[Bug bootstrap/15627] New: Sun CC cannot bootstrap GCC

[Bug bootstrap/15627] New: Sun CC cannot bootstrap GCC

[ebotcazou at gcc dot gnu dot org]

I get with Sun CC 5.0, 5.3 and 5.5:

cc -erroff -DHAVE_CONFIG_H -I. -I/home/eric/cvs/gcc/libbanshee/engine -I..   
-I/home/eric/cvs/gcc/libbanshee/engine/../libcompat
-I/home/eric/cvs/gcc/libbanshee/../include
-I/home/eric/cvs/gcc/libbanshee/engine/../include -I. -Ddeletes= -Dtraditional=
-Dsameregion= -Dparentptr=  -g  -c `test -f
'/home/eric/cvs/gcc/libbanshee/engine/util.c' || echo
'/home/eric/cvs/gcc/libbanshee/engine/'`/home/eric/cvs/gcc/libbanshee/engine/util.c
"/home/eric/cvs/gcc/libbanshee/engine/util.c", line 199: identifier redefined: min
        current : function(int, int) returning int
        previous: function(int, int) returning int :
"/home/eric/cvs/gcc/libbanshee/engine/util.h", line 104
"/home/eric/cvs/gcc/libbanshee/engine/util.c", line 200: identifier redefined: max
        current : function(int, int) returning int
        previous: function(int, int) returning int :
"/home/eric/cvs/gcc/libbanshee/engine/util.h", line 105
cc: acomp failed for /home/eric/cvs/gcc/libbanshee/engine/util.c
gmake[3]: *** [util.o] Error 2

-- 
           Summary: Sun CC cannot bootstrap GCC
           Product: gcc
           Version: 3.5.0
            Status: UNCONFIRMED
          Severity: critical
          Priority: P2
         Component: bootstrap
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: ebotcazou at gcc dot gnu dot org
                CC: gcc-bugs at gcc dot gnu dot org
 GCC build triplet: *-*-solaris2.*
  GCC host triplet: *-*-solaris2.*
GCC target triplet: *-*-solaris2.*


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug other/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-05-24 11:12 -------
Daniel,
This is because of the functions being declared extern inline in util.h.
Please fix this.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dberlin at gcc dot gnu dot
                   |                            |org
           Severity|critical                    |normal
             Status|UNCONFIRMED                 |NEW
          Component|bootstrap                   |other
     Ever Confirmed|                            |1
           Keywords|                            |build
   Last reconfirmed|0000-00-00 00:00:00         |2004-05-24 11:12:54
               date|                            |
            Summary|Sun CC cannot bootstrap GCC |[3.5 Regression] Sun CC
                   |                            |cannot bootstrap GCC
   Target Milestone|---                         |3.5.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug other/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-05-24 13:21 -------
Subject: Bug 15627

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	dberlin@gcc.gnu.org	2004-05-24 13:20:53

Modified files:
	libbanshee     : ChangeLog 
	libbanshee/engine: util.c util.h 

Log message:
	2004-05-24  Daniel Berlin  <dberlin@dberlin.org>
	
	PR other/15627
	* engine/util.h: Include ansidecl.h.
	Make max and min static inline.
	* engine/util.c: Remove max and min from here.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libbanshee/ChangeLog.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libbanshee/engine/util.c.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libbanshee/engine/util.h.diff?cvsroot=gcc&r1=1.2&r2=1.3



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug other/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[dberlin at gcc dot gnu dot org]

------- Additional Comments From dberlin at gcc dot gnu dot org  2004-05-24 13:21 -------
Fixed

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug other/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[ebotcazou at gcc dot gnu dot org]

------- Additional Comments From ebotcazou at gcc dot gnu dot org  2004-05-30 07:49 -------
Unfortunately, Sun CC still can't bootstrap the compiler:

cc -erroff -DHAVE_CONFIG_H -I. -I/home/eric/cvs/gcc/libbanshee/engine -I..   
-I/home/eric/cvs/gcc/libbanshee/engine/../libcompat
-I/home/eric/cvs/gcc/libbanshee/../include
-I/home/eric/cvs/gcc/libbanshee/engine/../include -I. -Ddeletes= -Dtraditional=
-Dsameregion= -Dparentptr=  -g  -c `test -f
'/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c' || echo
'/home/eric/cvs/gcc/libbanshee/engine/'`/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 199: syntax error
before or at: {
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 202: undefined symbol: v
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 234: undefined symbol: r
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 234: an address is not
allowed in a constant initializer for an integral type whose size is smaller
than the size of a pointer
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 235: identifier
redeclared: search_ubs_aux
        current : function() returning int
        previous: function(pointer to struct setif_var {}) returning void :
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 198
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 237: syntax error
before or at: return
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 246: syntax error
before or at: {
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 248: undefined symbol: v
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 280: identifier
redefined: found
        current : int
        previous: int : "/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c",
line 233
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 281: identifier
redefined: cycle
        current : int
        previous: int : "/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c",
line 234
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 281: an address is not
allowed in a constant initializer for an integral type whose size is smaller
than the size of a pointer
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 282: identifier
redeclared: search_lbs_aux
        current : function() returning int
        previous: function(pointer to struct setif_var {}) returning void :
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 245
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 284: syntax error
before or at: return
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 318: syntax error
before or at: {
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 330: undefined symbol:
scan_cycle
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 331: undefined symbol: var
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 333: undefined symbol:
lowest
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 334: undefined symbol:
scan_bounds
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 335: undefined symbol: lb
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 336: undefined symbol: b
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 339: undefined symbol:
witness
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 340: undefined symbol:
lowest
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 342: undefined symbol: b
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 342: undefined symbol:
scan_bounds
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 343: undefined symbol: lb
"/home/eric/cvs/gcc/libbanshee/engine/setif-sort.c", line 351: cannot recover
from previous errors
cc: acomp failed for /home/eric/cvs/gcc/libbanshee/engine/setif-sort.c
gmake[3]: *** [setif-sort.o] Error 2
gmake[3]: Leaving directory `/opt/build/eric/gcc-libbanshee/libbanshee/engine'
gmake[2]: *** [all-recursive] Error 1
gmake[2]: Leaving directory `/opt/build/eric/gcc-libbanshee/libbanshee'
gmake[1]: *** [all] Error 2
gmake[1]: Leaving directory `/opt/build/eric/gcc-libbanshee/libbanshee'
gmake: *** [all-libbanshee] Error 2


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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug other/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-05-30 12:58 -------
This is caused by nested functions.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |dberlin at gcc dot gnu dot
                   |dot org                     |org
             Status|REOPENED                    |ASSIGNED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug bootstrap/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[giovannibajo at libero dot it]

------- Additional Comments From giovannibajo at libero dot it  2004-06-27 01:35 -------
Eric, do we need also to get rid of VLAs in libbanshee? And what about C++-
style comments?

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|other                       |bootstrap


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug bootstrap/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[giovannibajo at libero dot it]

------- Additional Comments From giovannibajo at libero dot it  2004-06-27 02:45 -------
A first patch is here:
http://gcc.gnu.org/ml/gcc-patches/2004-06/msg02227.html

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |patch


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug bootstrap/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[ebotcazou at gcc dot gnu dot org]

------- Additional Comments From ebotcazou at gcc dot gnu dot org  2004-06-27 06:30 -------
> Eric, do we need also to get rid of VLAs in libbanshee? And what about C++-
> style comments?

I think libbanshee must be compilable by an ISO C90 compiler, like the compiler.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |giovannibajo at libero dot
                   |                            |it


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug bootstrap/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-06-27 11:21 -------
Subject: Bug 15627

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	giovannibajo@gcc.gnu.org	2004-06-27 11:21:14

Modified files:
	libbanshee     : ChangeLog 
	libbanshee/engine: flowrow-sort.c setif-sort.c 

Log message:
	PR bootstrap/15627
	* engine/flowrow-sort.c (update_upper_bound): Converted comment to
	C-style.
	(flowrow_inclusion): Likewise.
	(flowrow_extract_field): Unnest field_eq.
	* engine/setif-sort.c (search_ubs): Unnest search_ubs_aux.
	(search_lbs): Unnest search_lbs_aux.
	(setif_inclusion): Unnest collapse_cycle_lower, collapse_cycle_upper,
	update_lower_bound, update_upper_bound.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libbanshee/ChangeLog.diff?cvsroot=gcc&r1=1.5&r2=1.6
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libbanshee/engine/flowrow-sort.c.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libbanshee/engine/setif-sort.c.diff?cvsroot=gcc&r1=1.2&r2=1.3



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug bootstrap/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[giovannibajo at libero dot it]

------- Additional Comments From giovannibajo at libero dot it  2004-06-27 11:24 -------
Daniel tested it bootstrapping the compiler with PTA, so I committed the patch. 
I just removed the hunk about the 'inline' keyword.

To remove the VLAs, we need something like alloca(), but probably that's a 
configure test, so I won't touch it, somebody else will have to. I could change 
it to malloc() but I am not sure about the performance implications.

Also, I just noticed that there are compound literals in points-
to/andersen_terms.c, I guess those will have to be removed as well.

Eric, where does Sun CC stop now?

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug bootstrap/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[ebotcazou at gcc dot gnu dot org]

------- Additional Comments From ebotcazou at gcc dot gnu dot org  2004-06-27 13:53 -------
> Daniel tested it bootstrapping the compiler with PTA, so I committed the
> patch.

Thanks.

> Eric, where does Sun CC stop now?

cc -xarch=v9 -erroff -DHAVE_CONFIG_H -I.
-I/home/eric/cvs/gcc/libbanshee/libcompat -I..   
-I/home/eric/cvs/gcc/libbanshee/libcompat/../engine
-I/home/eric/cvs/gcc/libbanshee/libcompat/../include -I. -Ddeletes=
-Dtraditional= -Dsameregion= -Dparentptr=  -g -c
/home/eric/cvs/gcc/libbanshee/libcompat/regions.c
"/home/eric/cvs/gcc/libbanshee/libcompat/regions.c", line 374: address of a void
object cannot be taken
"/home/eric/cvs/gcc/libbanshee/libcompat/regions.c", line 374: address of a void
object cannot be taken
cc: acomp failed for /home/eric/cvs/gcc/libbanshee/libcompat/regions.c
gmake[3]: *** [regions.o] Error 2


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug bootstrap/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[dberlin at dberlin dot org]

------- Additional Comments From dberlin at dberlin dot org  2004-06-27 13:57 -------
Subject: Re:  [3.5 Regression] Sun CC cannot bootstrap
	GCC

On Sun, 2004-06-27 at 13:53 +0000, ebotcazou at gcc dot gnu dot org
wrote:

> > Eric, where does Sun CC stop now?
> 
> cc -xarch=v9 -erroff -DHAVE_CONFIG_H -I.
> -I/home/eric/cvs/gcc/libbanshee/libcompat -I..   
> -I/home/eric/cvs/gcc/libbanshee/libcompat/../engine
> -I/home/eric/cvs/gcc/libbanshee/libcompat/../include -I. -Ddeletes=
> -Dtraditional= -Dsameregion= -Dparentptr=  -g -c
> /home/eric/cvs/gcc/libbanshee/libcompat/regions.c
> "/home/eric/cvs/gcc/libbanshee/libcompat/regions.c", line 374: address of a void
> object cannot be taken
> "/home/eric/cvs/gcc/libbanshee/libcompat/regions.c", line 374: address of a void
> object cannot be taken
> cc: acomp failed for /home/eric/cvs/gcc/libbanshee/libcompat/regions.c
> gmake[3]: *** [regions.o] Error 2
> 

Can you simply delete that function/ifdef block (findgrefs), and see
what happens next?
The whole find* functions aren't used anywhere by us.




-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug bootstrap/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[giovannibajo at libero dot it]

------- Additional Comments From giovannibajo at libero dot it  2004-06-27 15:01 -------
Subject: Re:  [3.5 Regression] Sun CC cannot bootstrap GCC

dberlin at dberlin dot org wrote:

>>> Eric, where does Sun CC stop now?
>> 
>> cc -xarch=v9 -erroff -DHAVE_CONFIG_H -I.
>> -I/home/eric/cvs/gcc/libbanshee/libcompat -I..
>> -I/home/eric/cvs/gcc/libbanshee/libcompat/../engine
>> -I/home/eric/cvs/gcc/libbanshee/libcompat/../include -I. -Ddeletes=
>> -Dtraditional= -Dsameregion= -Dparentptr=  -g -c
>> /home/eric/cvs/gcc/libbanshee/libcompat/regions.c
>> "/home/eric/cvs/gcc/libbanshee/libcompat/regions.c", line 374:
>> address of a void object cannot be taken
>> "/home/eric/cvs/gcc/libbanshee/libcompat/regions.c", line 374:
>> address of a void object cannot be taken
>> cc: acomp failed for
>> /home/eric/cvs/gcc/libbanshee/libcompat/regions.c 
>> gmake[3]: *** [regions.o] Error 2
>> 
> 
> Can you simply delete that function/ifdef block (findgrefs), and see
> what happens next?
> The whole find* functions aren't used anywhere by us.

Or can you provide a login on a Solaris box where we can try directly?

Giovanni Bajo




-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug bootstrap/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-08-18 18:35 -------
*** Bug 17085 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |danglin at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug bootstrap/15627] [3.5 Regression] Sun CC cannot bootstrap GCC

[giovannibajo at libero dot it]

------- Additional Comments From giovannibajo at libero dot it  2004-09-10 00:24 -------
libbanshee is now gone, can you try again if bootstrap with Sun CC works now?

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |WAITING


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627

[Bug bootstrap/15627] [4.0 Regression] Sun CC cannot bootstrap GCC

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-09-10 07:09 -------
I know that disabling libbanshee fixed the error so closing as fixed.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |RESOLVED
         Resolution|                            |FIXED
            Summary|[3.5 Regression] Sun CC     |[4.0 Regression] Sun CC
                   |cannot bootstrap GCC        |cannot bootstrap GCC


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15627