[Bug target/12081] Gcc can't be compiled with -mregparm=3

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[stefan.kristiansson at saunalahti dot fi]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

Stefan Kristiansson <stefan.kristiansson at saunalahti dot fi> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |stefan.kristiansson@saunala
                   |                            |hti.fi

--- Comment #19 from Stefan Kristiansson <stefan.kristiansson at saunalahti dot fi> ---
Created attachment 30186
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=30186&action=edit
Updated version of Rask Ingemann Lambertsen's patch

I'm bumping this bug to make people aware that it still exists and that people
are affected by it.
On OpenRISC (not in main tree arch), the ABI states that var args are always
passed on the stack, thus making it impossible to build a working native
compiler
when this bug is present.

Attached is an updated version of the patch posted by Rask Ingemann Lambertsen,
bringing it up to date with the current svn.

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[mikpe at it dot uu.se]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

Mikael Pettersson <mikpe at it dot uu.se> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mikpe at it dot uu.se

--- Comment #20 from Mikael Pettersson <mikpe at it dot uu.se> ---
Created attachment 30188
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=30188&action=edit
fix up conversion in tilepro_expand_builtin, delete unused 11-arity stuff

Looking through the patch I noticed that the conversion in
tilepro_expand_builtin was broken (icode was lost).  Grepping around I also
noticed that nothing used GEN_FCN11 (or is that used by the out-of-tree
OpenRISC port?)  This add-on fixes those two issues.

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[mikpe at it dot uu.se]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

--- Comment #21 from Mikael Pettersson <mikpe at it dot uu.se> ---
(In reply to Mikael Pettersson from comment #20)
> Grepping around I also
> noticed that nothing used GEN_FCN11 (or is that used by the out-of-tree
> OpenRISC port?)  This add-on fixes those two issues.

i386/sync.md appears to generate code that needs the 11-arity version; consider
that part of the fixup patch revoked.

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[mikpe at it dot uu.se]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

--- Comment #22 from Mikael Pettersson <mikpe at it dot uu.se> ---
FWIW, the updated patch for gcc 4.9 bootstraps and regtests cleanly on several
hosts (x86_64, sparc64, powerpc64, armv5tel, m68k).

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[ubizjak at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

--- Comment #23 from Uroš Bizjak <ubizjak at gmail dot com> ---
(In reply to Mikael Pettersson from comment #22)
> FWIW, the updated patch for gcc 4.9 bootstraps and regtests cleanly on
> several hosts (x86_64, sparc64, powerpc64, armv5tel, m68k).

Please post the patch to gcc-patches@ mailing list for a review.
>From gcc-bugs-return-423873-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Jun 05 18:41:43 2013
Return-Path: <gcc-bugs-return-423873-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 22688 invoked by alias); 5 Jun 2013 18:41:43 -0000
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
Delivered-To: mailing list gcc-bugs@gcc.gnu.org
Received: (qmail 22666 invoked by uid 48); 5 Jun 2013 18:41:40 -0000
From: "paolo.carlini at oracle dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c++/57472] internal compiler error: in finish_member_declaration, at cp/semantics.c
Date: Wed, 05 Jun 2013 18:41:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: c++
X-Bugzilla-Version: 4.8.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: paolo.carlini at oracle dot com
X-Bugzilla-Status: NEW
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: bug_status cf_reconfirmed_on everconfirmed
Message-ID: <bug-57472-4-jnSzjeNynr@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-57472-4@http.gcc.gnu.org/bugzilla/>
References: <bug-57472-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
X-SW-Source: 2013-06/txt/msg00252.txt.bz2
Content-length: 506

http://gcc.gnu.org/bugzilla/show_bug.cgi?idW472

Paolo Carlini <paolo.carlini at oracle dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2013-06-05
     Ever confirmed|0                           |1

--- Comment #1 from Paolo Carlini <paolo.carlini at oracle dot com> ---
Confirmed on x86_64-linux.

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[stefan.kristiansson at saunalahti dot fi]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

--- Comment #24 from Stefan Kristiansson <stefan.kristiansson at saunalahti dot fi> ---
(In reply to Uroš Bizjak from comment #23)
> (In reply to Mikael Pettersson from comment #22)
> > FWIW, the updated patch for gcc 4.9 bootstraps and regtests cleanly on
> > several hosts (x86_64, sparc64, powerpc64, armv5tel, m68k).
> 
> Please post the patch to gcc-patches@ mailing list for a review.

Done: http://gcc.gnu.org/ml/gcc-patches/2013-07/msg00400.html
>From gcc-bugs-return-426067-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Jul 10 02:13:58 2013
Return-Path: <gcc-bugs-return-426067-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 26162 invoked by alias); 10 Jul 2013 02:13:58 -0000
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
Delivered-To: mailing list gcc-bugs@gcc.gnu.org
Received: (qmail 26125 invoked by uid 48); 10 Jul 2013 02:13:55 -0000
From: "hp at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug regression/57766] [4.9 Regression]: gcc.sum c-c++-common/cilk-plus/AN/sec_implicit_ex.c
Date: Wed, 10 Jul 2013 02:13:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: regression
X-Bugzilla-Version: unknown
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: hp at gcc dot gnu.org
X-Bugzilla-Status: RESOLVED
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: bug_status resolution
Message-ID: <bug-57766-4-9w1ZLrONRC@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-57766-4@http.gcc.gnu.org/bugzilla/>
References: <bug-57766-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
X-SW-Source: 2013-07/txt/msg00574.txt.bz2
Content-length: 588

http://gcc.gnu.org/bugzilla/show_bug.cgi?idW766

Hans-Peter Nilsson <hp at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|---                         |FIXED

--- Comment #3 from Hans-Peter Nilsson <hp at gcc dot gnu.org> ---
(In reply to Balaji V. Iyer from comment #2)
> This issue should be fixed in SVN revision 200581.
>
> Thanks,
>
> Balaji V. Iyer.

Confirmed; fixed at 200581.  Thanks.

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[olegendo at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

Oleg Endo <olegendo at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |olegendo at gcc dot gnu.org

--- Comment #25 from Oleg Endo <olegendo at gcc dot gnu.org> ---
I've proposed a simpler patch which doesn't require fixing up backend code etc.

http://gcc.gnu.org/ml/gcc-patches/2013-07/msg01315.html

[Bug other/12081] Gcc can't be compiled with -mregparm=3

[olegendo at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

Oleg Endo <olegendo at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|target                      |other

--- Comment #26 from Oleg Endo <olegendo at gcc dot gnu.org> ---
I don't think this is a 'target' issue, changed to 'other'.

I've just committed the following, which should fix the issue for 4.9.

http://gcc.gnu.org/viewcvs/gcc?view=revision&revision=201513

    PR other/12081
    * recog.h (rtx (*insn_gen_fn) (rtx, ...)): Replace typedef with    new
    class insn_gen_fn.
    * expr.c (move_by_pieces_1, store_by_pieces_2): Replace argument
    rtx (*) (rtx, ...) with insn_gen_fn.
    * genoutput.c (output_insn_data): Cast gen_? function pointers to
    insn_gen_fn::stored_funcptr.  Add initializer braces.


Since 4.8 also uses C++ it could be backported.

[Bug other/12081] Gcc can't be compiled with -mregparm=3

[olegendo at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

--- Comment #31 from Oleg Endo <olegendo at gcc dot gnu.org> ---
Author: olegendo
Date: Thu Aug 29 18:37:46 2013
New Revision: 202083

URL: http://gcc.gnu.org/viewcvs?rev=202083&root=gcc&view=rev
Log:
    Backport from mainline
    2013-08-05  Oleg Endo  <olegendo@gcc.gnu.org>

    PR other/12081
    * recog.h (rtx (*insn_gen_fn) (rtx, ...)): Replace typedef with    new
    class insn_gen_fn.
    * expr.c (move_by_pieces_1, store_by_pieces_2): Replace argument
    rtx (*) (rtx, ...) with insn_gen_fn.
    * genoutput.c (output_insn_data): Cast gen_? function pointers to
    insn_gen_fn::stored_funcptr.  Add initializer braces.

    Backport from mainline
    2013-08-07  Oleg Endo  <olegendo@gcc.gnu.org>

    PR other/12081
    * config/rs6000/rs6000.c (gen_2arg_fn_t): Remove typedef.
    (rs6000_emit_swdiv_high_precision, rs6000_emit_swdiv_low_precision,
    rs6000_emit_swrsqrt): Don't cast result of GEN_FCN to gen_2arg_fn_t.


Modified:
    branches/gcc-4_8-branch/gcc/ChangeLog
    branches/gcc-4_8-branch/gcc/config/rs6000/rs6000.c
    branches/gcc-4_8-branch/gcc/expr.c
    branches/gcc-4_8-branch/gcc/genoutput.c
    branches/gcc-4_8-branch/gcc/recog.h

[Bug other/12081] Gcc can't be compiled with -mregparm=3

[olegendo at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

Oleg Endo <olegendo at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|---                         |FIXED

--- Comment #32 from Oleg Endo <olegendo at gcc dot gnu.org> ---
Fixed on 4.8 and 4.9.  Won't backport to 4.7 since it doesn't use C++.

[Bug other/12081] Gcc can't be compiled with -mregparm=3

[oyvind.harboe at zylin dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

oyvind.harboe at zylin dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |oyvind.harboe at zylin dot com

--- Comment #33 from oyvind.harboe at zylin dot com ---
This may be of interest historically: The SPEC2006 403.gcc benchmark has this
bug. 403.gcc is based on gcc 3.2.

https://www.spec.org/cpu2006/Docs/403.gcc.html

[Bug other/12081] Gcc can't be compiled with -mregparm=3

[oyvind.harboe at zylin dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

--- Comment #34 from oyvind.harboe at zylin dot com ---
SPEC2017 502.gcc_r is based on gcc 4.5.0, which also has this bug.

[Bug other/12081] Gcc can't be compiled with -mregparm=3

[oyvind.harboe at zylin dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

--- Comment #35 from oyvind.harboe at zylin dot com ---
SPEC 2017 added SPEC_GCC_VARIADIC_FUNCTIONS_MISMATCH_WORKAROUND to cope with
this error.

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[pinskia at gmail dot com]

------- Comment #18 from pinskia at gmail dot com  2007-10-30 00:20 -------
Subject: Re:  Gcc can't be compiled with -mregparm=3

On 30 Oct 2007 00:14:21 -0000, rask at gcc dot gnu dot org
<gcc-bugzilla@gcc.gnu.org> wrote:
>
>
> ------- Comment #17 from rask at gcc dot gnu dot org  2007-10-30 00:14 -------
> Created an attachment (id=14438)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14438&action=view)
>  --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14438&action=view)
> patch v3, varargs free

This version is much better.

Thanks,
Andrew Pinski


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

Re: [Bug target/12081] Gcc can't be compiled with -mregparm=3

[Andrew Pinski]

On 30 Oct 2007 00:14:21 -0000, rask at gcc dot gnu dot org
<gcc-bugzilla@gcc.gnu.org> wrote:
>
>
> ------- Comment #17 from rask at gcc dot gnu dot org  2007-10-30 00:14 -------
> Created an attachment (id=14438)
>  --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14438&action=view)
> patch v3, varargs free

This version is much better.

Thanks,
Andrew Pinski

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[rask at gcc dot gnu dot org]

------- Comment #17 from rask at gcc dot gnu dot org  2007-10-30 00:14 -------
Created an attachment (id=14438)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14438&action=view)
patch v3, varargs free

In reply to comment #16:
> You can cast them at the time of calling and store them as void * in the
> table --- that is standard-compliant.

Casting is what caused the problem to go unnoticed in the first place. This
version will cause complaints from the compiler if there is a mixup.


-- 

rask at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #14419|0                           |1
        is obsolete|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[mikulas at artax dot karlin dot mff dot cuni dot cz]

------- Comment #16 from mikulas at artax dot karlin dot mff dot cuni dot cz  2007-10-28 16:09 -------
Subject: Re:  Gcc can't be compiled with -mregparm=3

> arguments the function receives. We have gen_* functions taking 0, 1, 2, 3, ...
> arguments and with GCC being designed the way it is, they need to be
> prototyped and defined with the same arguments.
> 
> You are most definitely welcome to post some non-varargs code which works.
> You are most definitely also welcome to time it against my patches.

You can cast them at the time of calling and store them as void * in the 
table --- that is standard-compliant.

Or you can define
union gen_function {
        rtx (*one_arg)(rtx);
        rtx (*two_args)(rtx, rtx);
        rtx (*three_args)(rtx, rtx, rtx);
        ... etc;
};

--- this will be correct C without the performance impact of varargs.

Mikulas


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[rask at gcc dot gnu dot org]

------- Comment #15 from rask at gcc dot gnu dot org  2007-10-28 15:54 -------
In reply to comment #14:
>  is your patch supposed to help with testcase presented in comment #6?

No, it's aimed at the problem from the description. That is, GCC itself doesn't
work if compiled with a compiler where f(int a, int b) is called differently
than f(int a, ...). I don't think your testcase is actually supposed to work.
Don't you get a warning if you take out (void *) and compile with -Wall?
As pinpointed in comment #2, GCC contains the same incorrect type cast. I
actually forgot to patch this part:

Index: gcc/genoutput.c
===================================================================
--- gcc/genoutput.c     (revision 129503)
+++ gcc/genoutput.c     (working copy)
@@ -386,7 +386,7 @@ output_insn_data (void)
        }

       if (d->name && d->name[0] != '*')
-       printf ("    (insn_gen_fn) gen_%s,\n", d->name);
+       printf ("    gen_%s,\n", d->name);
       else
        printf ("    0,\n");



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[sam at rfc1149 dot net]

------- Comment #14 from sam at rfc1149 dot net  2007-10-28 14:42 -------
Rask,

is your patch supposed to help with testcase presented in comment #6? It looks
like I still get "f" waiting (as expected) for arguments in registers when I
use -mregparm=3 while the call through g still uses the stack:

main:
[...]
        subl    $8, %esp
        movl    $2, 4(%esp)
        movl    $1, (%esp)
        call    f

(GCC at SVN/trunk 129695 + your patch)


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[rask at gcc dot gnu dot org]

------- Comment #13 from rask at gcc dot gnu dot org  2007-10-28 14:02 -------
In reply to comment #7:

> I need it to build GCC with OpenWatcom, which wants parameters on the stack by default.

Er, that's in registers by default, of course.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[rask at gcc dot gnu dot org]

------- Comment #12 from rask at gcc dot gnu dot org  2007-10-28 14:00 -------
In reply to comment #11:
> How many times do I have to say this is bad for most RISC targets (hosts)?

I don't particularily care how many times you say it. Show some code (which
works) and/or show some timings (of code that works).

> Really if the type being is used is wrong, they should be changed
> rather than changing to use var-args.

Will you please look at the code? You have gen_movsi() with 2 arguments and
gen_addsi3() with 3 arguments. Given that they are put into a table as a
function of type insn_gen_fn and called as such through the table (see the
GEN_FCN() macro in optabs.h and the table definition in recog.h), the
function definition has to match. If it doesn't, all bets are off as to what
arguments the function receives. We have gen_* functions taking 0, 1, 2, 3, ...
arguments and with GCC being designed the way it is, they need to be
prototyped and defined with the same arguments.

You are most definitely welcome to post some non-varargs code which works.
You are most definitely also welcome to time it against my patches.


-- 

rask at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to fail|                            |4.3.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[pinskia at gmail dot com]

------- Comment #11 from pinskia at gmail dot com  2007-10-28 11:57 -------
Subject: Re:  Gcc can't be compiled with -mregparm=3

On 28 Oct 2007 11:14:37 -0000, rask at gcc dot gnu dot org
<gcc-bugzilla@gcc.gnu.org> wrote:

How many times do I have to say this is bad for most RISC targets (hosts)?

Really if the type being is used is wrong, they should be changed
rather than changing to use var-args.

-- Pinski


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

Re: [Bug target/12081] Gcc can't be compiled with -mregparm=3

[Andrew Pinski]

On 28 Oct 2007 11:14:37 -0000, rask at gcc dot gnu dot org
<gcc-bugzilla@gcc.gnu.org> wrote:

How many times do I have to say this is bad for most RISC targets (hosts)?

Really if the type being is used is wrong, they should be changed
rather than changing to use var-args.

-- Pinski

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[rask at gcc dot gnu dot org]

------- Comment #10 from rask at gcc dot gnu dot org  2007-10-28 11:14 -------
Created an attachment (id=14419)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14419&action=view)
patch v2, i386 fix added


-- 

rask at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #14417|0                           |1
        is obsolete|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[rask at gcc dot gnu dot org]

------- Comment #9 from rask at gcc dot gnu dot org  2007-10-27 23:36 -------
It happens to work because all the compilers people use to build GCC pass
varargs the same way as non-varargs, at least for the number of arguments
received by the gen_* functions. IOW you shouldn't see any speed difference if
it works for you already.


-- 

rask at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to fail|4.3.0                       |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

Re: [Bug target/12081] Gcc can't be compiled with -mregparm=3

[Andrew Pinski]

On 27 Oct 2007 23:16:57 -0000, rask at gcc dot gnu dot org
<gcc-bugzilla@gcc.gnu.org> wrote:
> Please give this patch a try. I need it to build GCC with OpenWatcom, which
> wants parameters on the stack by default.

I think this makes some worse code in some cases.  Var-args is very
bad for code generation on targets where argments are passed via
registers (almost all RISC and IA64).

-- Pinski

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[pinskia at gmail dot com]

------- Comment #8 from pinskia at gmail dot com  2007-10-27 23:22 -------
Subject: Re:  Gcc can't be compiled with -mregparm=3

On 27 Oct 2007 23:16:57 -0000, rask at gcc dot gnu dot org
<gcc-bugzilla@gcc.gnu.org> wrote:
> Please give this patch a try. I need it to build GCC with OpenWatcom, which
> wants parameters on the stack by default.

I think this makes some worse code in some cases.  Var-args is very
bad for code generation on targets where argments are passed via
registers (almost all RISC and IA64).

-- Pinski


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[rask at gcc dot gnu dot org]

------- Comment #7 from rask at gcc dot gnu dot org  2007-10-27 23:16 -------
Created an attachment (id=14417)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14417&action=view)
possible patch

Please give this patch a try. I need it to build GCC with OpenWatcom, which
wants parameters on the stack by default.


-- 

rask at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |rask at gcc dot gnu dot org
                   |dot org                     |
             Status|NEW                         |ASSIGNED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[sam at rfc1149 dot net]

------- Comment #6 from sam at rfc1149 dot net  2007-10-27 18:32 -------
Here is a small test case for this program. Compile with -mregparm=3 (or any
value greater than 0 on x86).

#include <stdio.h>

typedef void (*func) (int a, ...);

void f (int a, int b)
{
   printf ("a = %d, b = %d\n", a, b);
}

int main ()
{
   func g = (void *) f;
   g (1, 2);
   return 0;
}

The execution on my machine prints:

a = 134513524, b = -1078190752

instead of the expected

a = 1, b = 2

obtained when using -mregparm=0.

The call to "f" (through variable "g") is generated as-is:

        movl    $2, 4(%esp)
        movl    $1, (%esp)
        call    f

instead of using registers.


-- 

sam at rfc1149 dot net changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |sam at rfc1149 dot net


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-10-28 12:09 -------
*** Bug 18193 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |rpedersen at atmel dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[dhazeghi at yahoo dot com]

------- Additional Comments From dhazeghi at yahoo dot com  2004-01-18 01:54 -------
I don't necessarily disagree. However without a testcase, the chances of somebody looking into/
fixing this aren't too good (I certainly don't have the expertise).

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[mikulas at artax dot karlin dot mff dot cuni dot cz]

------- Additional Comments From mikulas at artax dot karlin dot mff dot cuni dot cz  2004-01-18 01:42 -------
In example I wrote there should be
g(1,2)
instead of
f(1,2)

f(1,2) is correct, but g(1,2) will crash with -mrtd or pass invalid arguments
with -mregparm=3.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[mikulas at artax dot karlin dot mff dot cuni dot cz]

------- Additional Comments From mikulas at artax dot karlin dot mff dot cuni dot cz  2004-01-18 01:39 -------
Gcc is code written the wrong way

Let's have these declarations:

void f(int a, int b);
void (*g)(int a, ...);

and this code:

g = (void (*)(int, ...))f;
f(1,2);

--- the code is obviously wrong in C and might not work. It works on most
architectures, but it doesn't on x86 with regparm 3 or -mrtd.

Similar code exists in gcc in insn-output.c (there is:
"(insn_gen_fn)gen_cmpqi_ext_3_insn" and many more like this).

I "fixed" it for myself by adding attribute regparm(0),cdecl to these
functions, but it is not correct fix. Correct fix would be to not use these
kind of casts.

I can't show you how to reproduce the bug, because you don't have environment
where regparm(3) is default calling convention. I do have.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |NEW


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[dhazeghi at yahoo dot com]

------- Additional Comments From dhazeghi at yahoo dot com  2004-01-17 23:53 -------
Sorry that this report seems to have slipped through the cracks. What exactly do you mean by 
"doesn't work"? Does it produce buggy code, or just crash, or what? Also, I don't suppose you 
could come up with a small, self-contained testcase? Thanks.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081

[Bug target/12081] Gcc can't be compiled with -mregparm=3

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|other                       |target


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12081