GDB CVS won't build on OSF4.0's cc

GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

This patch fixes it.  I'm checking it in as obviously correct.

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On May 29, 2001, Alexandre Oliva <aoliva@redhat.com> wrote:

> This patch fixes it.  I'm checking it in as obviously correct.

> Index: gdb/ChangeLog
> from  Alexandre Oliva  <aoliva@redhat.com>

> 	* symfile.c (compare_psymbols, compare_symbols): Declare using
> 	PTR, as in the definition.

More of the same...  Checking in...

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

Another problem showed up while building gdbtk.  gdb/defs.h #includes
<stdarg.h>, but tcl/generic/tclInt.h #includes <varargs.h>, because
HAS_STDARG is not defined.

It turns out that the va_start macro defined in varargs.h ends up
overriding that of stdarg.h, so we call va_start with the wrong number
of arguments.

Since GDB assumes the existence of stdarg.h, I suppose this patch
would be acceptable.  Ok to install?

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

[-- Attachment #1: Type: text/plain, Size: 574 bytes --]

On May 29, 2001, Alexandre Oliva <aoliva@redhat.com> wrote:

> Another problem showed up while building gdbtk.  gdb/defs.h #includes
> <stdarg.h>, but tcl/generic/tclInt.h #includes <varargs.h>, because
> HAS_STDARG is not defined.

> It turns out that the va_start macro defined in varargs.h ends up
> overriding that of stdarg.h, so we call va_start with the wrong number
> of arguments.

> Since GDB assumes the existence of stdarg.h, I suppose this patch
> would be acceptable.  Ok to install?

This one seems to have never been acknowledged.  Ok for 5.2 and
mainline?


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: gdb-has-stdarg.patch --]
[-- Type: text/x-patch, Size: 513 bytes --]

Index: gdb/ChangeLog
from  Alexandre Oliva  <aoliva@redhat.com>

	* defs.h (HAS_STDARG): Define it.

Index: gdb/defs.h
===================================================================
RCS file: /cvs/src/src/gdb/defs.h,v
retrieving revision 1.50
diff -u -p -r1.50 defs.h
--- gdb/defs.h 2001/05/15 00:03:36 1.50
+++ gdb/defs.h 2001/05/29 11:10:38
@@ -54,6 +54,7 @@
 #include "ansidecl.h"
 
 #include <stdarg.h>		/* for va_list */
+#define HAS_STDARG 1		/* Avoid varargs.h in tcl.  */
 
 #include "libiberty.h"
 

[-- Attachment #3: Type: text/plain, Size: 289 bytes --]


-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Andrew Cagney]

[-- Attachment #1: Type: text/plain, Size: 855 bytes --]

[added insight]

> On May 29, 2001, Alexandre Oliva <aoliva@redhat.com> wrote:
> 
> 
>> Another problem showed up while building gdbtk.  gdb/defs.h #includes
>> <stdarg.h>, but tcl/generic/tclInt.h #includes <varargs.h>, because
>> HAS_STDARG is not defined.
> 
> 
>> It turns out that the va_start macro defined in varargs.h ends up
>> overriding that of stdarg.h, so we call va_start with the wrong number
>> of arguments.
> 
> 
>> Since GDB assumes the existence of stdarg.h, I suppose this patch
>> would be acceptable.  Ok to install?
> 
> 
> This one seems to have never been acknowledged.  Ok for 5.2 and
> mainline?

I don't know that this belongs in defs.h (definitly not without a big 
comment explaining the history :-).  As you note, gdb doesn't include 
tclInt.h, only GDBTK.  I suspect gdbtk/generic/gdbtk.h or autoconf is 
better?

Andrew


[-- Attachment #2: mailbox-message://ac131313@movemail/fsf/gdb/patches#899876 --]
[-- Type: message/rfc822, Size: 4176 bytes --]

[-- Attachment #2.1.1: Type: text/plain, Size: 574 bytes --]

On May 29, 2001, Alexandre Oliva <aoliva@redhat.com> wrote:

> Another problem showed up while building gdbtk.  gdb/defs.h #includes
> <stdarg.h>, but tcl/generic/tclInt.h #includes <varargs.h>, because
> HAS_STDARG is not defined.

> It turns out that the va_start macro defined in varargs.h ends up
> overriding that of stdarg.h, so we call va_start with the wrong number
> of arguments.

> Since GDB assumes the existence of stdarg.h, I suppose this patch
> would be acceptable.  Ok to install?

This one seems to have never been acknowledged.  Ok for 5.2 and
mainline?


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2.1.2: gdb-has-stdarg.patch --]
[-- Type: text/x-patch, Size: 513 bytes --]

Index: gdb/ChangeLog
from  Alexandre Oliva  <aoliva@redhat.com>

	* defs.h (HAS_STDARG): Define it.

Index: gdb/defs.h
===================================================================
RCS file: /cvs/src/src/gdb/defs.h,v
retrieving revision 1.50
diff -u -p -r1.50 defs.h
--- gdb/defs.h 2001/05/15 00:03:36 1.50
+++ gdb/defs.h 2001/05/29 11:10:38
@@ -54,6 +54,7 @@
 #include "ansidecl.h"
 
 #include <stdarg.h>		/* for va_list */
+#define HAS_STDARG 1		/* Avoid varargs.h in tcl.  */
 
 #include "libiberty.h"
 

[-- Attachment #2.1.3: Type: text/plain, Size: 289 bytes --]


-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

[-- Attachment #1: Type: text/plain, Size: 1144 bytes --]

On Mar  6, 2002, Andrew Cagney <ac131313@cygnus.com> wrote:

> I don't know that this belongs in defs.h (definitly not without a big
> comment explaining the history :-).  As you note, gdb doesn't include
> tclInt.h, only GDBTK.  I suspect gdbtk/generic/gdbtk.h or autoconf is
> better?

It would be a silly use of autoconf, since gdb just assumes stdarg.h
is there.  As for its not including tclInt.h, it's true that gdb
itself does not, but some of the gdbtk files include tk, itcl and tix
headers that end up including tclInt.h.  Unfortunately, they include
gdbtk.h only after including such headers, so adding a define in
gdbtk.h would be too late.

I agree we should try to keep gdbtk as separate from gdb as possible,
but I thought this define would be ok, since gdb's configure already
has code to look for tcl/tk.  Anyway, here's an alternate patch that I
dislike a bit on the grounds that it doesn't make sure the problem
won't show up again in the future, and it won't trigger any warnings
should gdb proper ever drop the assumption that stdarg.h is present
(unlikely, but... :-)

Ok to install?  Verified to build with OSF4.0's cc.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: gdb-has-stdarg.patch --]
[-- Type: text/x-patch, Size: 1988 bytes --]

Index: gdb/gdbtk/ChangeLog
from  Alexandre Oliva  <aoliva@redhat.com>

	* generic/gdbtk.c (HAS_STDARG): Define.
	* generic/gdbtk-cmds.c (HAS_STDARG): Likewise.
	* generic/gdbtk-hooks.c (HAS_STDARG): Likewise.

Index: gdb/gdbtk/generic/gdbtk-cmds.c
===================================================================
RCS file: /cvs/src/src/gdb/gdbtk/generic/gdbtk-cmds.c,v
retrieving revision 1.52
diff -u -p -r1.52 gdbtk-cmds.c
--- gdb/gdbtk/generic/gdbtk-cmds.c 2002/02/11 03:21:55 1.52
+++ gdb/gdbtk/generic/gdbtk-cmds.c 2002/03/07 02:47:19
@@ -35,6 +35,10 @@
 #include "top.h"
 #include "annotate.h"
 
+/* tcl header files includes varargs.h unless HAS_STDARG is defined,
+   but gdb uses stdarg.h, so make sure HAS_STDARG is defined.  */
+#define HAS_STDARG 1
+
 #include <tix.h>
 #include <itcl.h>
 
Index: gdb/gdbtk/generic/gdbtk-hooks.c
===================================================================
RCS file: /cvs/src/src/gdb/gdbtk/generic/gdbtk-hooks.c,v
retrieving revision 1.21
diff -u -p -r1.21 gdbtk-hooks.c
--- gdb/gdbtk/generic/gdbtk-hooks.c 2002/02/14 07:25:22 1.21
+++ gdb/gdbtk/generic/gdbtk-hooks.c 2002/03/07 02:47:20
@@ -37,6 +37,10 @@
 #include <windows.h>
 #endif
 
+/* tcl header files includes varargs.h unless HAS_STDARG is defined,
+   but gdb uses stdarg.h, so make sure HAS_STDARG is defined.  */
+#define HAS_STDARG 1
+
 #include <itcl.h>
 #include <tix.h>
 #include "guitcl.h"
Index: gdb/gdbtk/generic/gdbtk.c
===================================================================
RCS file: /cvs/src/src/gdb/gdbtk/generic/gdbtk.c,v
retrieving revision 1.26
diff -u -p -r1.26 gdbtk.c
--- gdb/gdbtk/generic/gdbtk.c 2002/02/07 21:56:07 1.26
+++ gdb/gdbtk/generic/gdbtk.c 2002/03/07 02:47:20
@@ -38,6 +38,10 @@
 #include <windows.h>
 #endif
 
+/* tcl header files includes varargs.h unless HAS_STDARG is defined,
+   but gdb uses stdarg.h, so make sure HAS_STDARG is defined.  */
+#define HAS_STDARG 1
+
 #include <tix.h>
 #include <itcl.h>
 #include <itk.h>

[-- Attachment #3: Type: text/plain, Size: 289 bytes --]


-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On Mar  6, 2002, Alexandre Oliva <aoliva@redhat.com> wrote:

> Index: gdb/gdbtk/ChangeLog
> from  Alexandre Oliva  <aoliva@redhat.com>

> 	* generic/gdbtk.c (HAS_STDARG): Define.
> 	* generic/gdbtk-cmds.c (HAS_STDARG): Likewise.
> 	* generic/gdbtk-hooks.c (HAS_STDARG): Likewise.

Ping?

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On Mar  6, 2002, Alexandre Oliva <aoliva@redhat.com> wrote:

> Ok to install?  Verified to build with OSF4.0's cc.


> Index: gdb/gdbtk/ChangeLog
> from  Alexandre Oliva  <aoliva@redhat.com>

> 	* generic/gdbtk.c (HAS_STDARG): Define.
> 	* generic/gdbtk-cmds.c (HAS_STDARG): Likewise.
> 	* generic/gdbtk-hooks.c (HAS_STDARG): Likewise.

Keith approved this patch in private, so I'm checking it in, mainline
and 5.2 branch.

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Elena Zannoni]

Alex, no. Sorry, but we are trying to get rid of PTR in the
gdb sources [as Kevin pointed out in another message].

Could you 'revert' these changes by changing the definitions to 
use void * and deleting the prototypes when they are not needed (everyone
of these can go, except for free_elfinfo, I think).

[see Andrew's patch: 
http://sources.redhat.com/ml/gdb-patches/2001-06/msg00077.html ]

Thanks and sorry again.

Elena

PS. I'll go through these files later and clean up some more of these.


Alexandre Oliva writes:

 > On May 29, 2001, Alexandre Oliva <aoliva@redhat.com> wrote:
 > 
 > > This patch fixes it.  I'm checking it in as obviously correct.
 > 
 > > Index: gdb/ChangeLog
 > > from  Alexandre Oliva  <aoliva@redhat.com>
 > 
 > > 	* symfile.c (compare_psymbols, compare_symbols): Declare using
 > > 	PTR, as in the definition.
 > 
 > More of the same...  Checking in...
 > 
 > Index: gdb/ChangeLog
 > 	* minsyms.c (compare_minimal_symbols): Likewise.
 > 	* coffread.c (find_targ_sec): Likewise.
 > 	* elfread.c (free_elfinfo, elf_locate_sections): Likewise.
 > 	* mipsread.c (alphacoff_locate_sections): Likewise.
 > 	* mdebugread.c (compare_blocks): Likewise.
 > Index: gdb/minsyms.c
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/minsyms.c,v
 > retrieving revision 1.16
 > diff -u -p -r1.16 minsyms.c
 > --- gdb/minsyms.c 2001/05/22 21:02:41 1.16
 > +++ gdb/minsyms.c 2001/05/29 10:33:09
 > @@ -77,7 +77,7 @@ static int msym_count;
 >  
 >  /* Prototypes for local functions. */
 >  
 > -static int compare_minimal_symbols (const void *, const void *);
 > +static int compare_minimal_symbols (const PTR, const PTR);
 >  
 >  static int
 >  compact_minimal_symbols (struct minimal_symbol *, int, struct objfile *);
 > Index: gdb/coffread.c
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/coffread.c,v
 > retrieving revision 1.17
 > diff -u -p -r1.17 coffread.c
 > --- gdb/coffread.c 2001/03/26 02:50:46 1.17
 > +++ gdb/coffread.c 2001/05/29 10:33:09
 > @@ -312,7 +312,7 @@ struct find_targ_sec_arg
 >      asection **resultp;
 >    };
 >  
 > -static void find_targ_sec (bfd *, asection *, void *);
 > +static void find_targ_sec (bfd *, asection *, PTR);
 >  
 >  static void
 >  find_targ_sec (bfd *abfd, asection *sect, PTR obj)
 > Index: gdb/elfread.c
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/elfread.c,v
 > retrieving revision 1.15
 > diff -u -p -r1.15 elfread.c
 > --- gdb/elfread.c 2001/03/06 08:21:07 1.15
 > +++ gdb/elfread.c 2001/05/29 10:33:09
 > @@ -76,7 +76,7 @@ static void elf_symfile_finish (struct o
 >  
 >  static void elf_symtab_read (struct objfile *, int);
 >  
 > -static void free_elfinfo (void *);
 > +static void free_elfinfo (PTR);
 >  
 >  static struct minimal_symbol *record_minimal_symbol_and_info (char *,
 >  							      CORE_ADDR,
 > @@ -88,7 +88,7 @@ static struct minimal_symbol *record_min
 >  							      struct objfile
 >  							      *);
 >  
 > -static void elf_locate_sections (bfd *, asection *, void *);
 > +static void elf_locate_sections (bfd *, asection *, PTR);
 >  
 >  /* We are called once per section from elf_symfile_read.  We
 >     need to examine each section we are passed, check to see
 > Index: gdb/mipsread.c
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/mipsread.c,v
 > retrieving revision 1.8
 > diff -u -p -r1.8 mipsread.c
 > --- gdb/mipsread.c 2001/03/07 02:57:08 1.8
 > +++ gdb/mipsread.c 2001/05/29 10:33:09
 > @@ -185,7 +185,7 @@ struct alphacoff_dynsecinfo
 >      asection *got_sect;		/* Section pointer for .got section */
 >    };
 >  
 > -static void alphacoff_locate_sections (bfd *, asection *, void *);
 > +static void alphacoff_locate_sections (bfd *, asection *, PTR);
 >  
 >  /* We are called once per section from read_alphacoff_dynamic_symtab.
 >     We need to examine each section we are passed, check to see
 > Index: gdb/mdebugread.c
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/mdebugread.c,v
 > retrieving revision 1.12
 > diff -u -p -r1.12 mdebugread.c
 > --- gdb/mdebugread.c 2001/03/07 02:57:08 1.12
 > +++ gdb/mdebugread.c 2001/05/29 10:33:09
 > @@ -378,7 +378,7 @@ static PTR xzalloc (unsigned int);
 >  
 >  static void sort_blocks (struct symtab *);
 >  
 > -static int compare_blocks (const void *, const void *);
 > +static int compare_blocks (const PTR, const PTR);
 >  
 >  static struct partial_symtab *new_psymtab (char *, struct objfile *);
 >  
 > 
 > -- 
 > Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
 > Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
 > CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
 > Free Software Evangelist    *Please* write to mailing lists, not to me

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On Jun  6, 2001, Elena Zannoni <ezannoni@cygnus.com> wrote:

> Alex, no. Sorry, but we are trying to get rid of PTR in the
> gdb sources [as Kevin pointed out in another message].

Sorry.

> Could you 'revert' these changes by changing the definitions to 
> use void * and deleting the prototypes when they are not needed (everyone
> of these can go, except for free_elfinfo, I think).

Fixed as follows.  Not tested on all relevant architectures.  I'd do
that after next Monday's GCC snapshot, since I've been building GDB
along with GDB lately.  Ok to install?

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On Jun 10, 2001, Eli Zaretskii <eliz@is.elta.co.il> wrote:

> On 9 Jun 2001, Alexandre Oliva wrote:

>> Fixed as follows.  Not tested on all relevant architectures.  I'd do
>> that after next Monday's GCC snapshot, since I've been building GDB
>> along with GDB lately.  Ok to install?
> [...]
>> -  memset ((PTR) objfile->sym_stab_info, 0, sizeof (struct dbx_symfile_info)); 
>> +  memset ((void *) objfile->sym_stab_info, 0,
>> +         sizeof (struct dbx_symfile_info));

> Do we _really_ have compilers which require to cast the first argument of
> memset to (void *)?

Dunno.  I was just dumbly replacing PTR with void*, as requested.

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist    *Please* write to mailing lists, not to me

Re: GDB CVS won't build on OSF4.0's cc

[Eli Zaretskii]

> From: Alexandre Oliva <aoliva@redhat.com>
> Date: 10 Jun 2001 18:25:42 -0300
> 
> On Jun 10, 2001, Eli Zaretskii <eliz@is.elta.co.il> wrote:
> 
> > On 9 Jun 2001, Alexandre Oliva wrote:
> 
> >> Fixed as follows.  Not tested on all relevant architectures.  I'd do
> >> that after next Monday's GCC snapshot, since I've been building GDB
> >> along with GDB lately.  Ok to install?
> > [...]
> >> -  memset ((PTR) objfile->sym_stab_info, 0, sizeof (struct dbx_symfile_info)); 
> >> +  memset ((void *) objfile->sym_stab_info, 0,
> >> +         sizeof (struct dbx_symfile_info));
> 
> > Do we _really_ have compilers which require to cast the first argument of
> > memset to (void *)?
> 
> Dunno.  I was just dumbly replacing PTR with void*, as requested.

Does anyone know?

IMHO, excess casting is evil, as it can mask programmatic errors.
However, if GDB standards don't frown on this, it's fine by me.

Re: GDB CVS won't build on OSF4.0's cc

[Russ Allbery]

Eli Zaretskii <eliz@is.elta.co.il> writes:

>> From: Alexandre Oliva <aoliva@redhat.com>
>> On Jun 10, 2001, Eli Zaretskii <eliz@is.elta.co.il> wrote:

>> >> -  memset ((PTR) objfile->sym_stab_info, 0, sizeof (struct dbx_symfile_info)); 
>> >> +  memset ((void *) objfile->sym_stab_info, 0,
>> >> +         sizeof (struct dbx_symfile_info));

>>> Do we _really_ have compilers which require to cast the first argument of
>>> memset to (void *)?

>> Dunno.  I was just dumbly replacing PTR with void*, as requested.

> Does anyone know?

> IMHO, excess casting is evil, as it can mask programmatic errors.
> However, if GDB standards don't frown on this, it's fine by me.

If memset is prototyped as taking a void *, this cast is unnecessary; the
C standard requires that the argument be converted appropriately due to
the prototype.  If memset is prototyped as taking a char *, this cast
neither hurts nor helps; the pointer has to be converted into a char *
anyway and that would be done by the prototype.

If memset isn't prototyped or has a K&R-style prototype without argument
types, then this could arguably help if pointers to different types of
objects were represented differently internally.  I'm a little skeptical
whether gdb supports such systems without proper function prototypes, and
I'd argue that this case is so insanely rare as to not bother worrying
about, particularly since gdb is now requiring ANSI C.

I've been systematically ripping those sorts of casts out of INN for a
while now.

-- 
Russ Allbery (rra@stanford.edu)             < http://www.eyrie.org/~eagle/ >

Re: GDB CVS won't build on OSF4.0's cc

[Andrew Cagney]

> I'd argue that this case is so insanely rare as to not bother worrying
> about, particularly since gdb is now requiring ANSI C.


I don't expect any one to disagree.  Ever since GDB started assuming 
ISO-C, people have been quietly removing these redundant casts.

Guess its time to extend the GDB mantra: Macro's are bad, casts are 
evil, m'kay.

	Andrew

Re: GDB CVS won't build on OSF4.0's cc

[Elena Zannoni]

Eli Zaretskii writes:
 > 
 > On 9 Jun 2001, Alexandre Oliva wrote:
 > 
 > > Fixed as follows.  Not tested on all relevant architectures.  I'd do
 > > that after next Monday's GCC snapshot, since I've been building GDB
 > > along with GDB lately.  Ok to install?
 > [...]
 > > -  memset ((PTR) objfile->sym_stab_info, 0, sizeof (struct dbx_symfile_info)); 
 > > +  memset ((void *) objfile->sym_stab_info, 0,
 > > +         sizeof (struct dbx_symfile_info));
 > 
 > Do we _really_ have compilers which require to cast the first argument of
 > memset to (void *)?  Which compiler(s) are those?  Isn't this a sign of a 
 > missing/conflicting prototype/header file?
 > 

Yes, I think we can safely get rid of these casts.
See a few line below, the call:

  memset (objfile->sym_private, 0, sizeof (struct coff_symfile_info));

Elena

Re: GDB CVS won't build on OSF4.0's cc

[Elena Zannoni]

Alexandre Oliva writes:
 > On Jun  6, 2001, Elena Zannoni <ezannoni@cygnus.com> wrote:
 > 
 > > Alex, no. Sorry, but we are trying to get rid of PTR in the
 > > gdb sources [as Kevin pointed out in another message].
 > 
 > Sorry.
 > 
 > > Could you 'revert' these changes by changing the definitions to 
 > > use void * and deleting the prototypes when they are not needed (everyone
 > > of these can go, except for free_elfinfo, I think).
 > 
 > Fixed as follows.  Not tested on all relevant architectures.  I'd do
 > that after next Monday's GCC snapshot, since I've been building GDB
 > along with GDB lately.  Ok to install?

Yes, but with the memset calls changes, i.e. remove the casts.

Elena

 > 
 > Index: gdb/ChangeLog
 > from  Alexandre Oliva  <aoliva@redhat.com>
 > 
 >       * coffread.c: Remove redundant static declarations.  Replace
 >       occurrences of `PTR' with `void *'.
 >       * elfread.c, mdebugread.c, minsyms.c, mipsread.c: Likewise.
 > 
 > Index: gdb/coffread.c
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/coffread.c,v
 > retrieving revision 1.18
 > diff -u -p -r1.18 coffread.c
 > --- gdb/coffread.c 2001/05/29 10:45:10 1.18
 > +++ gdb/coffread.c 2001/06/09 23:02:27
 > @@ -192,8 +192,6 @@ static struct symbol *process_coff_symbo
 >  
 >  static void patch_opaque_types (struct symtab *);
 >  
 > -static void patch_type (struct type *, struct type *);
 > -
 >  static void enter_linenos (long, int, int, struct objfile *);
 >  
 >  static void free_linetab (void);
 > @@ -216,32 +214,6 @@ static void read_one_sym (struct coff_sy
 >  			  struct internal_syment *, union internal_auxent *);
 >  
 >  static void coff_symtab_read (long, unsigned int, struct objfile *);
 > -
 > -static void find_linenos (bfd *, sec_ptr, PTR);
 > -
 > -static void coff_symfile_init (struct objfile *);
 > -
 > -static void coff_new_init (struct objfile *);
 > -
 > -static void coff_symfile_read (struct objfile *, int);
 > -
 > -static void coff_symfile_finish (struct objfile *);
 > -
 > -static void record_minimal_symbol (char *, CORE_ADDR,
 > -				   enum minimal_symbol_type,
 > -				   struct objfile *);
 > -
 > -static void coff_end_symtab (struct objfile *);
 > -
 > -static void complete_symtab (char *, CORE_ADDR, unsigned int);
 > -
 > -static void coff_start_symtab (char *);
 > -
 > -static struct type *coff_alloc_type (int);
 > -
 > -static struct type **coff_lookup_type (int);
 > -
 > -static void coff_locate_sections (bfd *, asection *, PTR);
 >  \f
 >  /* We are called once per section from coff_symfile_read.  We
 >     need to examine each section we are passed, check to see
 > @@ -254,7 +226,7 @@ static void coff_locate_sections (bfd *,
 >     -kingdon).  */
 >  
 >  static void
 > -coff_locate_sections (bfd *abfd, asection *sectp, PTR csip)
 > +coff_locate_sections (bfd *abfd, asection *sectp, void *csip)
 >  {
 >    register struct coff_symfile_info *csi;
 >    const char *name;
 > @@ -312,10 +284,8 @@ struct find_targ_sec_arg
 >      asection **resultp;
 >    };
 >  
 > -static void find_targ_sec (bfd *, asection *, PTR);
 > -
 >  static void
 > -find_targ_sec (bfd *abfd, asection *sect, PTR obj)
 > +find_targ_sec (bfd *abfd, asection *sect, void *obj)
 >  {
 >    struct find_targ_sec_arg *args = (struct find_targ_sec_arg *) obj;
 >    if (sect->target_index == args->targ_index)
 > @@ -509,7 +479,8 @@ coff_symfile_init (struct objfile *objfi
 >    objfile->sym_stab_info = (struct dbx_symfile_info *)
 >      xmmalloc (objfile->md, sizeof (struct dbx_symfile_info));
 >  
 > -  memset ((PTR) objfile->sym_stab_info, 0, sizeof (struct dbx_symfile_info));
 > +  memset ((void *) objfile->sym_stab_info, 0,
 > +	  sizeof (struct dbx_symfile_info));
 >  
 >    /* Allocate struct to keep track of the symfile */
 >    objfile->sym_private = xmmalloc (objfile->md,
 > @@ -531,7 +502,7 @@ coff_symfile_init (struct objfile *objfi
 >  
 >  /* ARGSUSED */
 >  static void
 > -find_linenos (bfd *abfd, sec_ptr asect, PTR vpinfo)
 > +find_linenos (bfd *abfd, sec_ptr asect, void *vpinfo)
 >  {
 >    struct coff_symfile_info *info;
 >    int size, count;
 > @@ -625,7 +596,7 @@ coff_symfile_read (struct objfile *objfi
 >    /* Read the line number table, all at once.  */
 >    info->min_lineno_offset = 0;
 >    info->max_lineno_offset = 0;
 > -  bfd_map_over_sections (abfd, find_linenos, (PTR) info);
 > +  bfd_map_over_sections (abfd, find_linenos, (void *) info);
 >  
 >    make_cleanup (free_linetab_cleanup, 0 /*ignore*/);
 >    val = init_lineno (abfd, info->min_lineno_offset,
 > @@ -662,7 +633,7 @@ coff_symfile_read (struct objfile *objfi
 >  
 >    install_minimal_symbols (objfile);
 >  
 > -  bfd_map_over_sections (abfd, coff_locate_sections, (PTR) info);
 > +  bfd_map_over_sections (abfd, coff_locate_sections, (void *) info);
 >  
 >    if (info->stabsects)
 >      {
 > Index: gdb/elfread.c
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/elfread.c,v
 > retrieving revision 1.16
 > diff -u -p -r1.16 elfread.c
 > --- gdb/elfread.c 2001/05/29 10:45:10 1.16
 > +++ gdb/elfread.c 2001/06/09 23:02:27
 > @@ -66,30 +66,8 @@ struct complaint stab_info_mismatch_comp
 >  struct complaint stab_info_questionable_complaint =
 >  {"elf/stab section information questionable for %s", 0, 0};
 >  
 > -static void elf_symfile_init (struct objfile *);
 > +static void free_elfinfo (void *);
 >  
 > -static void elf_new_init (struct objfile *);
 > -
 > -static void elf_symfile_read (struct objfile *, int);
 > -
 > -static void elf_symfile_finish (struct objfile *);
 > -
 > -static void elf_symtab_read (struct objfile *, int);
 > -
 > -static void free_elfinfo (PTR);
 > -
 > -static struct minimal_symbol *record_minimal_symbol_and_info (char *,
 > -							      CORE_ADDR,
 > -							      enum
 > -							      minimal_symbol_type,
 > -							      char *,
 > -							      asection *
 > -							      bfd_section,
 > -							      struct objfile
 > -							      *);
 > -
 > -static void elf_locate_sections (bfd *, asection *, PTR);
 > -
 >  /* We are called once per section from elf_symfile_read.  We
 >     need to examine each section we are passed, check to see
 >     if it is something we are interested in processing, and
 > @@ -110,7 +88,7 @@ static void elf_locate_sections (bfd *, 
 >     -kingdon).  */
 >  
 >  static void
 > -elf_locate_sections (bfd *ignore_abfd, asection *sectp, PTR eip)
 > +elf_locate_sections (bfd *ignore_abfd, asection *sectp, void *eip)
 >  {
 >    register struct elfinfo *ei;
 >  
 > @@ -445,7 +423,8 @@ elf_symtab_read (struct objfile *objfile
 >  			    {
 >  			      sectinfo = (struct stab_section_info *)
 >  				xmmalloc (objfile->md, sizeof (*sectinfo));
 > -			      memset ((PTR) sectinfo, 0, sizeof (*sectinfo));
 > +			      memset ((void *) sectinfo, 0,
 > +				      sizeof (*sectinfo));
 >  			      if (filesym == NULL)
 >  				{
 >  				  complain (&section_info_complaint,
 > @@ -513,7 +492,7 @@ elf_symtab_read (struct objfile *objfile
 >  	      size = ((elf_symbol_type *) sym)->internal_elf_sym.st_size;
 >  	      msym = record_minimal_symbol_and_info
 >  		((char *) sym->name, symaddr,
 > -		 ms_type, (PTR) size, sym->section, objfile);
 > +		 ms_type, (void *) size, sym->section, objfile);
 >  #ifdef SOFUN_ADDRESS_MAYBE_MISSING
 >  	      if (msym != NULL)
 >  		msym->filename = filesymname;
 > @@ -576,7 +555,7 @@ elf_symfile_read (struct objfile *objfil
 >    objfile->sym_stab_info = (struct dbx_symfile_info *)
 >      xmmalloc (objfile->md, sizeof (struct dbx_symfile_info));
 >    memset ((char *) objfile->sym_stab_info, 0, sizeof (struct dbx_symfile_info));
 > -  make_cleanup (free_elfinfo, (PTR) objfile);
 > +  make_cleanup (free_elfinfo, (void *) objfile);
 >  
 >    /* Process the normal ELF symbol table first.  This may write some 
 >       chain of info into the dbx_symfile_info in objfile->sym_stab_info,
 > @@ -601,7 +580,7 @@ elf_symfile_read (struct objfile *objfil
 >      }
 >  
 >    /* We first have to find them... */
 > -  bfd_map_over_sections (abfd, elf_locate_sections, (PTR) & ei);
 > +  bfd_map_over_sections (abfd, elf_locate_sections, (void *) & ei);
 >  
 >    /* ELF debugging information is inserted into the psymtab in the
 >       order of least informative first - most informative last.  Since
 > @@ -668,7 +647,7 @@ elf_symfile_read (struct objfile *objfil
 >     stab_section_info's, that might be dangling from it.  */
 >  
 >  static void
 > -free_elfinfo (PTR objp)
 > +free_elfinfo (void *objp)
 >  {
 >    struct objfile *objfile = (struct objfile *) objp;
 >    struct dbx_symfile_info *dbxinfo = objfile->sym_stab_info;
 > Index: gdb/mdebugread.c
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/mdebugread.c,v
 > retrieving revision 1.13
 > diff -u -p -r1.13 mdebugread.c
 > --- gdb/mdebugread.c 2001/05/29 10:45:10 1.13
 > +++ gdb/mdebugread.c 2001/06/09 23:02:27
 > @@ -325,30 +325,15 @@ static int found_ecoff_debugging_info;
 >  
 >  /* Forward declarations */
 >  
 > -static void add_pending (FDR *, char *, struct type *);
 > +static int upgrade_type (int, struct type **, int, union aux_ext *,
 > +			 int, char *);
 >  
 > -static struct mdebug_pending *is_pending_symbol (FDR *, char *);
 > -
 > -static void pop_parse_stack (void);
 > -
 > -static void push_parse_stack (void);
 > -
 > -static char *fdr_name (FDR *);
 > -
 > -static void mdebug_psymtab_to_symtab (struct partial_symtab *);
 > -
 > -static int
 > -upgrade_type (int, struct type **, int, union aux_ext *, int, char *);
 > -
 >  static void parse_partial_symbols (struct objfile *);
 >  
 > -static FDR * get_rfd (int, int);
 > -
 >  static int has_opaque_xref (FDR *, SYMR *);
 >  
 > -static int
 > -cross_ref (int, union aux_ext *, struct type **, enum type_code,
 > -	   char **, int, char *);
 > +static int cross_ref (int, union aux_ext *, struct type **, enum type_code,
 > +		      char **, int, char *);
 >  
 >  static struct symbol *new_symbol (char *);
 >  
 > @@ -362,10 +347,6 @@ static struct linetable *new_linetable (
 >  
 >  static struct blockvector *new_bvect (int);
 >  
 > -static int
 > -parse_symbol (SYMR *, union aux_ext *, char *, int, struct section_offsets *,
 > -	      struct objfile *);
 > -
 >  static struct type *parse_type (int, union aux_ext *, unsigned int, int *,
 >  				int, char *);
 >  
 > @@ -374,12 +355,8 @@ static struct symbol *mylookup_symbol (c
 >  
 >  static struct block *shrink_block (struct block *, struct symtab *);
 >  
 > -static PTR xzalloc (unsigned int);
 > -
 >  static void sort_blocks (struct symtab *);
 >  
 > -static int compare_blocks (const PTR, const PTR);
 > -
 >  static struct partial_symtab *new_psymtab (char *, struct objfile *);
 >  
 >  static void psymtab_to_symtab_1 (struct partial_symtab *, char *);
 > @@ -392,8 +369,8 @@ static int add_line (struct linetable *,
 >  
 >  static struct linetable *shrink_linetable (struct linetable *);
 >  
 > -static void
 > -handle_psymbol_enumerators (struct objfile *, FDR *, int, CORE_ADDR);
 > +static void handle_psymbol_enumerators (struct objfile *, FDR *, int,
 > +					CORE_ADDR);
 >  
 >  static char *mdebug_next_symbol_text (struct objfile *);
 >  \f
 > @@ -403,10 +380,10 @@ CORE_ADDR sigtramp_address, sigtramp_end
 >  
 >  /* Allocate zeroed memory */
 >  
 > -static PTR
 > +static void *
 >  xzalloc (unsigned int size)
 >  {
 > -  PTR p = xmalloc (size);
 > +  void *p = xmalloc (size);
 >  
 >    memset (p, 0, size);
 >    return p;
 > @@ -678,7 +655,7 @@ parse_symbol (SYMR *sh, union aux_ext *a
 >  	      struct section_offsets *section_offsets, struct objfile *objfile)
 >  {
 >    const bfd_size_type external_sym_size = debug_swap->external_sym_size;
 > -  void (*const swap_sym_in) (bfd *, PTR, SYMR *) = debug_swap->swap_sym_in;
 > +  void (*const swap_sym_in) (bfd *, void *, SYMR *) = debug_swap->swap_sym_in;
 >    char *name;
 >    struct symbol *s;
 >    struct block *b;
 > @@ -1117,7 +1094,7 @@ parse_symbol (SYMR *sh, union aux_ext *a
 >  		enum_sym = ((struct symbol *)
 >  			    obstack_alloc (&current_objfile->symbol_obstack,
 >  					   sizeof (struct symbol)));
 > -		memset ((PTR) enum_sym, 0, sizeof (struct symbol));
 > +		memset ((void *) enum_sym, 0, sizeof (struct symbol));
 >  		SYMBOL_NAME (enum_sym) =
 >  		  obsavestring (f->name, strlen (f->name),
 >  				&current_objfile->symbol_obstack);
 > @@ -1215,7 +1192,7 @@ parse_symbol (SYMR *sh, union aux_ext *a
 >  	  e = ((struct mips_extra_func_info *)
 >  	       obstack_alloc (&current_objfile->symbol_obstack,
 >  			      sizeof (struct mips_extra_func_info)));
 > -	  memset ((PTR) e, 0, sizeof (struct mips_extra_func_info));
 > +	  memset ((void *) e, 0, sizeof (struct mips_extra_func_info));
 >  	  SYMBOL_VALUE (s) = (long) e;
 >  	  e->numargs = top_stack->numargs;
 >  	  e->pdr.framereg = -1;
 > @@ -2209,9 +2186,9 @@ parse_partial_symbols (struct objfile *o
 >    const bfd_size_type external_sym_size = debug_swap->external_sym_size;
 >    const bfd_size_type external_rfd_size = debug_swap->external_rfd_size;
 >    const bfd_size_type external_ext_size = debug_swap->external_ext_size;
 > -  void (*const swap_ext_in) (bfd *, PTR, EXTR *) = debug_swap->swap_ext_in;
 > -  void (*const swap_sym_in) (bfd *, PTR, SYMR *) = debug_swap->swap_sym_in;
 > -  void (*const swap_rfd_in) (bfd *, PTR, RFDT *) = debug_swap->swap_rfd_in;
 > +  void (*const swap_ext_in) (bfd *, void *, EXTR *) = debug_swap->swap_ext_in;
 > +  void (*const swap_sym_in) (bfd *, void *, SYMR *) = debug_swap->swap_sym_in;
 > +  void (*const swap_rfd_in) (bfd *, void *, RFDT *) = debug_swap->swap_rfd_in;
 >    int f_idx, s_idx;
 >    HDRR *hdr = &debug_info->symbolic_header;
 >    /* Running pointers */
 > @@ -2298,7 +2275,7 @@ parse_partial_symbols (struct objfile *o
 >      ((struct mdebug_pending **)
 >       obstack_alloc (&objfile->psymbol_obstack,
 >  		    hdr->ifdMax * sizeof (struct mdebug_pending *)));
 > -  memset ((PTR) pending_list, 0,
 > +  memset ((void *) pending_list, 0,
 >  	  hdr->ifdMax * sizeof (struct mdebug_pending *));
 >  
 >    /* Pass 0 over external syms: swap them in.  */
 > @@ -2501,7 +2478,7 @@ parse_partial_symbols (struct objfile *o
 >        pst->read_symtab_private = ((char *)
 >  				  obstack_alloc (&objfile->psymbol_obstack,
 >  						 sizeof (struct symloc)));
 > -      memset ((PTR) pst->read_symtab_private, 0, sizeof (struct symloc));
 > +      memset ((void *) pst->read_symtab_private, 0, sizeof (struct symloc));
 >  
 >        save_pst = pst;
 >        FDR_IDX (pst) = f_idx;
 > @@ -3116,7 +3093,7 @@ handle_psymbol_enumerators (struct objfi
 >  			    CORE_ADDR svalue)
 >  {
 >    const bfd_size_type external_sym_size = debug_swap->external_sym_size;
 > -  void (*const swap_sym_in) (bfd *, PTR, SYMR *) = debug_swap->swap_sym_in;
 > +  void (*const swap_sym_in) (bfd *, void *, SYMR *) = debug_swap->swap_sym_in;
 >    char *ext_sym = ((char *) debug_info->external_sym
 >  		   + ((fh->isymBase + cur_sdx + 1) * external_sym_size));
 >    SYMR sh;
 > @@ -3207,8 +3184,8 @@ psymtab_to_symtab_1 (struct partial_symt
 >  {
 >    bfd_size_type external_sym_size;
 >    bfd_size_type external_pdr_size;
 > -  void (*swap_sym_in) (bfd *, PTR, SYMR *);
 > -  void (*swap_pdr_in) (bfd *, PTR, PDR *);
 > +  void (*swap_sym_in) (bfd *, void *, SYMR *);
 > +  void (*swap_pdr_in) (bfd *, void *, PDR *);
 >    int i;
 >    struct symtab *st;
 >    FDR *fh;
 > @@ -3338,7 +3315,7 @@ psymtab_to_symtab_1 (struct partial_symt
 >  				  sizeof (struct mips_extra_func_info)));
 >  		  struct symbol *s = new_symbol (MIPS_EFI_SYMBOL_NAME);
 >  
 > -		  memset ((PTR) e, 0, sizeof (struct mips_extra_func_info));
 > +		  memset ((void *) e, 0, sizeof (struct mips_extra_func_info));
 >  		  SYMBOL_NAMESPACE (s) = LABEL_NAMESPACE;
 >  		  SYMBOL_CLASS (s) = LOC_CONST;
 >  		  SYMBOL_TYPE (s) = mdebug_type_void;
 > @@ -3887,7 +3864,7 @@ add_block (struct block *b, struct symta
 >  {
 >    struct blockvector *bv = BLOCKVECTOR (s);
 >  
 > -  bv = (struct blockvector *) xrealloc ((PTR) bv,
 > +  bv = (struct blockvector *) xrealloc ((void *) bv,
 >  					(sizeof (struct blockvector)
 >  					 + BLOCKVECTOR_NBLOCKS (bv)
 >  					 * sizeof (bv->block)));
 > @@ -3935,7 +3912,7 @@ add_line (struct linetable *lt, int line
 >  /* Blocks with a smaller low bound should come first */
 >  
 >  static int
 > -compare_blocks (const PTR arg1, const PTR arg2)
 > +compare_blocks (const void *arg1, const void *arg2)
 >  {
 >    register int addr_diff;
 >    struct block **b1 = (struct block **) arg1;
 > @@ -4037,7 +4014,7 @@ new_psymtab (char *name, struct objfile 
 >    psymtab->read_symtab_private = ((char *)
 >  				  obstack_alloc (&objfile->psymbol_obstack,
 >  						 sizeof (struct symloc)));
 > -  memset ((PTR) psymtab->read_symtab_private, 0, sizeof (struct symloc));
 > +  memset ((void *) psymtab->read_symtab_private, 0, sizeof (struct symloc));
 >    CUR_BFD (psymtab) = cur_bfd;
 >    DEBUG_SWAP (psymtab) = debug_swap;
 >    DEBUG_INFO (psymtab) = debug_info;
 > @@ -4074,7 +4051,7 @@ static struct linetable *
 >  shrink_linetable (struct linetable *lt)
 >  {
 >  
 > -  return (struct linetable *) xrealloc ((PTR) lt,
 > +  return (struct linetable *) xrealloc ((void *) lt,
 >  					(sizeof (struct linetable)
 >  					 + ((lt->nitems - 1)
 >  					    * sizeof (lt->item))));
 > @@ -4118,7 +4095,7 @@ shrink_block (struct block *b, struct sy
 >  
 >    /* Just reallocate it and fix references to the old one */
 >  
 > -  new = (struct block *) xrealloc ((PTR) b,
 > +  new = (struct block *) xrealloc ((void *) b,
 >  				   (sizeof (struct block)
 >  				    + ((BLOCK_NSYMS (b) - 1)
 >  				       * sizeof (struct symbol *))));
 > @@ -4144,7 +4121,7 @@ new_symbol (char *name)
 >  		      obstack_alloc (&current_objfile->symbol_obstack,
 >  				     sizeof (struct symbol)));
 >  
 > -  memset ((PTR) s, 0, sizeof (*s));
 > +  memset ((void *) s, 0, sizeof (*s));
 >    SYMBOL_NAME (s) = obsavestring (name, strlen (name),
 >  				  &current_objfile->symbol_obstack);
 >    SYMBOL_LANGUAGE (s) = psymtab_language;
 > Index: gdb/minsyms.c
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/minsyms.c,v
 > retrieving revision 1.17
 > diff -u -p -r1.17 minsyms.c
 > --- gdb/minsyms.c 2001/05/29 10:45:10 1.17
 > +++ gdb/minsyms.c 2001/06/09 23:02:27
 > @@ -75,16 +75,6 @@ static int msym_bunch_index;
 >  
 >  static int msym_count;
 >  
 > -/* Prototypes for local functions. */
 > -
 > -static int compare_minimal_symbols (const PTR, const PTR);
 > -
 > -static int
 > -compact_minimal_symbols (struct minimal_symbol *, int, struct objfile *);
 > -
 > -static void add_minsym_to_demangled_hash_table (struct minimal_symbol *sym,
 > -						struct minimal_symbol **table);
 > -
 >  /* Compute a hash code based using the same criteria as `strcmp_iw'.  */
 >  
 >  unsigned int
 > @@ -686,7 +676,7 @@ prim_record_minimal_symbol_and_info (con
 >     Within groups with the same address, sort by name.  */
 >  
 >  static int
 > -compare_minimal_symbols (const PTR fn1p, const PTR fn2p)
 > +compare_minimal_symbols (const void *fn1p, const void *fn2p)
 >  {
 >    register const struct minimal_symbol *fn1;
 >    register const struct minimal_symbol *fn2;
 > Index: gdb/mipsread.c
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/mipsread.c,v
 > retrieving revision 1.9
 > diff -u -p -r1.9 mipsread.c
 > --- gdb/mipsread.c 2001/05/29 10:45:10 1.9
 > +++ gdb/mipsread.c 2001/06/09 23:02:27
 > @@ -185,15 +185,13 @@ struct alphacoff_dynsecinfo
 >      asection *got_sect;		/* Section pointer for .got section */
 >    };
 >  
 > -static void alphacoff_locate_sections (bfd *, asection *, PTR);
 > -
 >  /* We are called once per section from read_alphacoff_dynamic_symtab.
 >     We need to examine each section we are passed, check to see
 >     if it is something we are interested in processing, and
 >     if so, stash away some access information for the section.  */
 >  
 >  static void
 > -alphacoff_locate_sections (bfd *ignore_abfd, asection *sectp, PTR sip)
 > +alphacoff_locate_sections (bfd *ignore_abfd, asection *sectp, void *sip)
 >  {
 >    register struct alphacoff_dynsecinfo *si;
 >  
 > @@ -252,7 +250,7 @@ read_alphacoff_dynamic_symtab (struct se
 >  
 >    /* Locate the dynamic symbols sections and read them in.  */
 >    memset ((char *) &si, 0, sizeof (si));
 > -  bfd_map_over_sections (abfd, alphacoff_locate_sections, (PTR) & si);
 > +  bfd_map_over_sections (abfd, alphacoff_locate_sections, (void *) & si);
 >    if (si.sym_sect == NULL
 >        || si.str_sect == NULL
 >        || si.dyninfo_sect == NULL
 > 
 > -- 
 > Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
 > Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
 > CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
 > Free Software Evangelist    *Please* write to mailing lists, not to me

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

[-- Attachment #1: Type: text/plain, Size: 1428 bytes --]

On Jun 11, 2001, Elena Zannoni <ezannoni@cygnus.com> wrote:

> Alexandre Oliva writes:
>> On Jun  6, 2001, Elena Zannoni <ezannoni@cygnus.com> wrote:
>> 
>> > Alex, no. Sorry, but we are trying to get rid of PTR in the
>> > gdb sources [as Kevin pointed out in another message].
>> 
>> Sorry.
>> 
>> > Could you 'revert' these changes by changing the definitions to 
>> > use void * and deleting the prototypes when they are not needed (everyone
>> > of these can go, except for free_elfinfo, I think).
>> 
>> Fixed as follows.  Not tested on all relevant architectures.  I'd do
>> that after next Monday's GCC snapshot, since I've been building GDB
>> along with GDB lately.  Ok to install?

> Yes, but with the memset calls changes, i.e. remove the casts.

>> Index: gdb/ChangeLog
>> from  Alexandre Oliva  <aoliva@redhat.com>
>> 
>> * coffread.c: Remove redundant static declarations.  Replace
>> occurrences of `PTR' with `void *'.
>> * elfread.c, mdebugread.c, minsyms.c, mipsread.c: Likewise.

Ok, here's an updated patch, that applies cleanly into mainline and
5.2 branch, and that adds two more PTR->void* fixes that slipped in
after my original clean-up.  Oh, and that removes the (void *) casts
in memset calls that I had introduced in my previous patch.  Ok to
install in 5.2 and mainline?  With this and the other 2 patches I
posted today (for gdb/defs.h and expect), gdb 5.2 builds successfully
with OSF 4.0's cc.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: gdb-PTR-static-cleanup.patch --]
[-- Type: text/x-patch, Size: 19841 bytes --]

Index: gdb/ChangeLog
from  Alexandre Oliva  <aoliva@redhat.com>
	* coffread.c: Remove redundant static declarations.  Replace
	occurrences of `PTR' with `void *'.
	* elfread.c, mdebugread.c, minsyms.c, mipsread.c: Likewise.
	* top.h (quit_cover): Likewise.
	* defs.h (catch_errors): Likewise.

Index: gdb/coffread.c
===================================================================
RCS file: /cvs/src/src/gdb/coffread.c,v
retrieving revision 1.25
diff -u -p -r1.25 coffread.c
--- gdb/coffread.c 2002/02/06 01:20:23 1.25
+++ gdb/coffread.c 2002/03/07 00:23:24
@@ -192,8 +192,6 @@ static struct symbol *process_coff_symbo
 
 static void patch_opaque_types (struct symtab *);
 
-static void patch_type (struct type *, struct type *);
-
 static void enter_linenos (long, int, int, struct objfile *);
 
 static void free_linetab (void);
@@ -216,32 +214,6 @@ static void read_one_sym (struct coff_sy
 			  struct internal_syment *, union internal_auxent *);
 
 static void coff_symtab_read (long, unsigned int, struct objfile *);
-
-static void find_linenos (bfd *, sec_ptr, PTR);
-
-static void coff_symfile_init (struct objfile *);
-
-static void coff_new_init (struct objfile *);
-
-static void coff_symfile_read (struct objfile *, int);
-
-static void coff_symfile_finish (struct objfile *);
-
-static void record_minimal_symbol (char *, CORE_ADDR,
-				   enum minimal_symbol_type,
-				   struct objfile *);
-
-static void coff_end_symtab (struct objfile *);
-
-static void complete_symtab (char *, CORE_ADDR, unsigned int);
-
-static void coff_start_symtab (char *);
-
-static struct type *coff_alloc_type (int);
-
-static struct type **coff_lookup_type (int);
-
-static void coff_locate_sections (bfd *, asection *, PTR);
 \f
 /* We are called once per section from coff_symfile_read.  We
    need to examine each section we are passed, check to see
@@ -254,7 +226,7 @@ static void coff_locate_sections (bfd *,
    -kingdon).  */
 
 static void
-coff_locate_sections (bfd *abfd, asection *sectp, PTR csip)
+coff_locate_sections (bfd *abfd, asection *sectp, void *csip)
 {
   register struct coff_symfile_info *csi;
   const char *name;
@@ -312,10 +284,8 @@ struct find_targ_sec_arg
     asection **resultp;
   };
 
-static void find_targ_sec (bfd *, asection *, PTR);
-
 static void
-find_targ_sec (bfd *abfd, asection *sect, PTR obj)
+find_targ_sec (bfd *abfd, asection *sect, void *obj)
 {
   struct find_targ_sec_arg *args = (struct find_targ_sec_arg *) obj;
   if (sect->target_index == args->targ_index)
@@ -509,7 +479,8 @@ coff_symfile_init (struct objfile *objfi
   objfile->sym_stab_info = (struct dbx_symfile_info *)
     xmmalloc (objfile->md, sizeof (struct dbx_symfile_info));
 
-  memset ((PTR) objfile->sym_stab_info, 0, sizeof (struct dbx_symfile_info));
+  memset (objfile->sym_stab_info, 0,
+	  sizeof (struct dbx_symfile_info));
 
   /* Allocate struct to keep track of the symfile */
   objfile->sym_private = xmmalloc (objfile->md,
@@ -531,7 +502,7 @@ coff_symfile_init (struct objfile *objfi
 
 /* ARGSUSED */
 static void
-find_linenos (bfd *abfd, sec_ptr asect, PTR vpinfo)
+find_linenos (bfd *abfd, sec_ptr asect, void *vpinfo)
 {
   struct coff_symfile_info *info;
   int size, count;
@@ -625,7 +596,7 @@ coff_symfile_read (struct objfile *objfi
   /* Read the line number table, all at once.  */
   info->min_lineno_offset = 0;
   info->max_lineno_offset = 0;
-  bfd_map_over_sections (abfd, find_linenos, (PTR) info);
+  bfd_map_over_sections (abfd, find_linenos, (void *) info);
 
   make_cleanup (free_linetab_cleanup, 0 /*ignore*/);
   val = init_lineno (abfd, info->min_lineno_offset,
@@ -662,7 +633,7 @@ coff_symfile_read (struct objfile *objfi
 
   install_minimal_symbols (objfile);
 
-  bfd_map_over_sections (abfd, coff_locate_sections, (PTR) info);
+  bfd_map_over_sections (abfd, coff_locate_sections, (void *) info);
 
   if (info->stabsects)
     {
Index: gdb/elfread.c
===================================================================
RCS file: /cvs/src/src/gdb/elfread.c,v
retrieving revision 1.20
diff -u -p -r1.20 elfread.c
--- gdb/elfread.c 2002/02/06 01:20:23 1.20
+++ gdb/elfread.c 2002/03/07 00:23:26
@@ -66,30 +66,8 @@ struct complaint stab_info_mismatch_comp
 struct complaint stab_info_questionable_complaint =
 {"elf/stab section information questionable for %s", 0, 0};
 
-static void elf_symfile_init (struct objfile *);
+static void free_elfinfo (void *);
 
-static void elf_new_init (struct objfile *);
-
-static void elf_symfile_read (struct objfile *, int);
-
-static void elf_symfile_finish (struct objfile *);
-
-static void elf_symtab_read (struct objfile *, int);
-
-static void free_elfinfo (PTR);
-
-static struct minimal_symbol *record_minimal_symbol_and_info (char *,
-							      CORE_ADDR,
-							      enum
-							      minimal_symbol_type,
-							      char *,
-							      asection *
-							      bfd_section,
-							      struct objfile
-							      *);
-
-static void elf_locate_sections (bfd *, asection *, PTR);
-
 /* We are called once per section from elf_symfile_read.  We
    need to examine each section we are passed, check to see
    if it is something we are interested in processing, and
@@ -110,7 +88,7 @@ static void elf_locate_sections (bfd *, 
    -kingdon).  */
 
 static void
-elf_locate_sections (bfd *ignore_abfd, asection *sectp, PTR eip)
+elf_locate_sections (bfd *ignore_abfd, asection *sectp, void *eip)
 {
   register struct elfinfo *ei;
 
@@ -443,7 +421,8 @@ elf_symtab_read (struct objfile *objfile
 			    {
 			      sectinfo = (struct stab_section_info *)
 				xmmalloc (objfile->md, sizeof (*sectinfo));
-			      memset ((PTR) sectinfo, 0, sizeof (*sectinfo));
+			      memset (sectinfo, 0,
+				      sizeof (*sectinfo));
 			      if (filesym == NULL)
 				{
 				  complain (&section_info_complaint,
@@ -511,7 +490,7 @@ elf_symtab_read (struct objfile *objfile
 	      size = ((elf_symbol_type *) sym)->internal_elf_sym.st_size;
 	      msym = record_minimal_symbol_and_info
 		((char *) sym->name, symaddr,
-		 ms_type, (PTR) size, sym->section, objfile);
+		 ms_type, (void *) size, sym->section, objfile);
 #ifdef SOFUN_ADDRESS_MAYBE_MISSING
 	      if (msym != NULL)
 		msym->filename = filesymname;
@@ -572,7 +551,7 @@ elf_symfile_read (struct objfile *objfil
   objfile->sym_stab_info = (struct dbx_symfile_info *)
     xmmalloc (objfile->md, sizeof (struct dbx_symfile_info));
   memset ((char *) objfile->sym_stab_info, 0, sizeof (struct dbx_symfile_info));
-  make_cleanup (free_elfinfo, (PTR) objfile);
+  make_cleanup (free_elfinfo, (void *) objfile);
 
   /* Process the normal ELF symbol table first.  This may write some 
      chain of info into the dbx_symfile_info in objfile->sym_stab_info,
@@ -597,7 +576,7 @@ elf_symfile_read (struct objfile *objfil
     }
 
   /* We first have to find them... */
-  bfd_map_over_sections (abfd, elf_locate_sections, (PTR) & ei);
+  bfd_map_over_sections (abfd, elf_locate_sections, (void *) & ei);
 
   /* ELF debugging information is inserted into the psymtab in the
      order of least informative first - most informative last.  Since
@@ -667,7 +646,7 @@ elf_symfile_read (struct objfile *objfil
    stab_section_info's, that might be dangling from it.  */
 
 static void
-free_elfinfo (PTR objp)
+free_elfinfo (void *objp)
 {
   struct objfile *objfile = (struct objfile *) objp;
   struct dbx_symfile_info *dbxinfo = objfile->sym_stab_info;
Index: gdb/mdebugread.c
===================================================================
RCS file: /cvs/src/src/gdb/mdebugread.c,v
retrieving revision 1.23
diff -u -p -r1.23 mdebugread.c
--- gdb/mdebugread.c 2002/01/24 04:08:33 1.23
+++ gdb/mdebugread.c 2002/03/07 00:23:29
@@ -327,30 +327,15 @@ static int found_ecoff_debugging_info;
 
 /* Forward declarations */
 
-static void add_pending (FDR *, char *, struct type *);
+static int upgrade_type (int, struct type **, int, union aux_ext *,
+			 int, char *);
 
-static struct mdebug_pending *is_pending_symbol (FDR *, char *);
-
-static void pop_parse_stack (void);
-
-static void push_parse_stack (void);
-
-static char *fdr_name (FDR *);
-
-static void mdebug_psymtab_to_symtab (struct partial_symtab *);
-
-static int
-upgrade_type (int, struct type **, int, union aux_ext *, int, char *);
-
 static void parse_partial_symbols (struct objfile *);
 
-static FDR * get_rfd (int, int);
-
 static int has_opaque_xref (FDR *, SYMR *);
 
-static int
-cross_ref (int, union aux_ext *, struct type **, enum type_code,
-	   char **, int, char *);
+static int cross_ref (int, union aux_ext *, struct type **, enum type_code,
+		      char **, int, char *);
 
 static struct symbol *new_symbol (char *);
 
@@ -364,10 +349,6 @@ static struct linetable *new_linetable (
 
 static struct blockvector *new_bvect (int);
 
-static int
-parse_symbol (SYMR *, union aux_ext *, char *, int, struct section_offsets *,
-	      struct objfile *);
-
 static struct type *parse_type (int, union aux_ext *, unsigned int, int *,
 				int, char *);
 
@@ -376,12 +357,8 @@ static struct symbol *mylookup_symbol (c
 
 static struct block *shrink_block (struct block *, struct symtab *);
 
-static PTR xzalloc (unsigned int);
-
 static void sort_blocks (struct symtab *);
 
-static int compare_blocks (const PTR, const PTR);
-
 static struct partial_symtab *new_psymtab (char *, struct objfile *);
 
 static void psymtab_to_symtab_1 (struct partial_symtab *, char *);
@@ -394,8 +371,8 @@ static int add_line (struct linetable *,
 
 static struct linetable *shrink_linetable (struct linetable *);
 
-static void
-handle_psymbol_enumerators (struct objfile *, FDR *, int, CORE_ADDR);
+static void handle_psymbol_enumerators (struct objfile *, FDR *, int,
+					CORE_ADDR);
 
 static char *mdebug_next_symbol_text (struct objfile *);
 \f
@@ -405,10 +382,10 @@ CORE_ADDR sigtramp_address, sigtramp_end
 
 /* Allocate zeroed memory */
 
-static PTR
+static void *
 xzalloc (unsigned int size)
 {
-  PTR p = xmalloc (size);
+  void *p = xmalloc (size);
 
   memset (p, 0, size);
   return p;
@@ -698,7 +675,7 @@ parse_symbol (SYMR *sh, union aux_ext *a
 	      struct section_offsets *section_offsets, struct objfile *objfile)
 {
   const bfd_size_type external_sym_size = debug_swap->external_sym_size;
-  void (*const swap_sym_in) (bfd *, PTR, SYMR *) = debug_swap->swap_sym_in;
+  void (*const swap_sym_in) (bfd *, void *, SYMR *) = debug_swap->swap_sym_in;
   char *name;
   struct symbol *s;
   struct block *b;
@@ -1137,7 +1114,7 @@ parse_symbol (SYMR *sh, union aux_ext *a
 		enum_sym = ((struct symbol *)
 			    obstack_alloc (&current_objfile->symbol_obstack,
 					   sizeof (struct symbol)));
-		memset ((PTR) enum_sym, 0, sizeof (struct symbol));
+		memset (enum_sym, 0, sizeof (struct symbol));
 		SYMBOL_NAME (enum_sym) =
 		  obsavestring (f->name, strlen (f->name),
 				&current_objfile->symbol_obstack);
@@ -1235,7 +1212,7 @@ parse_symbol (SYMR *sh, union aux_ext *a
 	  e = ((struct mips_extra_func_info *)
 	       obstack_alloc (&current_objfile->symbol_obstack,
 			      sizeof (struct mips_extra_func_info)));
-	  memset ((PTR) e, 0, sizeof (struct mips_extra_func_info));
+	  memset (e, 0, sizeof (struct mips_extra_func_info));
 	  SYMBOL_VALUE (s) = (long) e;
 	  e->numargs = top_stack->numargs;
 	  e->pdr.framereg = -1;
@@ -2230,9 +2207,9 @@ parse_partial_symbols (struct objfile *o
   const bfd_size_type external_sym_size = debug_swap->external_sym_size;
   const bfd_size_type external_rfd_size = debug_swap->external_rfd_size;
   const bfd_size_type external_ext_size = debug_swap->external_ext_size;
-  void (*const swap_ext_in) (bfd *, PTR, EXTR *) = debug_swap->swap_ext_in;
-  void (*const swap_sym_in) (bfd *, PTR, SYMR *) = debug_swap->swap_sym_in;
-  void (*const swap_rfd_in) (bfd *, PTR, RFDT *) = debug_swap->swap_rfd_in;
+  void (*const swap_ext_in) (bfd *, void *, EXTR *) = debug_swap->swap_ext_in;
+  void (*const swap_sym_in) (bfd *, void *, SYMR *) = debug_swap->swap_sym_in;
+  void (*const swap_rfd_in) (bfd *, void *, RFDT *) = debug_swap->swap_rfd_in;
   int f_idx, s_idx;
   HDRR *hdr = &debug_info->symbolic_header;
   /* Running pointers */
@@ -2319,7 +2296,7 @@ parse_partial_symbols (struct objfile *o
     ((struct mdebug_pending **)
      obstack_alloc (&objfile->psymbol_obstack,
 		    hdr->ifdMax * sizeof (struct mdebug_pending *)));
-  memset ((PTR) pending_list, 0,
+  memset (pending_list, 0,
 	  hdr->ifdMax * sizeof (struct mdebug_pending *));
 
   /* Pass 0 over external syms: swap them in.  */
@@ -2522,7 +2499,7 @@ parse_partial_symbols (struct objfile *o
       pst->read_symtab_private = ((char *)
 				  obstack_alloc (&objfile->psymbol_obstack,
 						 sizeof (struct symloc)));
-      memset ((PTR) pst->read_symtab_private, 0, sizeof (struct symloc));
+      memset (pst->read_symtab_private, 0, sizeof (struct symloc));
 
       save_pst = pst;
       TEXTLOW (pst) = pst->textlow;
@@ -3673,7 +3650,7 @@ handle_psymbol_enumerators (struct objfi
 			    CORE_ADDR svalue)
 {
   const bfd_size_type external_sym_size = debug_swap->external_sym_size;
-  void (*const swap_sym_in) (bfd *, PTR, SYMR *) = debug_swap->swap_sym_in;
+  void (*const swap_sym_in) (bfd *, void *, SYMR *) = debug_swap->swap_sym_in;
   char *ext_sym = ((char *) debug_info->external_sym
 		   + ((fh->isymBase + cur_sdx + 1) * external_sym_size));
   SYMR sh;
@@ -3764,8 +3741,8 @@ psymtab_to_symtab_1 (struct partial_symt
 {
   bfd_size_type external_sym_size;
   bfd_size_type external_pdr_size;
-  void (*swap_sym_in) (bfd *, PTR, SYMR *);
-  void (*swap_pdr_in) (bfd *, PTR, PDR *);
+  void (*swap_sym_in) (bfd *, void *, SYMR *);
+  void (*swap_pdr_in) (bfd *, void *, PDR *);
   int i;
   struct symtab *st = NULL;
   FDR *fh;
@@ -3918,7 +3895,7 @@ psymtab_to_symtab_1 (struct partial_symt
 				  sizeof (struct mips_extra_func_info)));
 		  struct symbol *s = new_symbol (MIPS_EFI_SYMBOL_NAME);
 
-		  memset ((PTR) e, 0, sizeof (struct mips_extra_func_info));
+		  memset (e, 0, sizeof (struct mips_extra_func_info));
 		  SYMBOL_NAMESPACE (s) = LABEL_NAMESPACE;
 		  SYMBOL_CLASS (s) = LOC_CONST;
 		  SYMBOL_TYPE (s) = mdebug_type_void;
@@ -4467,7 +4444,7 @@ add_block (struct block *b, struct symta
 {
   struct blockvector *bv = BLOCKVECTOR (s);
 
-  bv = (struct blockvector *) xrealloc ((PTR) bv,
+  bv = (struct blockvector *) xrealloc ((void *) bv,
 					(sizeof (struct blockvector)
 					 + BLOCKVECTOR_NBLOCKS (bv)
 					 * sizeof (bv->block)));
@@ -4515,7 +4492,7 @@ add_line (struct linetable *lt, int line
 /* Blocks with a smaller low bound should come first */
 
 static int
-compare_blocks (const PTR arg1, const PTR arg2)
+compare_blocks (const void *arg1, const void *arg2)
 {
   register int addr_diff;
   struct block **b1 = (struct block **) arg1;
@@ -4617,7 +4594,7 @@ new_psymtab (char *name, struct objfile 
   psymtab->read_symtab_private = ((char *)
 				  obstack_alloc (&objfile->psymbol_obstack,
 						 sizeof (struct symloc)));
-  memset ((PTR) psymtab->read_symtab_private, 0, sizeof (struct symloc));
+  memset (psymtab->read_symtab_private, 0, sizeof (struct symloc));
   CUR_BFD (psymtab) = cur_bfd;
   DEBUG_SWAP (psymtab) = debug_swap;
   DEBUG_INFO (psymtab) = debug_info;
@@ -4654,7 +4631,7 @@ static struct linetable *
 shrink_linetable (struct linetable *lt)
 {
 
-  return (struct linetable *) xrealloc ((PTR) lt,
+  return (struct linetable *) xrealloc ((void *) lt,
 					(sizeof (struct linetable)
 					 + ((lt->nitems - 1)
 					    * sizeof (lt->item))));
@@ -4698,7 +4675,7 @@ shrink_block (struct block *b, struct sy
 
   /* Just reallocate it and fix references to the old one */
 
-  new = (struct block *) xrealloc ((PTR) b,
+  new = (struct block *) xrealloc ((void *) b,
 				   (sizeof (struct block)
 				    + ((BLOCK_NSYMS (b) - 1)
 				       * sizeof (struct symbol *))));
@@ -4724,7 +4701,7 @@ new_symbol (char *name)
 		      obstack_alloc (&current_objfile->symbol_obstack,
 				     sizeof (struct symbol)));
 
-  memset ((PTR) s, 0, sizeof (*s));
+  memset (s, 0, sizeof (*s));
   SYMBOL_NAME (s) = obsavestring (name, strlen (name),
 				  &current_objfile->symbol_obstack);
   SYMBOL_LANGUAGE (s) = psymtab_language;
Index: gdb/minsyms.c
===================================================================
RCS file: /cvs/src/src/gdb/minsyms.c,v
retrieving revision 1.19
diff -u -p -r1.19 minsyms.c
--- gdb/minsyms.c 2001/12/10 22:04:10 1.19
+++ gdb/minsyms.c 2002/03/07 00:23:30
@@ -75,16 +75,6 @@ static int msym_bunch_index;
 
 static int msym_count;
 
-/* Prototypes for local functions. */
-
-static int compare_minimal_symbols (const PTR, const PTR);
-
-static int
-compact_minimal_symbols (struct minimal_symbol *, int, struct objfile *);
-
-static void add_minsym_to_demangled_hash_table (struct minimal_symbol *sym,
-						struct minimal_symbol **table);
-
 /* Compute a hash code based using the same criteria as `strcmp_iw'.  */
 
 unsigned int
@@ -688,7 +678,7 @@ prim_record_minimal_symbol_and_info (con
    Within groups with the same address, sort by name.  */
 
 static int
-compare_minimal_symbols (const PTR fn1p, const PTR fn2p)
+compare_minimal_symbols (const void *fn1p, const void *fn2p)
 {
   register const struct minimal_symbol *fn1;
   register const struct minimal_symbol *fn2;
Index: gdb/mipsread.c
===================================================================
RCS file: /cvs/src/src/gdb/mipsread.c,v
retrieving revision 1.9
diff -u -p -r1.9 mipsread.c
--- gdb/mipsread.c 2001/05/29 10:45:10 1.9
+++ gdb/mipsread.c 2002/03/07 00:23:30
@@ -185,15 +185,13 @@ struct alphacoff_dynsecinfo
     asection *got_sect;		/* Section pointer for .got section */
   };
 
-static void alphacoff_locate_sections (bfd *, asection *, PTR);
-
 /* We are called once per section from read_alphacoff_dynamic_symtab.
    We need to examine each section we are passed, check to see
    if it is something we are interested in processing, and
    if so, stash away some access information for the section.  */
 
 static void
-alphacoff_locate_sections (bfd *ignore_abfd, asection *sectp, PTR sip)
+alphacoff_locate_sections (bfd *ignore_abfd, asection *sectp, void *sip)
 {
   register struct alphacoff_dynsecinfo *si;
 
@@ -252,7 +250,7 @@ read_alphacoff_dynamic_symtab (struct se
 
   /* Locate the dynamic symbols sections and read them in.  */
   memset ((char *) &si, 0, sizeof (si));
-  bfd_map_over_sections (abfd, alphacoff_locate_sections, (PTR) & si);
+  bfd_map_over_sections (abfd, alphacoff_locate_sections, (void *) & si);
   if (si.sym_sect == NULL
       || si.str_sect == NULL
       || si.dyninfo_sect == NULL
Index: gdb/top.h
===================================================================
RCS file: /cvs/src/src/gdb/top.h,v
retrieving revision 1.6
diff -u -p -r1.6 top.h
--- gdb/top.h 2001/10/21 19:20:30 1.6
+++ gdb/top.h 2002/03/07 00:23:30
@@ -44,7 +44,7 @@ extern void simplified_command_loop (cha
 extern int quit_confirm (void);
 extern void quit_force (char *, int);
 extern void quit_command (char *, int);
-extern int quit_cover (PTR);
+extern int quit_cover (void *);
 extern void execute_command (char *, int);
 
 /* This function returns a pointer to the string that is used
Index: gdb/defs.h
===================================================================
RCS file: /cvs/src/src/gdb/defs.h,v
retrieving revision 1.81
diff -u -p -r1.81 defs.h
--- gdb/defs.h 2002/02/26 03:29:55 1.81
+++ gdb/defs.h 2002/03/07 00:23:25
@@ -1152,7 +1152,7 @@ extern int catch_exceptions (struct ui_o
    This function is superseeded by catch_exceptions().  */
 
 typedef int (catch_errors_ftype) (PTR);
-extern int catch_errors (catch_errors_ftype *, PTR, char *, return_mask);
+extern int catch_errors (catch_errors_ftype *, void *, char *, return_mask);
 
 /* Template to catch_errors() that wraps calls to command
    functions. */

[-- Attachment #3: Type: text/plain, Size: 289 bytes --]


-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On Mar  6, 2002, Alexandre Oliva <aoliva@redhat.com> wrote:

> Index: gdb/ChangeLog
> from  Alexandre Oliva  <aoliva@redhat.com>
> 	* coffread.c: Remove redundant static declarations.  Replace
> 	occurrences of `PTR' with `void *'.
> 	* elfread.c, mdebugread.c, minsyms.c, mipsread.c: Likewise.
> 	* top.h (quit_cover): Likewise.
> 	* defs.h (catch_errors): Likewise.

Ping?

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Elena Zannoni]

OK, I think this is fine now, but you forgot one PTR in defs.h, see
below.

Check it in, thanks.

Elena


Alexandre Oliva writes:
 > On Jun 11, 2001, Elena Zannoni <ezannoni@cygnus.com> wrote:
 > 
 > > Alexandre Oliva writes:
 > >> On Jun  6, 2001, Elena Zannoni <ezannoni@cygnus.com> wrote:
 > >> 
 > >> > Alex, no. Sorry, but we are trying to get rid of PTR in the
 > >> > gdb sources [as Kevin pointed out in another message].
 > >> 
 > >> Sorry.
 > >> 
 > >> > Could you 'revert' these changes by changing the definitions to 
 > >> > use void * and deleting the prototypes when they are not needed (everyone
 > >> > of these can go, except for free_elfinfo, I think).
 > >> 
 > >> Fixed as follows.  Not tested on all relevant architectures.  I'd do
 > >> that after next Monday's GCC snapshot, since I've been building GDB
 > >> along with GDB lately.  Ok to install?
 > 
 > > Yes, but with the memset calls changes, i.e. remove the casts.
 > 
 > >> Index: gdb/ChangeLog
 > >> from  Alexandre Oliva  <aoliva@redhat.com>
 > >> 
 > >> * coffread.c: Remove redundant static declarations.  Replace
 > >> occurrences of `PTR' with `void *'.
 > >> * elfread.c, mdebugread.c, minsyms.c, mipsread.c: Likewise.
 > 
 > Ok, here's an updated patch, that applies cleanly into mainline and
 > 5.2 branch, and that adds two more PTR->void* fixes that slipped in
 > after my original clean-up.  Oh, and that removes the (void *) casts
 > in memset calls that I had introduced in my previous patch.  Ok to
 > install in 5.2 and mainline?  With this and the other 2 patches I
 > posted today (for gdb/defs.h and expect), gdb 5.2 builds successfully
 > with OSF 4.0's cc.
 > 
 > Index: gdb/ChangeLog
 > from  Alexandre Oliva  <aoliva@redhat.com>
 > 	* coffread.c: Remove redundant static declarations.  Replace
 > 	occurrences of `PTR' with `void *'.
 > 	* elfread.c, mdebugread.c, minsyms.c, mipsread.c: Likewise.
 > 	* top.h (quit_cover): Likewise.
 > 	* defs.h (catch_errors): Likewise.
 > 

 > Index: gdb/defs.h
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/defs.h,v
 > retrieving revision 1.81
 > diff -u -p -r1.81 defs.h
 > --- gdb/defs.h 2002/02/26 03:29:55 1.81
 > +++ gdb/defs.h 2002/03/07 00:23:25
 > @@ -1152,7 +1152,7 @@ extern int catch_exceptions (struct ui_o
 >     This function is superseeded by catch_exceptions().  */
 >  
 >  typedef int (catch_errors_ftype) (PTR);
                                      ^^^^

 > -extern int catch_errors (catch_errors_ftype *, PTR, char *, return_mask);
 > +extern int catch_errors (catch_errors_ftype *, void *, char *, return_mask);
 >  
 >  /* Template to catch_errors() that wraps calls to command
 >     functions. */
 > 
 > -- 
 > Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
 > Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
 > CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
 > Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On Mar 19, 2002, Elena Zannoni <ezannoni@redhat.com> wrote:

> OK, I think this is fine now, but you forgot one PTR in defs.h, see
> below.

It was not forgotten, but rather intentionally left in.  Since some
compilers are picky about differences between PTR and void*, and there
still are a number of function definitions and declarations that used
PTR, I thought I wouldn't take such a bold step at this time.

> Check it in, thanks.

Mainline and 5.2 branch?

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Elena Zannoni]

Alexandre Oliva writes:
 > On Mar 19, 2002, Elena Zannoni <ezannoni@redhat.com> wrote:
 > 
 > > OK, I think this is fine now, but you forgot one PTR in defs.h, see
 > > below.
 > 
 > It was not forgotten, but rather intentionally left in.  Since some
 > compilers are picky about differences between PTR and void*, and there
 > still are a number of function definitions and declarations that used
 > PTR, I thought I wouldn't take such a bold step at this time.
 > 

Ok, then. 

 > > Check it in, thanks.
 > 
 > Mainline and 5.2 branch?
 > 

I think build failures fixes qualify for 5.2. Is there a bug report
for this?

Elena

 > -- 
 > Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
 > Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
 > CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
 > Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On Mar 19, 2002, Elena Zannoni <ezannoni@redhat.com> wrote:

>> > Check it in, thanks.

>> Mainline and 5.2 branch?

> I think build failures fixes qualify for 5.2.

Great.

> Is there a bug report for this?

Other than the one that came with the patch posted to this mailing
list?  I don't think so?  Should I create one, install the patch and
then close it?  I'm not aware of such GDB policies :-(

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Andrew Cagney]

> 
> Great.
> 
> 
>> Is there a bug report for this?
> 
> 
> Other than the one that came with the patch posted to this mailing
> list?  I don't think so?  Should I create one, install the patch and
> then close it?

I counted 4 or 5!  You should look at gdb/390, there may still be 
``issues''.

 > I'm not aware of such GDB policies :-(

I think people doing this are viewed in a positive light (however there 
isn't any strict requirement :-).

The rewritten web page reads:

> GDB has a Bug Database. It is used to track bugs (Problem Reports or PRs) and enhancements (Change Requests or CRs). In addition to problems encountered when running GDB, bugs include: errors or missing documentation; missing test cases; web changes; work in progress; dummy reports for fixed but unreported bugs; entries of not-a-bugs; and even adminstrivia.

Andrew

Re: GDB CVS won't build on OSF4.0's cc

[Elena Zannoni]

Andrew Cagney writes:
 > > 
 > > Great.
 > > 
 > > 
 > >> Is there a bug report for this?
 > > 
 > > 
 > > Other than the one that came with the patch posted to this mailing
 > > list?  I don't think so?  Should I create one, install the patch and
 > > then close it?
 > 
 > I counted 4 or 5!  You should look at gdb/390, there may still be 
 > ``issues''.

There is also 294 (which is essentially a duplicate of 390).  Alex,
can you take a look? Maybe your configury test needs to be made a bit
more sophisticated.
http://sources.redhat.com/cgi-bin/gnatsweb.pl?database=gdb

Elena


 > 
 >  > I'm not aware of such GDB policies :-(
 > 
 > I think people doing this are viewed in a positive light (however there 
 > isn't any strict requirement :-).
 > 
 > The rewritten web page reads:
 > 
 > > GDB has a Bug Database. It is used to track bugs (Problem Reports or PRs) and enhancements (Change Requests or CRs). In addition to problems encountered when running GDB, bugs include: errors or missing documentation; missing test cases; web changes; work in progress; dummy reports for fixed but unreported bugs; entries of not-a-bugs; and even adminstrivia.
 > 
 > Andrew
 > 

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On Mar 20, 2002, Elena Zannoni <ezannoni@redhat.com> wrote:

> Andrew Cagney writes:

>> > Other than the one that came with the patch posted to this mailing
>> > list?  I don't think so?  Should I create one, install the patch and
>> > then close it?
>> 
>> I counted 4 or 5!  You should look at gdb/390, there may still be 
>> ``issues''.

> There is also 294 (which is essentially a duplicate of 390).  Alex,
> can you take a look? Maybe your configury test needs to be made a bit
> more sophisticated.

This is a totally different subject.  This patch is for a build error
on OSF1/DU/Tru64.  Both PRs you mention are about TOC problems on
AIX.

>> > I'm not aware of such GDB policies :-(

>> I think people doing this are viewed in a positive light (however there 
>> isn't any strict requirement :-).

I never thought such a simple patch would take so long to get in (my
fault; I dropped the ball in revising the patch as requested for a
long while).  Anyway, when a patch is ready, I see little value in
creating a PR.  To me, it feels like pure overhead.  If you have a
different opinion, please enlighten me :-)

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Elena Zannoni]

Alexandre Oliva writes:
 > On Mar 20, 2002, Elena Zannoni <ezannoni@redhat.com> wrote:
 > 
 > > Andrew Cagney writes:
 > 
 > >> > Other than the one that came with the patch posted to this mailing
 > >> > list?  I don't think so?  Should I create one, install the patch and
 > >> > then close it?
 > >> 
 > >> I counted 4 or 5!  You should look at gdb/390, there may still be 
 > >> ``issues''.
 > 
 > > There is also 294 (which is essentially a duplicate of 390).  Alex,
 > > can you take a look? Maybe your configury test needs to be made a bit
 > > more sophisticated.
 > 
 > This is a totally different subject.  This patch is for a build error
 > on OSF1/DU/Tru64.  Both PRs you mention are about TOC problems on
 > AIX.


Sorry! Both Andrew and I were thinking of this other thread:
http://sources.redhat.com/ml/gdb/2002-02/msg00059.html

 > 
 > >> > I'm not aware of such GDB policies :-(
 > 
 > >> I think people doing this are viewed in a positive light (however there 
 > >> isn't any strict requirement :-).
 > 
 > I never thought such a simple patch would take so long to get in (my
 > fault; I dropped the ball in revising the patch as requested for a
 > long while).  Anyway, when a patch is ready, I see little value in
 > creating a PR.  To me, it feels like pure overhead.  If you have a
 > different opinion, please enlighten me :-)
 > 

Are any of the priority==high bugs related to this (as opposed to the
minimal-toc) problem?

Elena


 > -- 
 > Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
 > Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
 > CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
 > Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On Mar 19, 2002, Andrew Cagney <ac131313@cygnus.com> wrote:

> I counted 4 or 5!

bug 324 can be closed.  bug 183 is unrelated with the problem I
fixed, and I don't have access to Tru64 5.1 AFAIK to check.  I
couldn't find other related problems.

Oh, and don't forget there's that other patch for Insight to build on
OSF that is yet to be reviewed.

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Elena Zannoni]

Alexandre Oliva writes:
 > On Mar 19, 2002, Andrew Cagney <ac131313@cygnus.com> wrote:
 > 
 > > I counted 4 or 5!
 > 
 > bug 324 can be closed.  bug 183 is unrelated with the problem I
 > fixed, and I don't have access to Tru64 5.1 AFAIK to check.  I
 > couldn't find other related problems.
 > 

If your patch fixes 324, I guess you can just close it, and mention the PR
number in the changelog entry.

Elena


 > Oh, and don't forget there's that other patch for Insight to build on
 > OSF that is yet to be reviewed.
 > 
 > -- 
 > Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
 > Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
 > CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
 > Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On Mar 20, 2002, Elena Zannoni <ezannoni@redhat.com> wrote:

> If your patch fixes 324, I guess you can just close it, and mention
> the PR number in the changelog entry.

I'd need write access to the GNATS GDB database to do so.

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On Mar 19, 2002, Elena Zannoni <ezannoni@redhat.com> wrote:

> I think build failures fixes qualify for 5.2.

I'm finally I'm installing this in the branch too:

Index: gdb/ChangeLog
from  Alexandre Oliva  <aoliva@redhat.com>
	* coffread.c: Remove redundant static declarations.  Replace
	occurrences of `PTR' with `void *'.
	* elfread.c, mdebugread.c, minsyms.c, mipsread.c: Likewise.
	* top.h (quit_cover): Likewise.
	* defs.h (catch_errors): Likewise.

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: GDB CVS won't build on OSF4.0's cc

[Kevin Buettner]

On May 29,  7:19am, Alexandre Oliva wrote:

> 	* symfile.c (compare_psymbols, compare_symbols): Declare using
> 	PTR, as in the definition.
> 
> Index: gdb/symfile.c
> ===================================================================
> RCS file: /cvs/src/src/gdb/symfile.c,v
> retrieving revision 1.32
> diff -u -p -r1.32 symfile.c
> --- gdb/symfile.c 2001/05/10 15:33:21 1.32
> +++ gdb/symfile.c 2001/05/29 10:18:16
> @@ -117,9 +117,9 @@ static void add_shared_symbol_files_comm
>  
>  static void cashier_psymtab (struct partial_symtab *);
>  
> -static int compare_psymbols (const void *, const void *);
> +static int compare_psymbols (const PTR, const PTR);
>  
> -static int compare_symbols (const void *, const void *);
> +static int compare_symbols (const PTR, const PTR);
>  
>  bfd *symfile_bfd_open (char *);

I'm not the maintainer of the file in question, but one of our long
term goals has been the elimination of PTR from the gdb sources.  I
think the fix should have been to change PTR in the definition to
void * rather than change the declaration to match the definition.

Kevin

Re: GDB CVS won't build on OSF4.0's cc

[Alexandre Oliva]

On May 29, 2001, Kevin Buettner <kevinb@cygnus.com> wrote:

> I think the fix should have been to change PTR in the definition to
> void * rather than change the declaration to match the definition.

Oh, well.  Sorry about that.  I'll try to keep that in mind next time
I stumble across this kind of problem.

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist    *Please* write to mailing lists, not to me