prep for release 0.5.0 (#206)

.github/workflows/docs.yml

@@ -16,7 +16,7 @@ jobs:
         with:
           version: '1.6'
       - name: Install dependencies
-        run: julia --project=docs/ -e 'using Pkg; Pkg.develop(PackageSpec(path=pwd())); Pkg.instantiate()'
+        run: julia --project=docs/ docs/install.jl
       - name: Build and deploy
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} # If authenticating with GitHub Actions token

NEWS.md

@@ -0,0 +1,50 @@
+# Release notes
+
+## v0.5.0 (August 30th, 2022)
+
+### Breaking changes
+
+- Change names to follow JSO conventions
+  - `AbstractInteriorPointSolver` -> `AbstractMadNLPSolver`
+  - `InteriorPointSolver` -> `MadNLPSolver`
+  - `IPMOptions` -> `MadNLPOptions`
+  - `Counters` -> `MadNLPCounters`
+  - `Logger` -> `MadNLPLogger`
+- Linear solvers are now defined as a struct, not a module
+  - `MadNLPLapackCPU.Solver` -> `LapackCPUSolver`
+  - `MadNLPUmfpack.Solver` -> `UmfpackSolver`
+  - `MadNLPLapackGPU.Solver` -> `LapackGPUSolver`
+  - `MadNLPMa27.Solver` -> `Ma27Solver`
+  - `MadNLPMa57.Solver` -> `Ma57Solver`
+  - `MadNLPMa77.Solver` -> `Ma77Solver`
+  - `MadNLPMa86.Solver` -> `Ma86Solver`
+  - `MadNLPMa97.Solver` -> `Ma97Solver`
+  - `MadNLPKrylov.Solver` -> `KrylovIterator`
+  - `MadNLPMumps.Solver` -> `MumpsSolver`
+  - `MadNLPPardiso.Solver` -> `PardisoSolver`
+  - `MadNLPPardisoMKL.Solver` -> `PardisoMKLSolver`
+- Refactor the way we pass options to MadNLP
+  - *Before:* `ips = MadNLP.InteriorPointSolver(nlp; option_dict=sparse_options)`
+  - *Now:* `ips = MadNLP.MadNLPSolver(nlp; sparse_options...)`
+
+### New features
+
+- Add support for `Float32`
+- Add support for NLSModels
+- Add a function `timing_madnlp` to decompose the time spent in the callbacks and in the linear solver at each iteration
+- Add a `CuMadNLPSolver` constructor to instantiate MadNLP on CUDA GPU
+
+### Bug fixes
+
+- Stop overwriting the number of threads used in BLAS
+- Use symmetric `mul!` when using `DENSE_CONDENSED_KKT_SYSTEM` on the GPU
+- Fix unscaling of the constraints during post-processing
+
+### Performance and maintenance
+
+- Add a toy nonlinear problem `HS15Model` in MadNLPTests
+- Improve the build of MadNLPHSL
+- Remove `StrideOneVector` alias in MadNLP
+- Update the documentation
+- Improve code coverage
+

Project.toml

@@ -2,7 +2,7 @@ name = "MadNLP"
 uuid = "2621e9c9-9eb4-46b1-8089-e8c72242dfb6"
 authors = ["Sungho Shin"]
 repo = "https://github.com/sshin23/MadNLP.jl.git"
-version = "0.4.2"
+version = "0.5.0"
 
 [deps]
 Libdl = "8f399da3-3557-5675-b5ff-fb832c97cbdb"
@@ -18,7 +18,7 @@ SuiteSparse = "4607b0f0-06f3-5cda-b6b1-a6196a1729e9"
 
 [compat]
 MINLPTests = "~0.5"
-MadNLPTests = "~0.2"
+MadNLPTests = "~0.3"
 MathOptInterface = "1"
 NLPModels = "~0.17.2, 0.18, 0.19"
 SolverCore = "~0.1,~0.2"

docs/Project.toml

@@ -1,6 +1,5 @@
 [deps]
 Documenter = "e30172f5-a6a5-5a46-863b-614d45cd2de4"
 MadNLP = "2621e9c9-9eb4-46b1-8089-e8c72242dfb6"
-MadNLPTests = "b52a2a03-04ab-4a5f-9698-6a2deff93217"
 NLPModels = "a4795742-8479-5a88-8948-cc11e1c8c1a6"
 SparseArrays = "2f01184e-e22b-5df5-ae63-d93ebab69eaf"

docs/install.jl

@@ -0,0 +1,8 @@
+using Pkg
+
+MADNLP_DIR = pwd()
+
+Pkg.develop(path=MADNLP_DIR)
+Pkg.develop(PackageSpec(path=joinpath(dirname(@__FILE__), "../lib/MadNLPTests/")))
+Pkg.instantiate()
+

docs/make.jl

@@ -1,10 +1,6 @@
 using Documenter
-using Pkg
 using MadNLP
 
-Pkg.develop(PackageSpec(path=joinpath(dirname(@__FILE__), "../lib/MadNLPTests/")))
-Pkg.instantiate()
-
 makedocs(
     sitename = "MadNLP.jl",
     format = Documenter.HTML(

lib/MadNLPGPU/Project.toml

@@ -1,7 +1,7 @@
 name = "MadNLPGPU"
 uuid = "d72a61cc-809d-412f-99be-fd81f4b8a598"
 authors = ["Sungho Shin <[email protected]>"]
-version = "0.3.0"
+version = "0.4.0"
 
 [deps]
 CUDA = "052768ef-5323-5732-b1bb-66c8b64840ba"
@@ -14,8 +14,8 @@ MadNLP = "2621e9c9-9eb4-46b1-8089-e8c72242dfb6"
 CUDA = "~3"
 CUDAKernels = "0.4"
 KernelAbstractions = "0.8"
-MadNLP = "~0.4.2"
-MadNLPTests = "~0.2"
+MadNLP = "~0.5"
+MadNLPTests = "~0.3"
 julia = "1.7"
 
 [extras]

lib/MadNLPHSL/Project.toml

@@ -1,7 +1,7 @@
 name = "MadNLPHSL"
 uuid = "7fb6135f-58fe-4112-84ca-653cf5be0c77"
 authors = ["Sungho Shin <[email protected]>"]
-version = "0.2.0"
+version = "0.3.0"
 
 [deps]
 MadNLP = "2621e9c9-9eb4-46b1-8089-e8c72242dfb6"
@@ -10,8 +10,8 @@ BinaryProvider = "b99e7846-7c00-51b0-8f62-c81ae34c0232"
 Libdl = "8f399da3-3557-5675-b5ff-fb832c97cbdb"
 
 [compat]
-MadNLP = "~0.4"
-MadNLPTests = "~0.2"
+MadNLP = "~0.5"
+MadNLPTests = "~0.3"
 BinaryProvider = "0.5"
 julia = "1.6"
 

lib/MadNLPKrylov/Project.toml

@@ -1,17 +1,17 @@
 name = "MadNLPKrylov"
 uuid = "1888cb03-ce40-4a36-8d45-ae8231a0e17c"
 authors = ["Sungho Shin <[email protected]>"]
-version = "0.2.0"
+version = "0.3.0"
 
 [deps]
 IterativeSolvers = "42fd0dbc-a981-5370-80f2-aaf504508153"
 MadNLP = "2621e9c9-9eb4-46b1-8089-e8c72242dfb6"
 
 [compat]
 IterativeSolvers = "~0.9"
-MadNLP = "~0.4"
+MadNLP = "~0.5"
 julia = "1.6"
-MadNLPTests = "~0.2"
+MadNLPTests = "~0.3"
 
 [extras]
 MadNLPTests = "b52a2a03-04ab-4a5f-9698-6a2deff93217"

lib/MadNLPMumps/Project.toml

@@ -1,7 +1,7 @@
 name = "MadNLPMumps"
 uuid = "3b83494e-c0a4-4895-918b-9157a7a085a1"
 authors = ["Sungho Shin <[email protected]>"]
-version = "0.2.0"
+version = "0.3.0"
 
 [deps]
 MUMPS_seq_jll = "d7ed1dd3-d0ae-5e8e-bfb4-87a502085b8d"
@@ -10,8 +10,8 @@ StaticArrays = "90137ffa-7385-5640-81b9-e52037218182"
 
 [compat]
 MUMPS_seq_jll = "~5.2.1, ~5.3.5"
-MadNLP = "~0.4"
-MadNLPTests = "~0.2"
+MadNLP = "~0.5"
+MadNLPTests = "~0.3"
 StaticArrays = "1"
 julia = "1.6"
 

lib/MadNLPPardiso/Project.toml

@@ -1,7 +1,7 @@
 name = "MadNLPPardiso"
 uuid = "312ee924-cb12-49df-b284-7304c3902fc0"
 authors = ["Sungho Shin <[email protected]>"]
-version = "0.2.0"
+version = "0.3.0"
 
 [deps]
 MadNLP = "2621e9c9-9eb4-46b1-8089-e8c72242dfb6"
@@ -11,9 +11,9 @@ Libdl = "8f399da3-3557-5675-b5ff-fb832c97cbdb"
 MKL_jll = "856f044c-d86e-5d09-b602-aeab76dc8ba7"
 
 [compat]
-MadNLP = "~0.4.1"
+MadNLP = "~0.5"
 BinaryProvider = "0.5"
-MadNLPTests = "~0.2"
+MadNLPTests = "~0.3"
 julia = "1.6"
 MKL_jll = "~2021,2022"
 

lib/MadNLPTests/Project.toml

@@ -1,7 +1,7 @@
 name = "MadNLPTests"
 uuid = "b52a2a03-04ab-4a5f-9698-6a2deff93217"
 authors = ["Sungho Shin <[email protected]>"]
-version = "0.2.1"
+version = "0.3.0"
 
 [deps]
 ADNLPModels = "54578032-b7ea-4c30-94aa-7cbd1cce6c9a"
@@ -15,7 +15,7 @@ Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
 
 [compat]
 ADNLPModels = "~0.3, 0.4"
-JuMP = "~0.23,~1"
-MadNLP = "~0.4"
+JuMP = "1"
+MadNLP = "~0.5"
 NLPModels = "~0.17.2, 0.18, 0.19"
 julia = "1.6"