKEND {1, nelem*elemsize, elemsize}

CUDA - add strided restrictions

ElemRestriction - make lmode a separate enum

ElemRestriction - remove lmode from backend apply signature

ElemRestriction - move lmode to constructor over apply

Style - rename files vec->vector for consistency

CUDA - fix up composite operator delegation

CUDA - drop QFAssemble delegation

Make cuda backends implement OperatorApplyAdd instead of OperatorApply.

Delegate AssembleLinearQF to ref/serial when not impl (#406)

* Operator - delegate AssembleLinearQF to ref/serial when not impl by backend

* Occa - Fix restriction summing error

* Tests - fix

Delegate AssembleLinearQF to ref/serial when not impl (#406)

* Operator - delegate AssembleLinearQF to ref/serial when not impl by backend

* Occa - Fix restriction summing error

* Tests - fix error in t534-f qfunction for CPU

* make style

* Operator - clarify fallback mechanism, allow backends to provide fallback other than /cpu/self/ref/serial

* Operator - update fallback to avoid copying vectors, restrictions

* Operator - move fallback to ceed level

* Operator - explicitly check for falling back to onself

* Update interface/ceed-operator.c

Co-Authored-By: Jed Brown <jed@jedbrown.org>

show more ...

Add Ceed_Cuda struct to Ceed_Cuda_ref/shared/gen.

Now Ceed_Cuda_ref/shared/gen act like subclasses and can be properly cast to Ceed_Cuda.

Basis - Use CEED_VECTOR_NONE for EVAL_MODE_WEIGHT

Style - Fix indentation errors

Make style