[Bug bootstrap/45381] [4.6 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[Bug bootstrap/45381] [4.6 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

Richard Guenther <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P4

[Bug bootstrap/45381] [4.6 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[rth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

--- Comment #6 from Richard Henderson <rth at gcc dot gnu.org> 2010-10-01 22:00:47 UTC ---
(In reply to comment #5)
> I think altivec should disabled with "gcc version 4.0.1 (Apple Inc. build
> 5493)". Otherwise this pr could be closed as wontfix.

I'd prefer it if you provided some amount of detail about *what*
fails with the patch attached to this PR, rather than just reporting
the bare fact of not working.

[Bug bootstrap/45381] [4.6 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

--- Comment #7 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2010-10-01 23:33:03 UTC ---
(In reply to comment #6)
> I'd prefer it if you provided some amount of detail about *what*
> fails with the patch attached to this PR, rather than just reporting
> the bare fact of not working.

gcc  -I../../gcc-4.6-work/libcpp -I. -I../../gcc-4.6-work/libcpp/../include
-I../../gcc-4.6-work/libcpp/include  -g -fkeep-inline-functions -W -Wall
-Wwrite-strings -Wmissing-format-attribute -Wstrict-prototypes
-Wmissing-prototypes -Wold-style-definition -pedantic -Wno-long-long 
-I../../gcc-4.6-work/libcpp -I. -I../../gcc-4.6-work/libcpp/../include
-I../../gcc-4.6-work/libcpp/include  -c -o lex.o -MT lex.o -MMD -MP -MF
.deps/lex.Tpo ../../gcc-4.6-work/libcpp/lex.c
../../gcc-4.6-work/libcpp/lex.c:789: error: conflicting types for
'warn_in_comment'
../../gcc-4.6-work/libcpp/lex.c:62: error: previous declaration of
'warn_in_comment' was here
../../gcc-4.6-work/libcpp/lex.c: In function 'warn_in_comment':
../../gcc-4.6-work/libcpp/lex.c:796: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:801: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:810: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:811: warning: control reaches end of non-void
function
../../gcc-4.6-work/libcpp/lex.c: In function '_cpp_process_line_notes':
../../gcc-4.6-work/libcpp/lex.c:851: error: invalid operands to binary ||
../../gcc-4.6-work/libcpp/lex.c: At top level:
../../gcc-4.6-work/libcpp/lex.c:883: error: conflicting types for
'_cpp_skip_block_comment'
../../gcc-4.6-work/libcpp/internal.h:596: error: previous declaration of
'_cpp_skip_block_comment' was here
../../gcc-4.6-work/libcpp/lex.c: In function '_cpp_skip_block_comment':
../../gcc-4.6-work/libcpp/lex.c:922: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:934: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:935: warning: control reaches end of non-void
function
../../gcc-4.6-work/libcpp/lex.c: In function 'skip_whitespace':
../../gcc-4.6-work/libcpp/lex.c:958: error: incompatible types in
initialization
../../gcc-4.6-work/libcpp/lex.c:967: error: incompatible types in assignment
../../gcc-4.6-work/libcpp/lex.c:979: error: invalid operands to binary !=
../../gcc-4.6-work/libcpp/lex.c: In function 'forms_identifier_p':
../../gcc-4.6-work/libcpp/lex.c:1035: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:1044: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:1055: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:1059: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:1060: warning: control reaches end of non-void
function
../../gcc-4.6-work/libcpp/lex.c: In function 'lex_identifier':
../../gcc-4.6-work/libcpp/lex.c:1130: error: wrong type argument to unary
exclamation mark
../../gcc-4.6-work/libcpp/lex.c:1137: error: invalid operands to binary !=
../../gcc-4.6-work/libcpp/lex.c:1146: error: invalid operands to binary !=
../../gcc-4.6-work/libcpp/lex.c: In function 'lex_number':
../../gcc-4.6-work/libcpp/lex.c:1209: error: invalid operands to binary !=
../../gcc-4.6-work/libcpp/lex.c: In function 'lex_string':
../../gcc-4.6-work/libcpp/lex.c:1523: error: incompatible types in
initialization
../../gcc-4.6-work/libcpp/lex.c:1579: error: incompatible types in assignment
../../gcc-4.6-work/libcpp/lex.c:1582: error: invalid operands to binary !=
../../gcc-4.6-work/libcpp/lex.c: At top level:
../../gcc-4.6-work/libcpp/lex.c:1863: error: conflicting types for
'_cpp_get_fresh_line'
../../gcc-4.6-work/libcpp/internal.h:595: error: previous declaration of
'_cpp_get_fresh_line' was here
../../gcc-4.6-work/libcpp/lex.c: In function '_cpp_get_fresh_line':
../../gcc-4.6-work/libcpp/lex.c:1868: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:1875: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:1880: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:1885: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c:1899: error: incompatible types in return
../../gcc-4.6-work/libcpp/lex.c: In function '_cpp_lex_direct':
../../gcc-4.6-work/libcpp/lex.c:1944: error: wrong type argument to unary
exclamation mark
../../gcc-4.6-work/libcpp/lex.c:2044: error: incompatible type for argument 3
of 'lex_identifier'
../../gcc-4.6-work/libcpp/lex.c:2068: error: invalid operands to binary !=
../../gcc-4.6-work/libcpp/lex.c:2270: error: invalid operands to binary !=
../../gcc-4.6-work/libcpp/lex.c:2273: error: incompatible type for argument 3
of 'lex_identifier'
../../gcc-4.6-work/libcpp/lex.c: At top level:
../../gcc-4.6-work/libcpp/lex.c:2356: error: conflicting types for
'cpp_spell_token'
../../gcc-4.6-work/libcpp/include/cpplib.h:723: error: previous declaration of
'cpp_spell_token' was here
../../gcc-4.6-work/libcpp/lex.c: In function 'cpp_spell_token':
../../gcc-4.6-work/libcpp/lex.c:2378: error: invalid operands to binary !=
../../gcc-4.6-work/libcpp/lex.c: In function 'cpp_token_as_text':
../../gcc-4.6-work/libcpp/lex.c:2422: error: incompatible type for argument 4
of 'cpp_spell_token'
../../gcc-4.6-work/libcpp/lex.c: In function 'cpp_output_line_to_string':
../../gcc-4.6-work/libcpp/lex.c:2634: error: incompatible type for argument 4
of 'cpp_spell_token'

[Bug bootstrap/45381] [4.6 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[iains at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

Iain Sandoe <iains at gcc dot gnu.org> changed:

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

--- Comment #8 from Iain Sandoe <iains at gcc dot gnu.org> 2011-02-08 12:17:28 UTC ---
what was the reason that the patch at comment #1 was not acceptable?

If the builtins were not available prior to 4.0.5, then the caveat seems
sensible ... I must be missing something ;)

===

if not, for all darwin 8 & 9 the system compiler is 4.0.1 
(3.3 for darwin 7, and I doubt trunk builds for earlier than that)

4.2.1 is not available on darwin 8 (so I'd imagine that bootstrap is broken
there without a ready workaround).  I don't have a ppc darwin 8 box available
right now - although I will try to crank one up before 4.6 forks.

I could not find a permutation of installed headers that gave the right result
- perhaps looking in the wrong place.

Do we want to pursue this further - or just accept a slower solution for the
(legacy) darwin versions?

[Bug bootstrap/45381] [4.6 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[rth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

--- Comment #9 from Richard Henderson <rth at gcc dot gnu.org> 2011-02-08 16:17:10 UTC ---
I don't think we really want to pursue this further.

I had hoped to find a solution that satisfied the older apple compiler -- it
is supposed to support altivec after all -- but it remains a mystery what is
broken with that compiler.

[Bug bootstrap/45381] [4.6 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[iains at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

--- Comment #10 from Iain Sandoe <iains at gcc dot gnu.org> 2011-02-08 16:44:15 UTC ---
(In reply to comment #9)
> I don't think we really want to pursue this further.
> 
> I had hoped to find a solution that satisfied the older apple compiler -- it
> is supposed to support altivec after all -- but it remains a mystery what is
> broken with that compiler.

OK, that's fair 
- so is the solution to apply (an updated) patch like comment #1  ?
- or must earlier darwin systems do that locally?

(I'm not clear whether the limitation of 4.0.5 applies generally or only to the
Apple Local versions)

[Bug bootstrap/45381] [4.6 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[rth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

--- Comment #11 from Richard Henderson <rth at gcc dot gnu.org> 2011-02-08 16:52:41 UTC ---
Yes, a patch like in #1 would be fine.

[Bug bootstrap/45381] [4.6 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[iains at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

--- Comment #12 from Iain Sandoe <iains at gcc dot gnu.org> 2011-02-08 21:25:05 UTC ---
(In reply to comment #11)
> Yes, a patch like in #1 would be fine.

I will produce one in due course (unless Dominique beats me to it...)

-- the hack below is purely to satisfy curiosity  :-)
-- I doubt it is worth unravelling how to make it work in reality..

The hack produces a working stage1 compiler built from gcc-4.0.1 or gcc-4.2.1

it seems that gcc-4.0.1 does not recognize __attribute__((altivec(vector))) and
altivec.h does not indeed define the functions it complains about...

---

Index: libcpp/lex.c
===================================================================
--- libcpp/lex.c        (revision 169914)
+++ libcpp/lex.c        (working copy)
@@ -21,6 +21,7 @@ along with this program; see the file COPYING3.  I
 <http://www.gnu.org/licenses/>.  */

 #include "config.h"
+#include <altivec.h>
 #include "system.h"
 #include "cpplib.h"
 #include "internal.h"
@@ -519,10 +520,11 @@ init_vectorized_lexer (void)
    so we can't compile this function without -maltivec on the command line
    (or implied by some other switch).  */

+
 static const uchar *
 search_line_fast (const uchar *s, const uchar *end ATTRIBUTE_UNUSED)
 {
-  typedef __attribute__((altivec(vector))) unsigned char vc;
+  typedef __vector unsigned char vc ;

   const vc repl_nl = {
     '\n', '\n', '\n', '\n', '\n', '\n', '\n', '\n', 
@@ -550,14 +552,14 @@ search_line_fast (const uchar *s, const uchar *end

   /* Altivec loads automatically mask addresses with -16.  This lets us
      issue the first load as early as possible.  */
-  data = __builtin_vec_ld(0, (const vc *)s);
+  data = vec_ld(0, (const vc *)s);

   /* Discard bytes before the beginning of the buffer.  Do this by
      beginning with all ones and shifting in zeros according to the
      mis-alignment.  The LVSR instruction pulls the exact shift we
      want from the address.  */
-  mask = __builtin_vec_lvsr(0, s);
-  mask = __builtin_vec_perm(zero, ones, mask);
+  mask = vec_lvsr(0, s);
+  mask = vec_perm(zero, ones, mask);
   data &= mask;

   /* While altivec loads mask addresses, we still need to align S so
@@ -571,20 +573,20 @@ search_line_fast (const uchar *s, const uchar *end
       vc m_nl, m_cr, m_bs, m_qm;

       s += 16;
-      data = __builtin_vec_ld(0, (const vc *)s);
+      data = vec_ld(0, (const vc *)s);

     start:
-      m_nl = (vc) __builtin_vec_cmpeq(data, repl_nl);
-      m_cr = (vc) __builtin_vec_cmpeq(data, repl_cr);
-      m_bs = (vc) __builtin_vec_cmpeq(data, repl_bs);
-      m_qm = (vc) __builtin_vec_cmpeq(data, repl_qm);
+      m_nl = (vc) vec_cmpeq(data, repl_nl);
+      m_cr = (vc) vec_cmpeq(data, repl_cr);
+      m_bs = (vc) vec_cmpeq(data, repl_bs);
+      m_qm = (vc) vec_cmpeq(data, repl_qm);
       t = (m_nl | m_cr) | (m_bs | m_qm);

       /* T now contains 0xff in bytes for which we matched one of the relevant
         characters.  We want to exit the loop if any byte in T is non-zero.
         Below is the expansion of vec_any_ne(t, zero).  */
     }
-  while (!__builtin_vec_vcmpeq_p(/*__CR6_LT_REV*/3, t, zero));
+  while (!vec_any_ne(t, zero));

   {
 #define N  (sizeof(vc) / sizeof(long))

[Bug bootstrap/45381] [4.6 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[iains at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

--- Comment #13 from Iain Sandoe <iains at gcc dot gnu.org> 2011-02-09 12:35:57 UTC ---
(In reply to comment #6)
> (In reply to comment #5)
> > I think altivec should disabled with "gcc version 4.0.1 (Apple Inc. build
> > 5493)". Otherwise this pr could be closed as wontfix.
> 
> I'd prefer it if you provided some amount of detail about *what*
> fails with the patch attached to this PR, rather than just reporting
> the bare fact of not working.

Sorry ... I should have read the audit trail better.
The fault is a very subtle header-ordering 'gotcha'...

[At least on Darwin] One cannot include external headers after system.h because
of the fact that system.h re-defines bool.

... the reason that comment #12 works (which is essentially the same thing as
your attachment) is that <altivec.h> is included before system.h.

I guess on ppc64-linux there is no difference in the size of bool - and I don't
wish to re-open the debate about Darwin's way of doing it ('tis out of our
control in any event).

I still don't know if it's worth doing any more than below, for 4.6:

Index: libcpp/lex.c
===================================================================
--- libcpp/lex.c        (revision 169914)
+++ libcpp/lex.c        (working copy)
@@ -512,7 +513,7 @@ init_vectorized_lexer (void)
   search_line_fast = impl;
 }

-#elif defined(__GNUC__) && defined(__ALTIVEC__)
+#elif defined(__GNUC__) && (GCC_VERSION >= 4005) && defined(__ALTIVEC__)

 /* A vection of the fast scanner using AltiVec vectorized byte compares.  */
 /* ??? Unfortunately, attribute(target("altivec")) is not yet supported,

[Bug bootstrap/45381] [4.6 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

--- Comment #14 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2011-02-15 21:06:59 UTC ---
Note that I did not have any problem to compile libcpp/lex.c on
powerpc-apple-darwin8 with
gcc version 4.0.1 (Apple Computer, Inc. build 5370):

...
gcc  -I../../work/libcpp -I. -I../../work/libcpp/../include
-I../../work/libcpp/include -I/sw/include -g -fkeep-inline-functions -W -Wall
-Wwrite-strings -Wmissing-format-attribute -Wstrict-prototypes
-Wmissing-prototypes -Wold-style-definition -pedantic -Wno-long-long 
-I../../work/libcpp -I. -I../../work/libcpp/../include
-I../../work/libcpp/include -I/sw/include -c -o lex.o -MT lex.o -MMD -MP -MF
.deps/lex.Tpo ../../work/libcpp/lex.c
...
/opt/gcc/build/./prev-gcc/xgcc -B/opt/gcc/build/./prev-gcc/
-B/opt/gcc/gcc4.6w/powerpc-apple-darwin8/bin/
-B/opt/gcc/gcc4.6w/powerpc-apple-darwin8/bin/
-B/opt/gcc/gcc4.6w/powerpc-apple-darwin8/lib/ -isystem
/opt/gcc/gcc4.6w/powerpc-apple-darwin8/include -isystem
/opt/gcc/gcc4.6w/powerpc-apple-darwin8/sys-include     -I../../work/libcpp -I.
-I../../work/libcpp/../include -I../../work/libcpp/include -I/sw/include -g -O2
-mdynamic-no-pic -gtoggle -W -Wall -Wwrite-strings -Wmissing-format-attribute
-Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition -Wc++-compat
-pedantic -Wno-long-long -Werror -I../../work/libcpp -I.
-I../../work/libcpp/../include -I../../work/libcpp/include -I/sw/include -c -o
lex.o -MT lex.o -MMD -MP -MF .deps/lex.Tpo ../../work/libcpp/lex.c

[Bug bootstrap/45381] [4.6 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[mrs at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

mrs@gcc.gnu.org <mrs at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |NEW
                 CC|                            |mrs at gcc dot gnu.org

--- Comment #15 from mrs at gcc dot gnu.org <mrs at gcc dot gnu.org> 2011-03-01 08:07:08 UTC ---
I don't think this is waiting on anything anymore.  It would be nice if someone
would check in the patch an unbreak the build, as it used to build.

[Bug bootstrap/45381] [4.6/4.7 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[Denis.Excoffier at airbus dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

Denis Excoffier <Denis.Excoffier at airbus dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |Denis.Excoffier at airbus
                   |                            |dot com

--- Comment #16 from Denis Excoffier <Denis.Excoffier at airbus dot com> 2011-03-17 21:16:40 UTC ---
Hello,

This one is still present for 4.6.0-RC-20110314, and the patch shown under
Comment #13 works perfectly for me.

Thank you.

[Bug bootstrap/45381] [4.6/4.7 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[rth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

--- Comment #17 from Richard Henderson <rth at gcc dot gnu.org> 2011-03-18 20:20:01 UTC ---
Author: rth
Date: Fri Mar 18 20:19:45 2011
New Revision: 171164

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=171164
Log:
PR bootstrap/45381
* lex.c [ALTIVEC] (search_line_fast): Require gcc version 4.5.

Modified:
    branches/gcc-4_6-branch/libcpp/ChangeLog
    branches/gcc-4_6-branch/libcpp/lex.c

[Bug bootstrap/45381] [4.6/4.7 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[rth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

--- Comment #18 from Richard Henderson <rth at gcc dot gnu.org> 2011-03-18 20:20:45 UTC ---
Author: rth
Date: Fri Mar 18 20:20:35 2011
New Revision: 171165

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=171165
Log:
PR bootstrap/45381
* lex.c [ALTIVEC] (search_line_fast): Require gcc version 4.5.

Modified:
    trunk/libcpp/ChangeLog
    trunk/libcpp/lex.c

[Bug bootstrap/45381] [4.6/4.7 Regression] Bootstrap failure for powerpc-apple-darwin9: error: AltiVec argument passed to unprototyped function

[rth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45381

Richard Henderson <rth at gcc dot gnu.org> changed:

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

--- Comment #19 from Richard Henderson <rth at gcc dot gnu.org> 2011-03-18 20:21:22 UTC ---
Patch committed.