Merge branch 'master' of github.com:JuliaLang/julia

Conflicts:
	.gitmodules
	base/multi.jl
	base/printf.jl
	base/show.jl
	base/start_image.jl
	src/julia.expmap
	src/support/Makefile

.gitmodules

@@ -1,6 +1,6 @@
 [submodule "deps/libuv"]
  path = deps/libuv
- url = http:https://github.com/JuliaLang/libuv.git
+ url = git:https://github.com/JuliaLang/libuv.git
 [submodule "deps/openlibm"]
  path = deps/openlibm
  url = https://github.com/JuliaLang/openlibm.git

LICENSE.md

@@ -1,7 +1,7 @@
 Julia is licensed under the MIT License:
 
 > Copyright (c) 2009-2012: Jeff Bezanson, Stefan Karpinski, Viral B. Shah, 
-> Stephan Boyer, George Xing, and other contributors:
+> and other contributors:
 > 
 > https://github.com/JuliaLang/julia/contributors
 > 

Make.inc

@@ -73,6 +73,7 @@ JCFLAGS += '-DJL_SYSTEM_IMAGE_PATH="../lib/julia/sys.ji"'
 # OPENBLAS build options
 OPENBLAS_DYNAMIC_ARCH=0
 OPENBLAS_USE_THREAD=1
+OPENBLAS_TARGET_ARCH=
 
 # Use libraries available on the system instead of building them
 

README.md

@@ -107,20 +107,20 @@ To rebuild a pre-built Julia source install with MKL support, delete from `deps/
 Building Julia requires that the following software be installed:
 
 - **[GNU make]** — building dependencies.
-- **[gcc, g++][gcc]**  — compiling and linking C, C++ (Need at least v4.6)
-- **[clang][clang]** - clang can be used instead of gcc (Need at least v3.1, Xcode 4.3.3 on OS X)
-- **[gfortran][gcc]** - compiling and linking fortran libraries
-- **[git]**  — contributions and version control.
+- **[gcc, g++][gcc]**  — compiling and linking C, C++ (Need at least v4.6)
+- **[clang][clang]** — clang can be used instead of gcc (Need at least v3.1, Xcode 4.3.3 on OS X)
+- **[gfortran][gcc]** — compiling and linking fortran libraries
+- **[git]**  — contributions and version control.
 - **[perl]** — preprocessing of header files of libraries.
-- **[wget]**, **[curl]**, or **fetch** — to automatically download external libraries.
+- **[wget]**, **[curl]**, or **fetch** (FreeBSD) — to automatically download external libraries.
 - **[m4]** — needed to build GMP.
-- **patch**  — for modifying source code.
+- **[patch]** — for modifying source code.
 
 Julia uses the following external libraries, which are automatically downloaded (or in a few cases, included in the Julia source repository) and then compiled from source the first time you run `make`:
 
 - **[LLVM]** — compiler infrastructure. Currently, julia requires LLVM 3.0.
 - **[FemtoLisp]** — packaged with julia source, and used to implement the compiler front-end.
-- **[GNU readline]** — library allowing shell-like line editing in the terminal, with history and familiar key bindings.
+- **[readline]**  — library allowing shell-like line editing in the terminal, with history and familiar key bindings.
 - **[fdlibm]** — a portable implementation of much of the system-dependent libm math library's functionality.
 - **[MT]** — a fast Mersenne Twister pseudorandom number generator library.
 - **[OpenBLAS]** — a fast, open, and maintained [basic linear algebra subprograms (BLAS)](https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms) library, based on [Kazushige Goto's](https://en.wikipedia.org/wiki/Kazushige_Goto) famous [GotoBLAS](https://www.tacc.utexas.edu/tacc-projects/gotoblas2/).
@@ -134,18 +134,19 @@ Julia uses the following external libraries, which are automatically downloaded
 - **[GMP]** — the GNU multiple precision arithmetic library, needed for bigint support.
 - **[D3]** — JavaScript visualization library.
 - **[double-conversion]** — efficient number-to-text conversion.
-- **[GLPK]**  - linear programming.
-- **[Rmath]**  - basic RNGs and distributions.
+- **[GLPK]**  — linear programming.
+- **[Rmath]**  — basic RNGs and distributions.
 
 
 [GNU make]: https://www.gnu.org/software/make/
-[gcc]: https://gcc.gnu.org/
-[clang]: https://clang.llvm.org/
+[patch]: https://www.gnu.org/software/patch/
 [wget]: https://www.gnu.org/software/wget/
+[m4]: https://www.gnu.org/software/m4/
+[gcc]: https://gcc.gnu.org/
+[clang]: https://clang.llvm.org/
 [curl]: https://curl.haxx.se/
 [git]: https://git-scm.com/
 [perl]: https://www.perl.org/
-[m4]: https://www.gnu.org/software/m4/
 [fdlibm]: https://www.netlib.org/fdlibm/readme
 [MT]: https://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/emt.html
 [OpenBLAS]: https://github.com/xianyi/OpenBLAS#readme
@@ -158,12 +159,16 @@ Julia uses the following external libraries, which are automatically downloaded
 [PCRE]: https://www.pcre.org/
 [LLVM]: https://www.llvm.org/
 [FemtoLisp]: https://github.com/JeffBezanson/femtolisp
-[GNU readline]: https://cnswww.cns.cwru.edu/php/chet/readline/rltop.html
+[readline]:  https://cnswww.cns.cwru.edu/php/chet/readline/rltop.html
 [GMP]: https://gmplib.org/
 [D3]: https://mbostock.github.com/d3/
 [double-conversion]: https://double-conversion.googlecode.com/
-[GLPK]: https://www.gnu.org/software/glpk/
-[Rmath]: https://cran.r-project.org/doc/manuals/R-admin.html#The-standalone-Rmath-library
+[GLPK]: https://www.gnu.org/software/glpk/
+[Rmath]: https://cran.r-project.org/doc/manuals/R-admin.html#The-standalone-Rmath-library
+
+If you already have one or more of these packages installed on your system, it is possible to pass `USE_SYSTEM_...=1` to `make` to prevent Julia from compiling duplicates of these libraries. The complete list of possible flags can be found in Make.inc (or pass `USE_DEBIAN=1` to make if you have all build dependencies and want the minimal Julia build). Please be aware that this proceedure is not officially supported, as it introduces additional variablity into the installation and versioning of the dependencies, and is recommended only for system package maintainers. Unexpected compile errors may result, as the build system will do no further checking to ensure the proper packages are installed.
+
+SuiteSparse is a special case, since it is typically only installed as a static library, while `USE_SYSTEM_SUITESPARSE=1` requires that it is a shared library. Running the script `contrib/repackage_system_suitesparse4.make` will copy your static system SuiteSparse installation into the shared library format required by Julia.
 
 <a name="Directories"/>
 ## Directories
@@ -229,4 +234,5 @@ Julia has a web REPL with very preliminary graphics capabilities. The web REPL i
 
 ### Try it Online
 
-Forio.com is generously hosting and maintaining an instance of Julia's web REPL here: [julia.forio.com](https://julia.forio.com)
+[Forio.com](https://forio.com/) is generously hosting and maintaining an instance of Julia's web REPL here: [julia.forio.com](https://julia.forio.com).
+This service is best-effort and may not always be up or stable. Be nice!

base/client.jl

@@ -195,7 +195,7 @@ function process_options(args::Array{Any,1})
  eval(parse_input_line(args[i]))
  elseif args[i]=="-L"
  i+=1
- include(args[i])
+ load(args[i])
  elseif args[i]=="-p"
  i+=1
  np = int32(args[i])
@@ -214,7 +214,7 @@ function process_options(args::Array{Any,1})
  repl = false
  # remove julia's arguments
  ARGS = args[i+1:end]
- include(args[i])
+ load(args[i])
  break
  else
  error("unknown option: ", args[i])

base/factorizations.jl

@@ -1,9 +1,9 @@
 ## solvers and inverses using factorizations
-for (getrs, potrs, trtrs, getri, potri, trtri, elty) in
- (("dgetrs_","dpotrs_","dtrtrs_","dgetri_","dpotri_","dtrtri_",:Float64),
- ("sgetrs_","spotrs_","strtrs_","sgetri_","spotri_","strtri_",:Float32),
- ("zgetrs_","zpotrs_","ztrtrs_","zgetri_","zpotri_","ztrtri_",:Complex128),
- ("cgetrs_","cpotrs_","ctrtrs_","cgetri_","cpotri_","ctrtri_",:Complex64))
+for (getrs, potrs, getri, potri, trtri, elty) in
+ (("dgetrs_","dpotrs_","dgetri_","dpotri_","dtrtri_",:Float64),
+ ("sgetrs_","spotrs_","sgetri_","spotri_","strtri_",:Float32),
+ ("zgetrs_","zpotrs_","zgetri_","zpotri_","ztrtri_",:Complex128),
+ ("cgetrs_","cpotrs_","cgetri_","cpotri_","ctrtri_",:Complex64))
 
  @eval begin
  # SUBROUTINE DGETRS( TRANS, N, NRHS, A, LDA, IPIV, B, LDB, INFO )
@@ -13,9 +13,9 @@ for (getrs, potrs, trtrs, getri, potri, trtri, elty) in
  # .. Array Arguments ..
  # INTEGER IPIV( * )
  # DOUBLE PRECISION A( LDA, * ), B( LDB, * )
- function _jl_lapack_getrs(trans::String, A::StridedMatrix{$elty}, ipiv::Vector{Int32}, B::StridedVecOrMat{$elty})
+ function _jl_lapack_getrs(trans::LapackChar, A::StridedMatrix{$elty}, ipiv::Vector{Int32}, B::StridedVecOrMat{$elty})
  if stride(A,1) != 1 || stride(B,1) != 1
- error("_jl_lapack_getrs: matrix columns must have contiguous elements");
+ error("_jl_lapack_getrs: matrix columns must have contiguous elements")
  end
  m, n = size(A)
  if m != n || size(B, 1) != m error("_jl_lapack_getrs: dimension mismatch") end
@@ -27,7 +27,7 @@ for (getrs, potrs, trtrs, getri, potri, trtri, elty) in
  Void,
  (Ptr{Uint8}, Ptr{Int32}, Ptr{Int32}, Ptr{$elty}, Ptr{Int32},
  Ptr{Int32}, Ptr{$elty}, Ptr{Int32}, Ptr{Int32}),
- trans, &n, &nrhs, A, &lda, ipiv, B, &ldb, info)
+ &trans, &n, &nrhs, A, &lda, ipiv, B, &ldb, info)
  if info[1] != 0 error("_jl_lapack_getrs: error $(info[1])") end
  B
  end
@@ -37,7 +37,7 @@ for (getrs, potrs, trtrs, getri, potri, trtri, elty) in
  # INTEGER INFO, LDA, LDB, N, NRHS
  # .. Array Arguments ..
  # DOUBLE PRECISION A( LDA, * ), B( LDB, * )
- function _jl_lapack_potrs(uplo::String, A::StridedMatrix{$elty}, B::StridedVecOrMat{$elty})
+ function _jl_lapack_potrs(uplo::LapackChar, A::StridedMatrix{$elty}, B::StridedVecOrMat{$elty})
  if stride(A,1) != 1 || stride(B,1) != 1
  error("_jl_lapack_potrs: matrix columns must have contiguous elements")
  end
@@ -51,32 +51,7 @@ for (getrs, potrs, trtrs, getri, potri, trtri, elty) in
  Void,
  (Ptr{Uint8}, Ptr{Int32}, Ptr{Int32}, Ptr{$elty}, Ptr{Int32},
  Ptr{$elty}, Ptr{Int32}, Ptr{Int32}),
- uplo, &n, &nrhs, A, &lda, B, &ldb, info)
- if info[1] != 0 error("_jl_lapack_potrs: error $(info[1])") end
- B
- end
- # SUBROUTINE DTRTRS( UPLO, TRANS, DIAG, N, NRHS, A, LDA, B, LDB,
- # INFO )
- # .. Scalar Arguments ..
- # CHARACTER DIAG, TRANS, UPLO
- # INTEGER INFO, LDA, LDB, N, NRHS
- # .. Array Arguments ..
- # DOUBLE PRECISION A( LDA, * ), B( LDB, * )
- function _jl_lapack_trtrs(uplo::String, trans::String, diag::String, A::StridedMatrix{$elty}, B::StridedVecOrMat{$elty})
- if stride(A,1) != 1 || stride(B,1) != 1
- error("_jl_lapack_trtrs: matrix columns must have contiguous elements");
- end
- m, n = size(A)
- if m != n || size(B, 1) != m error("_jl_lapack_trtrs: dimension mismatch") end
- nrhs = size(B, 2)
- lda = stride(A, 2)
- ldb = isa(B, Vector) ? m : stride(B, 2)
- info = Array(Int32, 1)
- ccall(dlsym(_jl_liblapack, $trtrs),
- Void,
- (Ptr{Uint8}, Ptr{Uint8}, Ptr{Uint8}, Ptr{Int32}, Ptr{Int32},
- Ptr{$elty}, Ptr{Int32}, Ptr{$elty}, Ptr{Int32}, Ptr{Int32}),
- uplo, trans, diag, &n, &nrhs, A, &lda, B, &ldb, info)
+ &uplo, &n, &nrhs, A, &lda, B, &ldb, info)
  if info[1] != 0 error("_jl_lapack_potrs: error $(info[1])") end
  B
  end
@@ -116,7 +91,7 @@ for (getrs, potrs, trtrs, getri, potri, trtri, elty) in
  # INTEGER INFO, LDA, N
  # .. Array Arguments ..
  # DOUBLE PRECISION A( LDA, * )
- function _jl_lapack_trtri(uplo::String, diag::String, A::StridedMatrix{$elty})
+ function _jl_lapack_trtri(uplo::LapackChar, diag::LapackChar, A::StridedMatrix{$elty})
  if stride(A,1) != 1
  error("_jl_lapack_trtri: matrix columns must have contiguous elements");
  end
@@ -128,7 +103,7 @@ for (getrs, potrs, trtrs, getri, potri, trtri, elty) in
  Void,
  (Ptr{Uint8}, Ptr{Uint8}, Ptr{Int32}, Ptr{$elty}, Ptr{Int32},
  Ptr{Int32}),
- uplo, diag, &n, A, &lda, info)
+ &uplo, &diag, &n, A, &lda, info)
  if info[1] != 0 error("_jl_lapack_trtri: error $(info[1])") end
  A
  end
@@ -139,16 +114,15 @@ abstract Factorization{T}
 
 type Cholesky{T} <: Factorization{T}
  LR::Matrix{T}
- uplo::String
- function Cholesky(A::Matrix{T}, ul::String)
- UL = uppercase(ul)
- if UL[1] != 'U' && UL[1] != 'L' error("Cholesky: uplo must be 'U' or 'L'") end
+ uplo::LapackChar
+ function Cholesky(A::Matrix{T}, ul::LapackChar)
+ if ul != 'U' && ul != 'L' error("Cholesky: uplo must be 'U' or 'L'") end
  Acopy = copy(A)
- _jl_lapack_potrf(UL, Acopy) == 0 ? new(UL[1] == 'U' ? triu(Acopy) : tril(Acopy), UL) : error("Cholesky: Matrix is not positive-definite")
+ _jl_lapack_potrf(ul, Acopy) == 0 ? new(ul == 'U' ? triu(Acopy) : tril(Acopy), ul) : error("Cholesky: Matrix is not positive-definite")
  end
 end
 
-Cholesky(A::Matrix) = Cholesky{eltype(A)}(A, "U")
+Cholesky(A::Matrix) = Cholesky{eltype(A)}(A, 'U')
 
 (\){T<:Union(Float64,Float32,Complex128,Complex64)}(C::Cholesky{T}, B::StridedVecOrMat{T}) =
  _jl_lapack_potrs(C.uplo, C.LR, copy(B))
@@ -180,7 +154,7 @@ end
 det(A::Matrix) = det(LU(A))
 
 (\){T<:Union(Float64,Float32,Complex128,Complex64)}(lu::LU{T}, B::StridedVecOrMat{T}) =
- _jl_lapack_getrs("N", lu.lu, lu.ipiv, copy(B))
+ _jl_lapack_getrs('N', lu.lu, lu.ipiv, copy(B))
 inv(lu::LU) = _jl_lapack_getri(copy(lu.lu), lu.ipiv)
 
 ## Multiplication by Q or Q' from a QR factorization
@@ -197,7 +171,7 @@ for (orm2r, elty) in
  # INTEGER INFO, K, LDA, LDC, M, N
  # .. Array Arguments ..
  # DOUBLE PRECISION A( LDA, * ), C( LDC, * ), TAU( * ), WORK( * )
- function _jl_lapack_orm2r(side::String, trans::String, A::StridedMatrix{$elty}, k::Integer, tau::Vector{$elty}, C::StridedVecOrMat{$elty})
+ function _jl_lapack_orm2r(side::LapackChar, trans::LapackChar, A::StridedMatrix{$elty}, k::Integer, tau::Vector{$elty}, C::StridedVecOrMat{$elty})
  if stride(A,1) != 1 || stride(C,1) != 1
  error("_jl_lapack_orm2r: matrix columns must have contiguous elements");
  end
@@ -206,14 +180,13 @@ for (orm2r, elty) in
  if size(A, 1) != m error("_jl_lapack_orm2r: dimension mismatch") end
  lda = stride(A, 2)
  ldc = isa(C, Vector) ? m : stride(C, 2)
- SIDE = uppercase(side)
- work = Array($elty, SIDE[1] == 'L' ? n : m)
+ work = Array($elty, side == 'L' ? n : m)
  info = Array(Int32, 1)
  ccall(dlsym(_jl_liblapack, $orm2r),
  Void,
  (Ptr{Uint8}, Ptr{Uint8}, Ptr{Int32}, Ptr{Int32}, Ptr{Int32}, Ptr{$elty},
  Ptr{Int32}, Ptr{$elty}, Ptr{$elty}, Ptr{Int32}, Ptr{$elty}, Ptr{Int32}),
- side, trans, &m, &n, &k, A, &lda, tau, C, &ldc, work, info)
+ &side, &trans, &m, &n, &k, A, &lda, tau, C, &ldc, work, info)
  if info[1] != 0 error("_jl_lapack_orm2r: error $(info[1])") end
  C
  end
@@ -239,13 +212,13 @@ size{T<:Union(Float64,Float32,Complex128,Complex64)}(A::QR{T},n) = size(A.hh,n)
 ## Multiplication by Q from the QR decomposition
 function (*){T<:Union(Float64,Float32,Complex128,Complex64)}(A::QR{T},
  B::StridedVecOrMat{T})
- _jl_lapack_orm2r("L", "N", A.hh, size(A, 2), A.tau, copy(B))
+ _jl_lapack_orm2r('L', 'N', A.hh, size(A, 2), A.tau, copy(B))
 end
 
 ## Multiplication by Q' from the QR decomposition
 function Ac_mul_B{T<:Union(Float64,Float32,Complex128,Complex64)}(A::QR{T},
  B::StridedVecOrMat{T})
- _jl_lapack_orm2r("L", iscomplex(A.tau) ? "C" : "T", A.hh, size(A, 2), A.tau, copy(B))
+ _jl_lapack_orm2r('L', iscomplex(A.tau) ? 'C' : 'T', A.hh, size(A, 2), A.tau, copy(B))
 end
 
 ## Least squares solution. Should be more careful about cases with m < n
@@ -254,7 +227,7 @@ function (\){T<:Union(Float64,Float32,Complex128,Complex64)}(A::QR{T},
  n = numel(A.tau)
  qtb = isa(B, Vector) ? (A' * B)[1:n] : (A' * B)[1:n, :]
  ## Not sure if this avoids copying A.hh[1:n,:] but at least it is not all of A.hh
- _jl_lapack_trtrs("U", "N", "N", A.hh[1:n,:], qtb)
+ _jl_lapack_trtrs('U','N','N', A.hh[1:n,:], qtb)
 end
 
 type QRP{T} <: Factorization{T}
@@ -279,12 +252,12 @@ size{T<:Union(Float64,Float32,Complex128,Complex64)}(A::QRP{T},n) = size(A.hh,n)
 
 function (*){T<:Union(Float64,Float32,Complex128,Complex64)}(A::QRP{T},
  B::StridedVecOrMat{T})
- _jl_lapack_orm2r("L", "N", A.hh, size(A, 2), A.tau, copy(B))
+ _jl_lapack_orm2r('L', 'N', A.hh, size(A, 2), A.tau, copy(B))
 end
 
 function Ac_mul_B{T<:Union(Float64,Float32,Complex128,Complex64)}(A::QRP{T},
  B::StridedVecOrMat{T})
- _jl_lapack_orm2r("L", iscomplex(A.tau) ? "C" : "T", A.hh, size(A, 2), A.tau, copy(B))
+ _jl_lapack_orm2r('L', iscomplex(A.tau) ? 'C' : 'T', A.hh, size(A, 2), A.tau, copy(B))
 end
 
 function (\){T<:Union(Float64,Float32,Complex128,Complex64)}(A::QRP{T},
@@ -301,7 +274,7 @@ function (\){T<:Union(Float64,Float32,Complex128,Complex64)}(A::QRP{T},
  BV = isa(B, Vector)
  n = numel(A.tau)
  qtb = BV ? (A' * B)[1:n] : (A' * B)[1:n, :]
- ans = _jl_lapack_trtrs("U", "N", "N", A.hh[1:n,:], qtb)
+ ans = _jl_lapack_trtrs('U', 'N', 'N', A.hh[1:n,:], qtb)
  BV ? ans[invperm(A.jpvt)] : ans[invperm(A.jpvt), :]
 end