* [Ada] Fix PR ada/63225
@ 2014-10-13 8:34 Eric Botcazou
0 siblings, 0 replies; only message in thread
From: Eric Botcazou @ 2014-10-13 8:34 UTC (permalink / raw)
To: gcc-patches; +Cc: Alan Modra
[-- Attachment #1: Type: text/plain, Size: 467 bytes --]
As reported by Alan, the Ada compiler doesn't build anymore if you compile it
with -fno-inline because cuintp.c references a private function of uintp.adb.
Tested on x86_64-suse-linux, applied on all active branches.
2014-10-13 Eric Botcazou <ebotcazou@adacore.com>
Alan Modra <amodra@gmail.com>
PR ada/63225
* uintp.adb (Vector_To_Uint): Move from here to...
* uintp.ads (UI_Vector): Make public.
(Vector_To_Uint): ...here.
--
Eric Botcazou
[-- Attachment #2: pr63225.diff --]
[-- Type: text/x-patch, Size: 4404 bytes --]
Index: uintp.ads
===================================================================
--- uintp.ads (revision 216116)
+++ uintp.ads (working copy)
@@ -90,6 +90,18 @@ package Uintp is
Uint_Minus_80 : constant Uint;
Uint_Minus_128 : constant Uint;
+ type UI_Vector is array (Pos range <>) of Int;
+ -- Vector containing the integer values of a Uint value
+
+ -- Note: An earlier version of this package used pointers of arrays of Ints
+ -- (dynamically allocated) for the Uint type. The change leads to a few
+ -- less natural idioms used throughout this code, but eliminates all uses
+ -- of the heap except for the table package itself. For example, Uint
+ -- parameters are often converted to UI_Vectors for internal manipulation.
+ -- This is done by creating the local UI_Vector using the function N_Digits
+ -- on the Uint to find the size needed for the vector, and then calling
+ -- Init_Operand to copy the values out of the table into the vector.
+
-----------------
-- Subprograms --
-----------------
@@ -252,6 +264,22 @@ package Uintp is
-- function is used for capacity checks, and it can be one bit off
-- without affecting its usage.
+ function Vector_To_Uint
+ (In_Vec : UI_Vector;
+ Negative : Boolean) return Uint;
+ -- Functions that calculate values in UI_Vectors, call this function to
+ -- create and return the Uint value. In_Vec contains the multiple precision
+ -- (Base) representation of a non-negative value. Leading zeroes are
+ -- permitted. Negative is set if the desired result is the negative of the
+ -- given value. The result will be either the appropriate directly
+ -- represented value, or a table entry in the proper canonical format is
+ -- created and returned.
+ --
+ -- Note that Init_Operand puts a signed value in the result vector, but
+ -- Vector_To_Uint is always presented with a non-negative value. The
+ -- processing of signs is something that is done by the caller before
+ -- calling Vector_To_Uint.
+
---------------------
-- Output Routines --
---------------------
@@ -494,18 +522,6 @@ private
-- UI_Vector is defined for this purpose and some internal subprograms
-- used for converting from one to the other are defined.
- type UI_Vector is array (Pos range <>) of Int;
- -- Vector containing the integer values of a Uint value
-
- -- Note: An earlier version of this package used pointers of arrays of Ints
- -- (dynamically allocated) for the Uint type. The change leads to a few
- -- less natural idioms used throughout this code, but eliminates all uses
- -- of the heap except for the table package itself. For example, Uint
- -- parameters are often converted to UI_Vectors for internal manipulation.
- -- This is done by creating the local UI_Vector using the function N_Digits
- -- on the Uint to find the size needed for the vector, and then calling
- -- Init_Operand to copy the values out of the table into the vector.
-
type Uint_Entry is record
Length : Pos;
-- Length of entry in Udigits table in digits (i.e. in words)
Index: uintp.adb
===================================================================
--- uintp.adb (revision 216116)
+++ uintp.adb (working copy)
@@ -171,22 +171,6 @@ package body Uintp is
-- If Discard_Quotient is True, Quotient is set to No_Uint
-- If Discard_Remainder is True, Remainder is set to No_Uint
- function Vector_To_Uint
- (In_Vec : UI_Vector;
- Negative : Boolean) return Uint;
- -- Functions that calculate values in UI_Vectors, call this function to
- -- create and return the Uint value. In_Vec contains the multiple precision
- -- (Base) representation of a non-negative value. Leading zeroes are
- -- permitted. Negative is set if the desired result is the negative of the
- -- given value. The result will be either the appropriate directly
- -- represented value, or a table entry in the proper canonical format is
- -- created and returned.
- --
- -- Note that Init_Operand puts a signed value in the result vector, but
- -- Vector_To_Uint is always presented with a non-negative value. The
- -- processing of signs is something that is done by the caller before
- -- calling Vector_To_Uint.
-
------------
-- Direct --
------------
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2014-10-13 8:22 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-10-13 8:34 [Ada] Fix PR ada/63225 Eric Botcazou
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).