------------------------------------------------------------------------------
-- --
-- GNAT RUN-TIME COMPONENTS --
-- --
-- S Y S T E M . P A C K _ 1 0 1 --
-- --
-- B o d y --
-- --
-- Copyright (C) 1992-2025, 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 System.Address_To_Access_Conversions;
with System.Storage_Elements;
with System.Unsigned_Types;
package body System.Pack_101 is
-- The high-level idea of the implementation is to overlay a record
-- containing components of the same size as that of the component
-- of the array, in order words 101 bits, and to access the indexed
-- component of the array through the appropriate selected component
-- of the record.
-- The record must be of a fixed size for technical reasons, so we
-- effectively overlay a series of contiguous records containing 8
-- components (so that their size in bits is a multiple of a byte)
-- at the start of the array and access the component in the last
-- of them. However, this component in the last record may also be
-- mapped to the last component of the array, which means that the
-- generated code cannot safely access past it (or its last byte).
-- That's why the last record of the series is shortened, so the
-- accessed component is always the last component of the record.
-- A (0) A (N)
-- | |
-- V V
-- ---------------------------------------------------------------
-- | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
-- ---------------------------------------------------------------
--
-- component K
-- |
-- V
-- ---------------------------------------------------------
-- | | | | |
-- ---------------------------------------------------------
-- | | | |
-- Cluster7 Cluster7 Cluster7 ClusterK
--
-- where the number of Cluster7 is N / 8 and K is N mod 8.
subtype Bit_Order is System.Bit_Order;
Reverse_Bit_Order : constant Bit_Order :=
Bit_Order'Val (1 - Bit_Order'Pos (System.Default_Bit_Order));
subtype Ofs is System.Storage_Elements.Storage_Offset;
subtype Uns is System.Unsigned_Types.Unsigned;
subtype N07 is System.Unsigned_Types.Unsigned range 0 .. 7;
use type System.Storage_Elements.Storage_Offset;
use type System.Unsigned_Types.Unsigned;
--------------
-- Cluster0 --
--------------
type Cluster0 is record
E0 : Bits_101;
end record;
for Cluster0 use record
E0 at 0 range 0 * Bits .. 0 * Bits + Bits - 1;
end record;
for Cluster0'Size use Bits * (1 + 0);
for Cluster0'Alignment use Integer'Min (Standard'Maximum_Alignment,
1 +
1 * Boolean'Pos (Bits mod 2 = 0) +
2 * Boolean'Pos (Bits mod 4 = 0));
-- Use maximum possible alignment, given the bit field size, since this
-- will result in the most efficient code possible for the field.
package AAC0 is new Address_To_Access_Conversions (Cluster0);
-- We convert addresses to access values and dereference them instead of
-- directly using overlays in order to work around the implementation of
-- the RM 13.3(19) clause, which would pessimize the generated code.
type Rev_Cluster0 is new Cluster0
with Bit_Order => Reverse_Bit_Order,
Scalar_Storage_Order => Reverse_Bit_Order;
package Rev_AAC0 is new Address_To_Access_Conversions (Rev_Cluster0);
--------------
-- Cluster1 --
--------------
type Cluster1 is record
E0, E1 : Bits_101;
end record;
for Cluster1 use record
E0 at 0 range 0 * Bits .. 0 * Bits + Bits - 1;
E1 at 0 range 1 * Bits .. 1 * Bits + Bits - 1;
end record;
for Cluster1'Size use Bits * (1 + 1);
for Cluster1'Alignment use Integer'Min (Standard'Maximum_Alignment,
1 +
1 * Boolean'Pos (Bits mod 2 = 0) +
2 * Boolean'Pos (Bits mod 4 = 0));
-- Use maximum possible alignment, given the bit field size, since this
-- will result in the most efficient code possible for the field.
package AAC1 is new Address_To_Access_Conversions (Cluster1);
-- We convert addresses to access values and dereference them instead of
-- directly using overlays in order to work around the implementation of
-- the RM 13.3(19) clause, which would pessimize the generated code.
type Rev_Cluster1 is new Cluster1
with Bit_Order => Reverse_Bit_Order,
Scalar_Storage_Order => Reverse_Bit_Order;
package Rev_AAC1 is new Address_To_Access_Conversions (Rev_Cluster1);
--------------
-- Cluster2 --
--------------
type Cluster2 is record
E0, E1, E2 : Bits_101;
end record;
for Cluster2 use record
E0 at 0 range 0 * Bits .. 0 * Bits + Bits - 1;
E1 at 0 range 1 * Bits .. 1 * Bits + Bits - 1;
E2 at 0 range 2 * Bits .. 2 * Bits + Bits - 1;
end record;
for Cluster2'Size use Bits * (1 + 2);
for Cluster2'Alignment use Integer'Min (Standard'Maximum_Alignment,
1 +
1 * Boolean'Pos (Bits mod 2 = 0) +
2 * Boolean'Pos (Bits mod 4 = 0));
-- Use maximum possible alignment, given the bit field size, since this
-- will result in the most efficient code possible for the field.
package AAC2 is new Address_To_Access_Conversions (Cluster2);
-- We convert addresses to access values and dereference them instead of
-- directly using overlays in order to work around the implementation of
-- the RM 13.3(19) clause, which would pessimize the generated code.
type Rev_Cluster2 is new Cluster2
with Bit_Order => Reverse_Bit_Order,
Scalar_Storage_Order => Reverse_Bit_Order;
package Rev_AAC2 is new Address_To_Access_Conversions (Rev_Cluster2);
--------------
-- Cluster3 --
--------------
type Cluster3 is record
E0, E1, E2, E3 : Bits_101;
end record;
for Cluster3 use record
E0 at 0 range 0 * Bits .. 0 * Bits + Bits - 1;
E1 at 0 range 1 * Bits .. 1 * Bits + Bits - 1;
E2 at 0 range 2 * Bits .. 2 * Bits + Bits - 1;
E3 at 0 range 3 * Bits .. 3 * Bits + Bits - 1;
end record;
for Cluster3'Size use Bits * (1 + 3);
for Cluster3'Alignment use Integer'Min (Standard'Maximum_Alignment,
1 +
1 * Boolean'Pos (Bits mod 2 = 0) +
2 * Boolean'Pos (Bits mod 4 = 0));
-- Use maximum possible alignment, given the bit field size, since this
-- will result in the most efficient code possible for the field.
package AAC3 is new Address_To_Access_Conversions (Cluster3);
-- We convert addresses to access values and dereference them instead of
-- directly using overlays in order to work around the implementation of
-- the RM 13.3(19) clause, which would pessimize the generated code.
type Rev_Cluster3 is new Cluster3
with Bit_Order => Reverse_Bit_Order,
Scalar_Storage_Order => Reverse_Bit_Order;
package Rev_AAC3 is new Address_To_Access_Conversions (Rev_Cluster3);
--------------
-- Cluster4 --
--------------
type Cluster4 is record
E0, E1, E2, E3, E4 : Bits_101;
end record;
for Cluster4 use record
E0 at 0 range 0 * Bits .. 0 * Bits + Bits - 1;
E1 at 0 range 1 * Bits .. 1 * Bits + Bits - 1;
E2 at 0 range 2 * Bits .. 2 * Bits + Bits - 1;
E3 at 0 range 3 * Bits .. 3 * Bits + Bits - 1;
E4 at 0 range 4 * Bits .. 4 * Bits + Bits - 1;
end record;
for Cluster4'Size use Bits * (1 + 4);
for Cluster4'Alignment use Integer'Min (Standard'Maximum_Alignment,
1 +
1 * Boolean'Pos (Bits mod 2 = 0) +
2 * Boolean'Pos (Bits mod 4 = 0));
-- Use maximum possible alignment, given the bit field size, since this
-- will result in the most efficient code possible for the field.
package AAC4 is new Address_To_Access_Conversions (Cluster4);
-- We convert addresses to access values and dereference them instead of
-- directly using overlays in order to work around the implementation of
-- the RM 13.3(19) clause, which would pessimize the generated code.
type Rev_Cluster4 is new Cluster4
with Bit_Order => Reverse_Bit_Order,
Scalar_Storage_Order => Reverse_Bit_Order;
package Rev_AAC4 is new Address_To_Access_Conversions (Rev_Cluster4);
--------------
-- Cluster5 --
--------------
type Cluster5 is record
E0, E1, E2, E3, E4, E5 : Bits_101;
end record;
for Cluster5 use record
E0 at 0 range 0 * Bits .. 0 * Bits + Bits - 1;
E1 at 0 range 1 * Bits .. 1 * Bits + Bits - 1;
E2 at 0 range 2 * Bits .. 2 * Bits + Bits - 1;
E3 at 0 range 3 * Bits .. 3 * Bits + Bits - 1;
E4 at 0 range 4 * Bits .. 4 * Bits + Bits - 1;
E5 at 0 range 5 * Bits .. 5 * Bits + Bits - 1;
end record;
for Cluster5'Size use Bits * (1 + 5);
for Cluster5'Alignment use Integer'Min (Standard'Maximum_Alignment,
1 +
1 * Boolean'Pos (Bits mod 2 = 0) +
2 * Boolean'Pos (Bits mod 4 = 0));
-- Use maximum possible alignment, given the bit field size, since this
-- will result in the most efficient code possible for the field.
package AAC5 is new Address_To_Access_Conversions (Cluster5);
-- We convert addresses to access values and dereference them instead of
-- directly using overlays in order to work around the implementation of
-- the RM 13.3(19) clause, which would pessimize the generated code.
type Rev_Cluster5 is new Cluster5
with Bit_Order => Reverse_Bit_Order,
Scalar_Storage_Order => Reverse_Bit_Order;
package Rev_AAC5 is new Address_To_Access_Conversions (Rev_Cluster5);
--------------
-- Cluster6 --
--------------
type Cluster6 is record
E0, E1, E2, E3, E4, E5, E6 : Bits_101;
end record;
for Cluster6 use record
E0 at 0 range 0 * Bits .. 0 * Bits + Bits - 1;
E1 at 0 range 1 * Bits .. 1 * Bits + Bits - 1;
E2 at 0 range 2 * Bits .. 2 * Bits + Bits - 1;
E3 at 0 range 3 * Bits .. 3 * Bits + Bits - 1;
E4 at 0 range 4 * Bits .. 4 * Bits + Bits - 1;
E5 at 0 range 5 * Bits .. 5 * Bits + Bits - 1;
E6 at 0 range 6 * Bits .. 6 * Bits + Bits - 1;
end record;
for Cluster6'Size use Bits * (1 + 6);
for Cluster6'Alignment use Integer'Min (Standard'Maximum_Alignment,
1 +
1 * Boolean'Pos (Bits mod 2 = 0) +
2 * Boolean'Pos (Bits mod 4 = 0));
-- Use maximum possible alignment, given the bit field size, since this
-- will result in the most efficient code possible for the field.
package AAC6 is new Address_To_Access_Conversions (Cluster6);
-- We convert addresses to access values and dereference them instead of
-- directly using overlays in order to work around the implementation of
-- the RM 13.3(19) clause, which would pessimize the generated code.
type Rev_Cluster6 is new Cluster6
with Bit_Order => Reverse_Bit_Order,
Scalar_Storage_Order => Reverse_Bit_Order;
package Rev_AAC6 is new Address_To_Access_Conversions (Rev_Cluster6);
--------------
-- Cluster7 --
--------------
type Cluster7 is record
E0, E1, E2, E3, E4, E5, E6, E7 : Bits_101;
end record;
for Cluster7 use record
E0 at 0 range 0 * Bits .. 0 * Bits + Bits - 1;
E1 at 0 range 1 * Bits .. 1 * Bits + Bits - 1;
E2 at 0 range 2 * Bits .. 2 * Bits + Bits - 1;
E3 at 0 range 3 * Bits .. 3 * Bits + Bits - 1;
E4 at 0 range 4 * Bits .. 4 * Bits + Bits - 1;
E5 at 0 range 5 * Bits .. 5 * Bits + Bits - 1;
E6 at 0 range 6 * Bits .. 6 * Bits + Bits - 1;
E7 at 0 range 7 * Bits .. 7 * Bits + Bits - 1;
end record;
for Cluster7'Size use Bits * (1 + 7);
for Cluster7'Alignment use Integer'Min (Standard'Maximum_Alignment,
1 +
1 * Boolean'Pos (Bits mod 2 = 0) +
2 * Boolean'Pos (Bits mod 4 = 0));
-- Use maximum possible alignment, given the bit field size, since this
-- will result in the most efficient code possible for the field.
package AAC7 is new Address_To_Access_Conversions (Cluster7);
-- We convert addresses to access values and dereference them instead of
-- directly using overlays in order to work around the implementation of
-- the RM 13.3(19) clause, which would pessimize the generated code.
type Rev_Cluster7 is new Cluster7
with Bit_Order => Reverse_Bit_Order,
Scalar_Storage_Order => Reverse_Bit_Order;
package Rev_AAC7 is new Address_To_Access_Conversions (Rev_Cluster7);
------------
-- Get_101 --
------------
function Get_101
(Arr : System.Address;
N : Natural;
Rev_SSO : Boolean) return Bits_101
is
A : constant System.Address := Arr + Bits * Ofs (Uns (N) / 8);
C0 : constant AAC0.Object_Pointer := AAC0.To_Pointer (A);
C1 : constant AAC1.Object_Pointer := AAC1.To_Pointer (A);
C2 : constant AAC2.Object_Pointer := AAC2.To_Pointer (A);
C3 : constant AAC3.Object_Pointer := AAC3.To_Pointer (A);
C4 : constant AAC4.Object_Pointer := AAC4.To_Pointer (A);
C5 : constant AAC5.Object_Pointer := AAC5.To_Pointer (A);
C6 : constant AAC6.Object_Pointer := AAC6.To_Pointer (A);
C7 : constant AAC7.Object_Pointer := AAC7.To_Pointer (A);
RC0 : constant Rev_AAC0.Object_Pointer := Rev_AAC0.To_Pointer (A);
RC1 : constant Rev_AAC1.Object_Pointer := Rev_AAC1.To_Pointer (A);
RC2 : constant Rev_AAC2.Object_Pointer := Rev_AAC2.To_Pointer (A);
RC3 : constant Rev_AAC3.Object_Pointer := Rev_AAC3.To_Pointer (A);
RC4 : constant Rev_AAC4.Object_Pointer := Rev_AAC4.To_Pointer (A);
RC5 : constant Rev_AAC5.Object_Pointer := Rev_AAC5.To_Pointer (A);
RC6 : constant Rev_AAC6.Object_Pointer := Rev_AAC6.To_Pointer (A);
RC7 : constant Rev_AAC7.Object_Pointer := Rev_AAC7.To_Pointer (A);
begin
return
(if Rev_SSO then
(case N07 (Uns (N) mod 8) is
when 0 => RC0.E0,
when 1 => RC1.E1,
when 2 => RC2.E2,
when 3 => RC3.E3,
when 4 => RC4.E4,
when 5 => RC5.E5,
when 6 => RC6.E6,
when 7 => RC7.E7)
else
(case N07 (Uns (N) mod 8) is
when 0 => C0.E0,
when 1 => C1.E1,
when 2 => C2.E2,
when 3 => C3.E3,
when 4 => C4.E4,
when 5 => C5.E5,
when 6 => C6.E6,
when 7 => C7.E7)
);
end Get_101;
------------
-- Set_101 --
------------
procedure Set_101
(Arr : System.Address;
N : Natural;
E : Bits_101;
Rev_SSO : Boolean)
is
A : constant System.Address := Arr + Bits * Ofs (Uns (N) / 8);
C0 : constant AAC0.Object_Pointer := AAC0.To_Pointer (A);
C1 : constant AAC1.Object_Pointer := AAC1.To_Pointer (A);
C2 : constant AAC2.Object_Pointer := AAC2.To_Pointer (A);
C3 : constant AAC3.Object_Pointer := AAC3.To_Pointer (A);
C4 : constant AAC4.Object_Pointer := AAC4.To_Pointer (A);
C5 : constant AAC5.Object_Pointer := AAC5.To_Pointer (A);
C6 : constant AAC6.Object_Pointer := AAC6.To_Pointer (A);
C7 : constant AAC7.Object_Pointer := AAC7.To_Pointer (A);
RC0 : constant Rev_AAC0.Object_Pointer := Rev_AAC0.To_Pointer (A);
RC1 : constant Rev_AAC1.Object_Pointer := Rev_AAC1.To_Pointer (A);
RC2 : constant Rev_AAC2.Object_Pointer := Rev_AAC2.To_Pointer (A);
RC3 : constant Rev_AAC3.Object_Pointer := Rev_AAC3.To_Pointer (A);
RC4 : constant Rev_AAC4.Object_Pointer := Rev_AAC4.To_Pointer (A);
RC5 : constant Rev_AAC5.Object_Pointer := Rev_AAC5.To_Pointer (A);
RC6 : constant Rev_AAC6.Object_Pointer := Rev_AAC6.To_Pointer (A);
RC7 : constant Rev_AAC7.Object_Pointer := Rev_AAC7.To_Pointer (A);
begin
if Rev_SSO then
case N07 (Uns (N) mod 8) is
when 0 => RC0.E0 := E;
when 1 => RC1.E1 := E;
when 2 => RC2.E2 := E;
when 3 => RC3.E3 := E;
when 4 => RC4.E4 := E;
when 5 => RC5.E5 := E;
when 6 => RC6.E6 := E;
when 7 => RC7.E7 := E;
end case;
else
case N07 (Uns (N) mod 8) is
when 0 => C0.E0 := E;
when 1 => C1.E1 := E;
when 2 => C2.E2 := E;
when 3 => C3.E3 := E;
when 4 => C4.E4 := E;
when 5 => C5.E5 := E;
when 6 => C6.E6 := E;
when 7 => C7.E7 := E;
end case;
end if;
end Set_101;
end System.Pack_101;