obtained gcc-4.6.4.tar.bz2 from upstream website;upstream

verified gcc-4.6.4.tar.bz2.sig;
imported gcc-4.6.4 source tree from verified upstream tarball.

downloading a git-generated archive based on the 'upstream' tag
should provide you with a source tree that is binary identical
to the one extracted from the above tarball.

if you have obtained the source via the command 'git clone',
however, do note that line-endings of files in your working
directory might differ from line-endings of the respective
files in the upstream repository.

1 files changed, 311 insertions, 0 deletions

diff --git a/gcc/ada/s-gerebl.adb b/gcc/ada/s-gerebl.adb
new file mode 100644
index 000000000..fc2f5d7d6
--- /dev/null
+++ b/gcc/ada/s-gerebl.adb
@@ -0,0 +1,311 @@
+------------------------------------------------------------------------------
+--                                                                          --
+--                         GNAT RUN-TIME COMPONENTS                         --
+--                                                                          --
+--             S Y S T E M . G E N E R I C _ R E A L _ B L A S              --
+--                                                                          --
+--                                 B o d y                                  --
+--                                                                          --
+--          Copyright (C) 2006-2009, Free Software Foundation, Inc.         --
+--                                                                          --
+-- GNAT is free software;  you can  redistribute it  and/or modify it under --
+-- terms of the  GNU General Public License as published  by the Free Soft- --
+-- ware  Foundation;  either version 3,  or (at your option) any later ver- --
+-- sion.  GNAT is distributed in the hope that it will be useful, but WITH- --
+-- OUT ANY WARRANTY;  without even the  implied warranty of MERCHANTABILITY --
+-- or FITNESS FOR A PARTICULAR PURPOSE.                                     --
+--                                                                          --
+-- As a special exception under Section 7 of GPL version 3, you are granted --
+-- additional permissions described in the GCC Runtime Library Exception,   --
+-- version 3.1, as published by the Free Software Foundation.               --
+--                                                                          --
+-- You should have received a copy of the GNU General Public License and    --
+-- a copy of the GCC Runtime Library Exception along with this program;     --
+-- see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see    --
+-- <http://www.gnu.org/licenses/>.                                          --
+--                                                                          --
+-- GNAT was originally developed  by the GNAT team at  New York University. --
+-- Extensive contributions were provided by Ada Core Technologies Inc.      --
+--                                                                          --
+------------------------------------------------------------------------------
+
+with Ada.Unchecked_Conversion;        use Ada;
+with Interfaces;                      use Interfaces;
+with Interfaces.Fortran;              use Interfaces.Fortran;
+with Interfaces.Fortran.BLAS;         use Interfaces.Fortran.BLAS;
+with System.Generic_Array_Operations; use System.Generic_Array_Operations;
+
+package body System.Generic_Real_BLAS is
+
+   Is_Single : constant Boolean :=
+                 Real'Machine_Mantissa = Fortran.Real'Machine_Mantissa
+                  and then Fortran.Real (Real'First) = Fortran.Real'First
+                  and then Fortran.Real (Real'Last) = Fortran.Real'Last;
+
+   Is_Double : constant Boolean :=
+                 Real'Machine_Mantissa = Double_Precision'Machine_Mantissa
+                  and then
+                    Double_Precision (Real'First) = Double_Precision'First
+                  and then
+                    Double_Precision (Real'Last) = Double_Precision'Last;
+
+   --  Local subprograms
+
+   function To_Double_Precision (X : Real) return Double_Precision;
+   pragma Inline_Always (To_Double_Precision);
+
+   function To_Real (X : Double_Precision) return Real;
+   pragma Inline_Always (To_Real);
+
+   --  Instantiations
+
+   function To_Double_Precision is new
+     Vector_Elementwise_Operation
+       (X_Scalar      => Real,
+        Result_Scalar => Double_Precision,
+        X_Vector      => Real_Vector,
+        Result_Vector => Double_Precision_Vector,
+        Operation     => To_Double_Precision);
+
+   function To_Real is new
+     Vector_Elementwise_Operation
+       (X_Scalar      => Double_Precision,
+        Result_Scalar => Real,
+        X_Vector      => Double_Precision_Vector,
+        Result_Vector => Real_Vector,
+        Operation     => To_Real);
+
+   function To_Double_Precision is new
+     Matrix_Elementwise_Operation
+       (X_Scalar      => Real,
+        Result_Scalar => Double_Precision,
+        X_Matrix      => Real_Matrix,
+        Result_Matrix => Double_Precision_Matrix,
+        Operation     => To_Double_Precision);
+
+   function To_Real is new
+     Matrix_Elementwise_Operation
+       (X_Scalar      => Double_Precision,
+        Result_Scalar => Real,
+        X_Matrix      => Double_Precision_Matrix,
+        Result_Matrix => Real_Matrix,
+        Operation     => To_Real);
+
+   function To_Double_Precision (X : Real) return Double_Precision is
+   begin
+      return Double_Precision (X);
+   end To_Double_Precision;
+
+   function To_Real (X : Double_Precision) return Real is
+   begin
+      return Real (X);
+   end To_Real;
+
+   ---------
+   -- dot --
+   ---------
+
+   function dot
+     (N     : Positive;
+      X     : Real_Vector;
+      Inc_X : Integer := 1;
+      Y     : Real_Vector;
+      Inc_Y : Integer := 1) return Real
+   is
+   begin
+      if Is_Single then
+         declare
+            type X_Ptr is access all BLAS.Real_Vector (X'Range);
+            type Y_Ptr is access all BLAS.Real_Vector (Y'Range);
+            function Conv_X is new Unchecked_Conversion (Address, X_Ptr);
+            function Conv_Y is new Unchecked_Conversion (Address, Y_Ptr);
+         begin
+            return Real (sdot (N, Conv_X (X'Address).all, Inc_X,
+                                  Conv_Y (Y'Address).all, Inc_Y));
+         end;
+
+      elsif Is_Double then
+         declare
+            type X_Ptr is access all BLAS.Double_Precision_Vector (X'Range);
+            type Y_Ptr is access all BLAS.Double_Precision_Vector (Y'Range);
+            function Conv_X is new Unchecked_Conversion (Address, X_Ptr);
+            function Conv_Y is new Unchecked_Conversion (Address, Y_Ptr);
+         begin
+            return Real (ddot (N, Conv_X (X'Address).all, Inc_X,
+                                  Conv_Y (Y'Address).all, Inc_Y));
+         end;
+
+      else
+         return Real (ddot (N, To_Double_Precision (X), Inc_X,
+                               To_Double_Precision (Y), Inc_Y));
+      end if;
+   end dot;
+
+   ----------
+   -- gemm --
+   ----------
+
+   procedure gemm
+     (Trans_A : access constant Character;
+      Trans_B : access constant Character;
+      M       : Positive;
+      N       : Positive;
+      K       : Positive;
+      Alpha   : Real := 1.0;
+      A       : Real_Matrix;
+      Ld_A    : Integer;
+      B       : Real_Matrix;
+      Ld_B    : Integer;
+      Beta    : Real := 0.0;
+      C       : in out Real_Matrix;
+      Ld_C    : Integer)
+   is
+   begin
+      if Is_Single then
+         declare
+            subtype A_Type is BLAS.Real_Matrix (A'Range (1), A'Range (2));
+            subtype B_Type is BLAS.Real_Matrix (B'Range (1), B'Range (2));
+            type C_Ptr is
+              access all BLAS.Real_Matrix (C'Range (1), C'Range (2));
+            function Conv_A is new Unchecked_Conversion (Real_Matrix, A_Type);
+            function Conv_B is new Unchecked_Conversion (Real_Matrix, B_Type);
+            function Conv_C is new Unchecked_Conversion (Address, C_Ptr);
+         begin
+            sgemm (Trans_A, Trans_B, M, N, K, Fortran.Real (Alpha),
+                   Conv_A (A), Ld_A, Conv_B (B), Ld_B, Fortran.Real (Beta),
+                   Conv_C (C'Address).all, Ld_C);
+         end;
+
+      elsif Is_Double then
+         declare
+            subtype A_Type is
+               Double_Precision_Matrix (A'Range (1), A'Range (2));
+            subtype B_Type is
+               Double_Precision_Matrix (B'Range (1), B'Range (2));
+            type C_Ptr is
+              access all Double_Precision_Matrix (C'Range (1), C'Range (2));
+            function Conv_A is new Unchecked_Conversion (Real_Matrix, A_Type);
+            function Conv_B is new Unchecked_Conversion (Real_Matrix, B_Type);
+            function Conv_C is new Unchecked_Conversion (Address, C_Ptr);
+         begin
+            dgemm (Trans_A, Trans_B, M, N, K, Double_Precision (Alpha),
+                   Conv_A (A), Ld_A, Conv_B (B), Ld_B, Double_Precision (Beta),
+                   Conv_C (C'Address).all, Ld_C);
+         end;
+
+      else
+         declare
+            DP_C : Double_Precision_Matrix (C'Range (1), C'Range (2));
+         begin
+            if Beta /= 0.0 then
+               DP_C := To_Double_Precision (C);
+            end if;
+
+            dgemm (Trans_A, Trans_B, M, N, K, Double_Precision (Alpha),
+                   To_Double_Precision (A), Ld_A,
+                   To_Double_Precision (B), Ld_B, Double_Precision (Beta),
+                   DP_C, Ld_C);
+
+            C := To_Real (DP_C);
+         end;
+      end if;
+   end gemm;
+
+   ----------
+   -- gemv --
+   ----------
+
+   procedure gemv
+     (Trans : access constant Character;
+      M     : Natural := 0;
+      N     : Natural := 0;
+      Alpha : Real := 1.0;
+      A     : Real_Matrix;
+      Ld_A  : Positive;
+      X     : Real_Vector;
+      Inc_X : Integer := 1;
+      Beta  : Real := 0.0;
+      Y     : in out Real_Vector;
+      Inc_Y : Integer := 1)
+   is
+   begin
+      if Is_Single then
+         declare
+            subtype A_Type is BLAS.Real_Matrix (A'Range (1), A'Range (2));
+            subtype X_Type is BLAS.Real_Vector (X'Range);
+            type Y_Ptr is access all BLAS.Real_Vector (Y'Range);
+            function Conv_A is new Unchecked_Conversion (Real_Matrix, A_Type);
+            function Conv_X is new Unchecked_Conversion (Real_Vector, X_Type);
+            function Conv_Y is new Unchecked_Conversion (Address, Y_Ptr);
+         begin
+            sgemv (Trans, M, N, Fortran.Real (Alpha),
+                   Conv_A (A), Ld_A, Conv_X (X), Inc_X, Fortran.Real (Beta),
+                   Conv_Y (Y'Address).all, Inc_Y);
+         end;
+
+      elsif Is_Double then
+         declare
+            subtype A_Type is
+               Double_Precision_Matrix (A'Range (1), A'Range (2));
+            subtype X_Type is Double_Precision_Vector (X'Range);
+            type Y_Ptr is access all Double_Precision_Vector (Y'Range);
+            function Conv_A is new Unchecked_Conversion (Real_Matrix, A_Type);
+            function Conv_X is new Unchecked_Conversion (Real_Vector, X_Type);
+            function Conv_Y is new Unchecked_Conversion (Address, Y_Ptr);
+         begin
+            dgemv (Trans, M, N, Double_Precision (Alpha),
+                   Conv_A (A), Ld_A, Conv_X (X), Inc_X,
+                   Double_Precision (Beta),
+                   Conv_Y (Y'Address).all, Inc_Y);
+         end;
+
+      else
+         declare
+            DP_Y : Double_Precision_Vector (Y'Range);
+         begin
+            if Beta /= 0.0 then
+               DP_Y := To_Double_Precision (Y);
+            end if;
+
+            dgemv (Trans, M, N, Double_Precision (Alpha),
+                   To_Double_Precision (A), Ld_A,
+                   To_Double_Precision (X), Inc_X, Double_Precision (Beta),
+                   DP_Y, Inc_Y);
+
+            Y := To_Real (DP_Y);
+         end;
+      end if;
+   end gemv;
+
+   ----------
+   -- nrm2 --
+   ----------
+
+   function nrm2
+     (N     : Natural;
+      X     : Real_Vector;
+      Inc_X : Integer := 1) return Real
+   is
+   begin
+      if Is_Single then
+         declare
+            subtype X_Type is BLAS.Real_Vector (X'Range);
+            function Conv_X is new Unchecked_Conversion (Real_Vector, X_Type);
+         begin
+            return Real (snrm2 (N, Conv_X (X), Inc_X));
+         end;
+
+      elsif Is_Double then
+         declare
+            subtype X_Type is Double_Precision_Vector (X'Range);
+            function Conv_X is new Unchecked_Conversion (Real_Vector, X_Type);
+         begin
+            return Real (dnrm2 (N, Conv_X (X), Inc_X));
+         end;
+
+      else
+         return Real (dnrm2 (N, To_Double_Precision (X), Inc_X));
+      end if;
+   end nrm2;
+
+end System.Generic_Real_BLAS;