Re: ada/6483: ada fails to build under hppa*-hp-hpux*

Re: ada/6483: ada fails to build under hppa*-hp-hpux*

[danglin]

Synopsis: ada fails to build under hppa*-hp-hpux*

Responsible-Changed-From-To: unassigned->danglin
Responsible-Changed-By: danglin
Responsible-Changed-When: Fri Apr 26 16:50:25 2002
Responsible-Changed-Why:
    Mine.  Regression.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=6483

Re: ada/6483: ada fails to build under hppa*-hp-hpux*

[zack]

The following reply was made to PR ada/6483; it has been noted by GNATS.

From: "zack@codesourcery.com" <zack@codesourcery.com>
To: Mark Mitchell <mark@codesourcery.com>
Cc: John David Anglin <dave@hiauly1.hia.nrc.ca>,
	"fw@deneb.enyo.de" <fw@deneb.enyo.de>,
	"dave.anglin@nrc.ca" <dave.anglin@nrc.ca>,
	"gcc-gnats@gcc.gnu.org" <gcc-gnats@gcc.gnu.org>
Subject: Re: ada/6483: ada fails to build under hppa*-hp-hpux*
Date: Tue, 30 Apr 2002 14:46:57 -0700

 On Tue, Apr 30, 2002 at 11:15:14AM -0700, Mark Mitchell wrote:
 > 
 > >As far as I understand the situation, it only needs to be applied
 > >to the c and ada files in the ada subdirectory.  We don't need
 > >to build all of gcc with -mdisable-indexing.
 > 
 > Thanks to both of you for explaining this situation.
 > 
 > >This is what I suggested
 > ><http://gcc.gnu.org/ml/gcc-patches/2002-04/msg01625.html>.
 > 
 > OK.  Check that version in, and let's close the PR.
 
 I'd suggest renaming ADA_CFLAGS to X_ADA_CFLAGS so it is clear that
 this is for use in x-fragments.  Probably only worth doing on the
 mainline.
 
 zw

Re: ada/6483: ada fails to build under hppa*-hp-hpux*

[John David Anglin]

The following reply was made to PR ada/6483; it has been noted by GNATS.

From: "John David Anglin" <dave@hiauly1.hia.nrc.ca>
To: zack@codesourcery.com
Cc: mark@codesourcery.com, fw@deneb.enyo.de, dave.anglin@nrc.ca,
   gcc-gnats@gcc.gnu.org
Subject: Re: ada/6483: ada fails to build under hppa*-hp-hpux*
Date: Tue, 30 Apr 2002 17:53:15 -0400 (EDT)

 > I'd suggest renaming ADA_CFLAGS to X_ADA_CFLAGS so it is clear that
 > this is for use in x-fragments.  Probably only worth doing on the
 > mainline.
 
 That seems reasonable.
 
 Dave
 -- 
 J. David Anglin                                  dave.anglin@nrc.ca
 National Research Council of Canada              (613) 990-0752 (FAX: 952-6605)

Re: ada/6483: ada fails to build under hppa*-hp-hpux*

[Mark Mitchell]

The following reply was made to PR ada/6483; it has been noted by GNATS.

From: Mark Mitchell <mark@codesourcery.com>
To: "zack@codesourcery.com" <zack@codesourcery.com>
Cc: John David Anglin <dave@hiauly1.hia.nrc.ca>,
   "fw@deneb.enyo.de" <fw@deneb.enyo.de>,
   "dave.anglin@nrc.ca" <dave.anglin@nrc.ca>,
   "gcc-gnats@gcc.gnu.org" <gcc-gnats@gcc.gnu.org>
Subject: Re: ada/6483: ada fails to build under hppa*-hp-hpux*
Date: Tue, 30 Apr 2002 14:49:26 -0700

 > I'd suggest renaming ADA_CFLAGS to X_ADA_CFLAGS so it is clear that
 > this is for use in x-fragments.  Probably only worth doing on the
 > mainline.
 
 Make sense, indeed.
 
 Dave, would you do this please (on the mainline, as Zack says)?
 
 Preapproved, and with apologies to Zack and Dave: Zack, because he pointed
 this out before, and Dave, because he probably already committed the
 change.
 
 Thanks,
 
 --
 Mark Mitchell                   mark@codesourcery.com
 CodeSourcery, LLC               http://www.codesourcery.com
 

Re: ada/6483: ada fails to build under hppa*-hp-hpux*

[danglin]

Synopsis: ada fails to build under hppa*-hp-hpux*

State-Changed-From-To: open->closed
State-Changed-By: danglin
State-Changed-When: Tue Apr 30 12:21:11 2002
State-Changed-Why:
    Fix applied.
    <http://gcc.gnu.org/ml/gcc-patches/2002-04/msg01625.html>.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=6483

Re: ada/6483: ada fails to build under hppa*-hp-hpux*

[Mark Mitchell]

The following reply was made to PR ada/6483; it has been noted by GNATS.

From: Mark Mitchell <mark@codesourcery.com>
To: John David Anglin <dave@hiauly1.hia.nrc.ca>,
   "zack@codesourcery.com" <zack@codesourcery.com>
Cc: "fw@deneb.enyo.de" <fw@deneb.enyo.de>,
   "dave.anglin@nrc.ca" <dave.anglin@nrc.ca>,
   "gcc-gnats@gcc.gnu.org" <gcc-gnats@gcc.gnu.org>
Subject: Re: ada/6483: ada fails to build under hppa*-hp-hpux*
Date: Tue, 30 Apr 2002 11:15:14 -0700

 > As far as I understand the situation, it only needs to be applied
 > to the c and ada files in the ada subdirectory.  We don't need
 > to build all of gcc with -mdisable-indexing.
 
 Thanks to both of you for explaining this situation.
 
 > This is what I suggested
 > <http://gcc.gnu.org/ml/gcc-patches/2002-04/msg01625.html>.
 
 OK.  Check that version in, and let's close the PR.
 
 Thanks,
 
 --
 Mark Mitchell                   mark@codesourcery.com
 CodeSourcery, LLC               http://www.codesourcery.com
 

Re: ada/6483: ada fails to build under hppa*-hp-hpux*

[John David Anglin]

The following reply was made to PR ada/6483; it has been noted by GNATS.

From: "John David Anglin" <dave@hiauly1.hia.nrc.ca>
To: zack@codesourcery.com
Cc: mark@codesourcery.com, fw@deneb.enyo.de, dave.anglin@nrc.ca,
   gcc-gnats@gcc.gnu.org
Subject: Re: ada/6483: ada fails to build under hppa*-hp-hpux*
Date: Tue, 30 Apr 2002 14:13:59 -0400 (EDT)

 > On Tue, Apr 30, 2002 at 10:14:34AM -0700, Mark Mitchell wrote:
 > > 
 > > >2001-10-08  Zack Weinberg  <zack@codesourcery.com>
 > > >
 > > >        * aclocal.m4 (gcc_AC_PROG_GNAT): New.
 > > >	* configure.in: Use it.
 > > >	* configure: Regenerated.
 > > >	* config/pa/t-linux, config/pa/t-pa, config/pa/t-pa64,
 > > >	config/pa/t-pro: Set T_ADAFLAGS, not ADA_CFLAGS.
 > > 
 > > Zack, can you comment on this?
 > 
 > I do not fully understand the issue here.  
 > 
 > My *impression* is that this is a host problem, i.e. that switch
 
 Yes.
 
 > should appear in an x-fragment not a t-fragment.  My impression is
 
 Yes.
 
 > also that -mdisable-indexing needs to get applied to everything
 > compiled into gnat1, not just Ada source code.
 
 As far as I understand the situation, it only needs to be applied
 to the c and ada files in the ada subdirectory.  We don't need
 to build all of gcc with -mdisable-indexing.
 
 > 
 > Therefore, I'd suggest that the right fix is to nuke T_ADAFLAGS and
 > just set -mdisable-indexing with X_CFLAGS in x-pa.  (There isn't an
 > x-pa right now, but we could create one.)
 
 I would prefer using something like X_ADA_CFLAGS so that the option
 only is applied building files in the ada subdirectory.
 
 > 
 > I can provide a patch if you'd like.
 
 This is what I suggested
 <http://gcc.gnu.org/ml/gcc-patches/2002-04/msg01625.html>.
 
 > p.s. The description of the effect of -mdisable-indexing, given by
 > various people on the list, does *not* match what it actually does.
 > What it actually does is prevent GCC from generating [reg+reg] loads.
 > (Well, that and this lovely piece of gobbledygook:
 > 
 >   /* Since move_operand is only used for source operands, we can always
 >      allow scaled indexing!  */
 >   if (! TARGET_DISABLE_INDEXING
 >       && GET_CODE (op) == PLUS
 >       && ((GET_CODE (XEXP (op, 0)) == MULT
 >            && GET_CODE (XEXP (XEXP (op, 0), 0)) == REG
 >            && GET_CODE (XEXP (XEXP (op, 0), 1)) == CONST_INT
 >            && INTVAL (XEXP (XEXP (op, 0), 1))
 >               == (HOST_WIDE_INT) GET_MODE_SIZE (mode)
 >            && GET_CODE (XEXP (op, 1)) == REG)
 >           || (GET_CODE (XEXP (op, 1)) == MULT
 >               &&GET_CODE (XEXP (XEXP (op, 1), 0)) == REG
 >               && GET_CODE (XEXP (XEXP (op, 1), 1)) == CONST_INT
 >               && INTVAL (XEXP (XEXP (op, 1), 1))
 >                  == (HOST_WIDE_INT) GET_MODE_SIZE (mode)
 >               && GET_CODE (XEXP (op, 0)) == REG)))
 >     return 1;
 > 
 > but I think that boils down to the same thing.)
 > 
 > I would think that if [reg+reg] loads are causing problems, there's a
 > bug in the machine description, and -mdisable-indexing is merely
 > papering it over.
 
 This has been discussed.  What it boils down to is ada's use of virtual
 arrays don't play well with the segmented memory runtime used under
 hpux.  Essentially, the gnat implementation assumes that these arrays
 can be accessed using a base register which can be anywhere in the virtual
 address space.
 
 Dave
 -- 
 J. David Anglin                                  dave.anglin@nrc.ca
 National Research Council of Canada              (613) 990-0752 (FAX: 952-6605)

Re: ada/6483: ada fails to build under hppa*-hp-hpux*

[zack]

The following reply was made to PR ada/6483; it has been noted by GNATS.

From: "zack@codesourcery.com" <zack@codesourcery.com>
To: Mark Mitchell <mark@codesourcery.com>
Cc: John David Anglin <dave@hiauly1.hia.nrc.ca>,
	Florian Weimer <fw@deneb.enyo.de>,
	"dave.anglin@nrc.ca" <dave.anglin@nrc.ca>,
	"gcc-gnats@gcc.gnu.org" <gcc-gnats@gcc.gnu.org>
Subject: Re: ada/6483: ada fails to build under hppa*-hp-hpux*
Date: Tue, 30 Apr 2002 10:55:49 -0700

 On Tue, Apr 30, 2002 at 10:14:34AM -0700, Mark Mitchell wrote:
 > 
 > >2001-10-08  Zack Weinberg  <zack@codesourcery.com>
 > >
 > >        * aclocal.m4 (gcc_AC_PROG_GNAT): New.
 > >	* configure.in: Use it.
 > >	* configure: Regenerated.
 > >	* config/pa/t-linux, config/pa/t-pa, config/pa/t-pa64,
 > >	config/pa/t-pro: Set T_ADAFLAGS, not ADA_CFLAGS.
 > 
 > Zack, can you comment on this?
 
 I do not fully understand the issue here.  
 
 My *impression* is that this is a host problem, i.e. that switch
 should appear in an x-fragment not a t-fragment.  My impression is
 also that -mdisable-indexing needs to get applied to everything
 compiled into gnat1, not just Ada source code.
 
 Therefore, I'd suggest that the right fix is to nuke T_ADAFLAGS and
 just set -mdisable-indexing with X_CFLAGS in x-pa.  (There isn't an
 x-pa right now, but we could create one.)
 
 I can provide a patch if you'd like.
 
 zw
 
 p.s. The description of the effect of -mdisable-indexing, given by
 various people on the list, does *not* match what it actually does.
 What it actually does is prevent GCC from generating [reg+reg] loads.
 (Well, that and this lovely piece of gobbledygook:
 
   /* Since move_operand is only used for source operands, we can always
      allow scaled indexing!  */
   if (! TARGET_DISABLE_INDEXING
       && GET_CODE (op) == PLUS
       && ((GET_CODE (XEXP (op, 0)) == MULT
            && GET_CODE (XEXP (XEXP (op, 0), 0)) == REG
            && GET_CODE (XEXP (XEXP (op, 0), 1)) == CONST_INT
            && INTVAL (XEXP (XEXP (op, 0), 1))
               == (HOST_WIDE_INT) GET_MODE_SIZE (mode)
            && GET_CODE (XEXP (op, 1)) == REG)
           || (GET_CODE (XEXP (op, 1)) == MULT
               &&GET_CODE (XEXP (XEXP (op, 1), 0)) == REG
               && GET_CODE (XEXP (XEXP (op, 1), 1)) == CONST_INT
               && INTVAL (XEXP (XEXP (op, 1), 1))
                  == (HOST_WIDE_INT) GET_MODE_SIZE (mode)
               && GET_CODE (XEXP (op, 0)) == REG)))
     return 1;
 
 but I think that boils down to the same thing.)
 
 I would think that if [reg+reg] loads are causing problems, there's a
 bug in the machine description, and -mdisable-indexing is merely
 papering it over.

Re: ada/6483: ada fails to build under hppa*-hp-hpux*

[John David Anglin]

The following reply was made to PR ada/6483; it has been noted by GNATS.

From: "John David Anglin" <dave@hiauly1.hia.nrc.ca>
To: fw@deneb.enyo.de (Florian Weimer)
Cc: dave.anglin@nrc.ca, gcc-gnats@gcc.gnu.org, mark@codesourcery.com,
   zack@codesourcery.com
Subject: Re: ada/6483: ada fails to build under hppa*-hp-hpux*
Date: Tue, 30 Apr 2002 13:11:21 -0400 (EDT)

 > dave.anglin@nrc.ca writes:
 > 
 > > The PR has been marked high-priority because it is a
 > > regression.
 > 
 > I don't think it's a regression, from the FSF point of view.
 
 Then, I don't understand the FSF point of view on what's a regression.
 
 The following patch broke native builds of the ada compiler under hpux:
 
 2001-10-08  Zack Weinberg  <zack@codesourcery.com>
 
         * aclocal.m4 (gcc_AC_PROG_GNAT): New.
 	* configure.in: Use it.
 	* configure: Regenerated.
 	* config/pa/t-linux, config/pa/t-pa, config/pa/t-pa64,
 	config/pa/t-pro: Set T_ADAFLAGS, not ADA_CFLAGS.
 
 This patch was made after the release of 3.0.  If my patch is not
 acceptable, then I request that the above patch be reverted or
 corrected.
 
 This first came to light when Matthew Wilcox did a cross build to
 a hppa-linux target and found T_ADAFLAGS was incorrectly defined.
 In the ensuing discussion, it became clear that "-mdisable-indxing"
 needed to used in host builds for both C and ADA files.  T_ADAFLAGS
 is only used for ADA files.  We had the correct behaviour in native
 builds before Zack's patch using ADA_CFLAGS.  However, this is not
 a proper target or host define, so I can see why Zack applied the
 above patch.
 
 I personally believe the makefile infrastructure should be fixed on
 the 3.1 branch.  Without it being fixed, ada will not build under
 hpux.  hppa2.0w-hp-hpux11.00 is one of the primary evaluation
 targets for the release.
 
 Dave
 -- 
 J. David Anglin                                  dave.anglin@nrc.ca
 National Research Council of Canada              (613) 990-0752 (FAX: 952-6605)

Re: ada/6483: ada fails to build under hppa*-hp-hpux*

[John David Anglin]

The following reply was made to PR ada/6483; it has been noted by GNATS.

From: "John David Anglin" <dave@hiauly1.hia.nrc.ca>
To: mark@codesourcery.com (Mark Mitchell)
Cc: fw@deneb.enyo.de, dave.anglin@nrc.ca, gcc-gnats@gcc.gnu.org,
   zack@codesourcery.com
Subject: Re: ada/6483: ada fails to build under hppa*-hp-hpux*
Date: Tue, 30 Apr 2002 13:32:15 -0400 (EDT)

 > > Then, I don't understand the FSF point of view on what's a regression.
 > 
 > It's not a regression because we never had a release where this worked.
 
 OK, I see.  ADA wasn't integrated into the GCC tree before.  However,
 the makefile infrastructure existed which enabled you to build ada if it
 was inserted into the tree.
 
 Possibly the other high priority ada PRs should not be marked as high
 priority.
 
 In looking at <http://gcc.gnu.org/gcc-3.1/changes.html>, there is no
 mention that ADA is new.
 
 Dave
 -- 
 J. David Anglin                                  dave.anglin@nrc.ca
 National Research Council of Canada              (613) 990-0752 (FAX: 952-6605)

Re: ada/6483: ada fails to build under hppa*-hp-hpux*

[Mark Mitchell]

The following reply was made to PR ada/6483; it has been noted by GNATS.

From: Mark Mitchell <mark@codesourcery.com>
To: John David Anglin <dave@hiauly1.hia.nrc.ca>,
   Florian Weimer <fw@deneb.enyo.de>
Cc: "dave.anglin@nrc.ca" <dave.anglin@nrc.ca>,
   "gcc-gnats@gcc.gnu.org" <gcc-gnats@gcc.gnu.org>,
   "zack@codesourcery.com" <zack@codesourcery.com>
Subject: Re: ada/6483: ada fails to build under hppa*-hp-hpux*
Date: Tue, 30 Apr 2002 10:14:34 -0700

 --On Tuesday, April 30, 2002 01:11:21 PM -0400 John David Anglin 
 <dave@hiauly1.hia.nrc.ca> wrote:
 
 >> dave.anglin@nrc.ca writes:
 >>
 >> > The PR has been marked high-priority because it is a
 >> > regression.
 >>
 >> I don't think it's a regression, from the FSF point of view.
 >
 > Then, I don't understand the FSF point of view on what's a regression.
 
 It's not a regression because we never had a release where this worked.
 
 On the other hand, since it worked at some time, it would be nice if
 it still did.
 
 > 2001-10-08  Zack Weinberg  <zack@codesourcery.com>
 >
 >         * aclocal.m4 (gcc_AC_PROG_GNAT): New.
 > 	* configure.in: Use it.
 > 	* configure: Regenerated.
 > 	* config/pa/t-linux, config/pa/t-pa, config/pa/t-pa64,
 > 	config/pa/t-pro: Set T_ADAFLAGS, not ADA_CFLAGS.
 
 Zack, can you comment on this?
 
 --
 Mark Mitchell                   mark@codesourcery.com
 CodeSourcery, LLC               http://www.codesourcery.com
 

Re: ada/6483: ada fails to build under hppa*-hp-hpux*

[Florian Weimer]

The following reply was made to PR ada/6483; it has been noted by GNATS.

From: Florian Weimer <fw@deneb.enyo.de>
To: dave.anglin@nrc.ca
Cc: gcc-gnats@gcc.gnu.org,  mark@codesourcery.com
Subject: Re: ada/6483: ada fails to build under hppa*-hp-hpux*
Date: Tue, 30 Apr 2002 17:27:04 +0200

 dave.anglin@nrc.ca writes:
 
 > The PR has been marked high-priority because it is a
 > regression.
 
 I don't think it's a regression, from the FSF point of view.

ada/6483: ada fails to build under hppa*-hp-hpux*

[dave.anglin]

>Number:         6483
>Category:       ada
>Synopsis:       ada fails to build under hppa*-hp-hpux*
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Apr 26 16:46:00 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator:     Dave Anglin
>Release:        gcc 3.1 and 3.2
>Organization:
>Environment:
hppa*-hp-hpux*
>Description:
Segmentation fault in gnat compiling ada.ads in stage 2.
The problem is caused by not specify "-mdisable-indexing"
when building the C files comprising the ada compiling.

The build configuration was changed last October by
this patch:

2001-10-08  Zack Weinberg  <zack@codesourcery.com>

        * aclocal.m4 (gcc_AC_PROG_GNAT): New.
        * configure.in: Use it.
        * configure: Regenerated.
        * config/pa/t-linux, config/pa/t-pa, config/pa/t-pa64,
        config/pa/t-pro: Set T_ADAFLAGS, not ADA_CFLAGS.

ADA_CFLAGS is used in building both c and ada files in the
ada compiler.  T_ADAFLAGS is only used when compiling ada
files.  The above option needs to be specified for both
c and ada files for correct operation of gnat.

A further problem is that T_ADAFLAGS is a target configuration
flag.  The "-mdisable-indexing" option needs to be specified
for the host not the target.  This was observed recently
when a cross compiler was used to port ada to hppa-linux.

The PR has been marked high-priority because it is a
regression.
>How-To-Repeat:

>Fix:
<http://gcc.gnu.org/ml/gcc-patches/2002-04/msg01625.html>
>Release-Note:
>Audit-Trail:
>Unformatted: