[Bug rtl-optimization/33552] [4.3 Regression] wrong code for multiple output asm, wrong df?
--- Comment #16 from bonzini at gnu dot org 2007-09-27 06:28 --- I will implement something along the lines that Jakub discussed. In the meanwhile, could anybody figure a self-contained execution testcase based on comment #14? Thanks! -- bonzini at gnu dot org changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |bonzini at gnu dot org |dot org | Status|NEW |ASSIGNED Last reconfirmed|2007-09-25 18:57:01 |2007-09-27 06:28:00 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33552
[Bug rtl-optimization/33552] [4.3 Regression] wrong code for multiple output asm, wrong df?
--- Comment #17 from belyshev at depni dot sinp dot msu dot ru 2007-09-27
08:01 ---
(In reply to comment #16)
something like this (for gcc.target/i386):
/* { dg-do run } */
/* { dg-require-effective-target lp64 } */
/* { dg-options -O2} */
unsigned long a [100];
int main (void)
{
unsigned long v = 0x5UL, c = 13, low, high;
long j;
a [0] = 0x2UL;
a [1] = 0x3UL;
for (j = 0; j 100; j++)
{
asm (mulq %3
: =a (low), =d (high)
: 0 (a [j]), rm (v));
asm (addq %5,%q1\n\tadcq %3,%q0
: =r (c), =r (a [j])
: 0 (high), rme (0), 1 (low), rme (c));
}
if (a [0] != 13 || a [1] != 10 || a [2] != 15 || a [3] != 0)
__builtin_abort ();
return 0;
}
--
belyshev at depni dot sinp dot msu dot ru changed:
What|Removed |Added
CC||belyshev at depni dot sinp
||dot msu dot ru
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33552
[Bug fortran/33568] New: ICE with ANINT (with KIND and an array)
The following test fails: $ cat test.f90 PROGRAM Test IMPLICIT NONE INTEGER, PARAMETER :: DP=8 REAL(KIND=DP), DIMENSION(1:3) :: A A = ANINT ( A , DP ) END PROGRAM Test $ gfortran -c test.f90 test.f90: In function 'MAIN__': test.f90:1: internal compiler error: in gfc_trans_assignment_1, at fortran/trans-expr.c:4052 Please submit a full bug report, with preprocessed source if appropriate. See http://gcc.gnu.org/bugs.html for instructions. $ gfortran -v Using built-in specs. Target: i386-pc-linux-gnu Configured with: /home/fx/gfortran_nightbuild/trunk/configure --prefix=/home/fx/gfortran_nightbuild/irun-20070926 --enable-languages=c,fortran --build=i386-pc-linux-gnu --enable-checking=release --with-gmp=/home/fx/gfortran_nightbuild/software Thread model: posix gcc version 4.3.0 20070926 (experimental) [trunk revision 128777] (GCC) It compiles fine if A is a scalar (not an array) or if I don't specify the kind in the ANINT function. AINT fails as well, not so NINT, FLOOR or CEILING. -- Summary: ICE with ANINT (with KIND and an array) Product: gcc Version: 4.3.0 Status: UNCONFIRMED Severity: major Priority: P3 Component: fortran AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: jellby at yahoo dot com http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33568
[Bug rtl-optimization/33552] [4.3 Regression] wrong code for multiple output asm, wrong df?
--- Comment #18 from jakub at gcc dot gnu dot org 2007-09-27 08:12 ---
Sure, no problem.
/* PR rtl-optimization/33552 */
/* { dg-do run } */
/* { dg-options -O2 } */
extern void abort (void);
void
__attribute__((noinline))
foo (unsigned long *wp, unsigned long *up, long un, unsigned long *vp)
{
long j;
unsigned long prod_low, prod_high;
unsigned long cy_dig;
unsigned long v_limb;
v_limb = vp[0];
cy_dig = 64;
for (j = un; j 0; j--)
{
unsigned long u_limb, w_limb;
u_limb = *up++;
__asm__ (
: =r (prod_low), =r (prod_high)
: 0 (u_limb), 1 (v_limb));
__asm__ (mov %5, %1; add %5, %0
: =r (cy_dig), =r (w_limb)
: 0 (prod_high), rm (0), 1 (prod_low), rm (cy_dig));
*wp++ = w_limb;
}
}
int
main (void)
{
unsigned long wp[4];
unsigned long up[4] = { 0x1248, 0x248a, 0x1745, 0x1853 };
unsigned long vp = 0xdead;
foo (wp, up, 4, vp);
if (wp[0] != 0x40 || wp[1] != 0xdeed || wp[2] != 0x1bd9a || wp[3] != 0x29c47)
abort ();
return 0;
}
For dg.target/i386, unless we add a bunch of target variants for the only
asm arch dependent string in there.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33552
[Bug c++/33553] [4.1/4.2/4.3 Regression] Bogus array bound is not an integer constant for parameter in template method of template class
--- Comment #4 from jakub at gcc dot gnu dot org 2007-09-27 08:58 --- Introduced by PR28595 fix. http://gcc.gnu.org/viewcvs?root=gccview=revrev=116468 or for 4.1: http://gcc.gnu.org/viewcvs?root=gccview=revrev=116471 -- jakub at gcc dot gnu dot org changed: What|Removed |Added Known to fail|4.2.0 4.3.0 |4.1.2 4.2.0 4.3.0 Summary|[4.2/4.3 Regression] Bogus |[4.1/4.2/4.3 Regression] |array bound is not an |Bogus array bound is not an |integer constant for |integer constant for |parameter in template method|parameter in template method |of template class |of template class http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33553
[Bug tree-optimization/33562] [4.3 Regression] aggregate DSE disabled
--- Comment #2 from rguenther at suse dot de 2007-09-27 09:08 --- Subject: Re: [4.3 Regression] aggregate DSE disabled On Wed, 26 Sep 2007, pinskia at gcc dot gnu dot org wrote: --- Comment #1 from pinskia at gcc dot gnu dot org 2007-09-26 19:04 --- * gcc.dg/tree-ssa/complex-4.c: XFAIL. is a regression then because the testcase was added back in 2006-02-18 (by me). I did wonder what optimized that before... (maybe a separate bug for this is more appropriate) Richard. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33562
Re: [Bug tree-optimization/33562] [4.3 Regression] aggregate DSE disabled
On 27 Sep 2007 09:08:17 -, rguenther at suse dot de [EMAIL PROTECTED] wrote: I did wonder what optimized that before... (maybe a separate bug for this is more appropriate) Must_def cause the optimization to work IIRC. In fact this is the reason why aggregate DSE was added was specifically to fix this testcase. See http://gcc.gnu.org/ml/gcc-patches/2006-05/msg01115.html which specifically mentions this. I am still trying to understand why we removed must_def anyways. Everything points to the removal of must_def caused many different regressions. Thanks, Andrew Pinski
[Bug tree-optimization/33562] [4.3 Regression] aggregate DSE disabled
--- Comment #3 from pinskia at gmail dot com 2007-09-27 09:19 --- Subject: Re: [4.3 Regression] aggregate DSE disabled On 27 Sep 2007 09:08:17 -, rguenther at suse dot de [EMAIL PROTECTED] wrote: I did wonder what optimized that before... (maybe a separate bug for this is more appropriate) Must_def cause the optimization to work IIRC. In fact this is the reason why aggregate DSE was added was specifically to fix this testcase. See http://gcc.gnu.org/ml/gcc-patches/2006-05/msg01115.html which specifically mentions this. I am still trying to understand why we removed must_def anyways. Everything points to the removal of must_def caused many different regressions. Thanks, Andrew Pinski -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33562
[Bug c++/33506] [4.2 Regression] TYPE_RAISES_EXCEPTIONS dumped with attributes
--- Comment #2 from jakub at gcc dot gnu dot org 2007-09-27 09:20 --- Fixed on the trunk. -- jakub at gcc dot gnu dot org changed: What|Removed |Added Known to fail||4.2.0 Known to work||4.3.0 Summary|[4.2/4.3 Regression]|[4.2 Regression] |TYPE_RAISES_EXCEPTIONS |TYPE_RAISES_EXCEPTIONS |dumped with attributes |dumped with attributes http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33506
[Bug c++/33493] [4.1/4.2/4.3 regression] Broken diagnostic: 'dl_expr' not supported by dump_expr
--- Comment #1 from paolo at gcc dot gnu dot org 2007-09-27 09:21 --- Subject: Bug 33493 Author: paolo Date: Thu Sep 27 09:20:59 2007 New Revision: 128831 URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=128831 Log: /cp 2007-09-27 Paolo Carlini [EMAIL PROTECTED] PR c++/33493 * error.c (dump_expr): Deal with DELETE_EXPR and VEC_DELETE_EXPR. * cxx-pretty-print.c (pp_cxx_delete_expression): Add missing spaces in the formatting. * cxx-pretty-print.h (pp_cxx_delete_expression): Declare. /testsuite 2007-09-27 Paolo Carlini [EMAIL PROTECTED] PR c++/33493 * g++.dg/template/error31.C: New. Added: trunk/gcc/testsuite/g++.dg/template/error31.C Modified: trunk/gcc/cp/ChangeLog trunk/gcc/cp/cxx-pretty-print.c trunk/gcc/cp/cxx-pretty-print.h trunk/gcc/cp/error.c trunk/gcc/testsuite/ChangeLog -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33493
[Bug c++/33493] [4.1/4.2 regression] Broken diagnostic: 'dl_expr' not supported by dump_expr
--- Comment #2 from pcarlini at suse dot de 2007-09-27 09:24 --- Fixed in mainline. -- pcarlini at suse dot de changed: What|Removed |Added AssignedTo|pcarlini at suse dot de |unassigned at gcc dot gnu ||dot org Status|ASSIGNED|NEW Summary|[4.1/4.2/4.3 regression]|[4.1/4.2 regression] Broken |Broken diagnostic: 'dl_expr'|diagnostic: 'dl_expr' not |not supported by dump_expr |supported by dump_expr http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33493
[Bug c++/33501] [4.1/4.2/4.3 regression] Copy constructor assumed to exist for undefined class
--- Comment #2 from jakub at gcc dot gnu dot org 2007-09-27 09:41 --- Most probably introduced by http://gcc.gnu.org/ml/gcc-patches/2004-07/msg00902.html -- jakub at gcc dot gnu dot org changed: What|Removed |Added CC||nathan at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33501
[Bug c++/33569] New: A bug in inlining static const members.
Linkage of the following example reports an undefined reference to `A::anInt'.
example.cpp
struct A
{
templatetypename T
void operator%( T const object ) {}
static int const anInt = 5;
};
int main( void )
{
A() % A::anInt;
return 0;
}
--
Summary: A bug in inlining static const members.
Product: gcc
Version: 4.1.2
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c++
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: juraj dot ivancic at gmail dot com
GCC build triplet: i686 i386 GNU/Linux
GCC host triplet: i686 i386 GNU/Linux
GCC target triplet: i686 i386 GNU/Linux
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33569
[Bug c++/33569] A bug in inlining static const members.
--- Comment #1 from pcarlini at suse dot de 2007-09-27 10:10 --- Indeed, in general you have to add: const int A::anInt; At high optimization levels the the static int is inlined but this is just implementation defined behavior... -- pcarlini at suse dot de changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||INVALID http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33569
[Bug tree-optimization/33562] [4.3 Regression] aggregate DSE disabled
--- Comment #4 from rguenth at gcc dot gnu dot org 2007-09-27 10:17 ---
Created an attachment (id=14254)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14254action=view)
restore DCE of killing defs
some ssa updating is broken in dce though:
/space/rguenther/src/svn/pointer_plus/gcc/testsuite/gcc.dg/tree-ssa/complex-4.c:
In function 'f':
/space/rguenther/src/svn/pointer_plus/gcc/testsuite/gcc.dg/tree-ssa/complex-4.c:5:
error: type mismatch between an SSA_NAME and its symbol
/space/rguenther/src/svn/pointer_plus/gcc/testsuite/gcc.dg/tree-ssa/complex-4.c:5:
error: in statement
# SFT.1_12 = VDEF SFT.1_11 { SFT.1 }
REALPART_EXPR t = 2;
/space/rguenther/src/svn/pointer_plus/gcc/testsuite/gcc.dg/tree-ssa/complex-4.c:5:
internal compiler error: verify_ssa failed
Please submit a full bug report,
with preprocessed source if appropriate.
See http://gcc.gnu.org/bugs.html for instructions.
but dceloop now removes the dead store and we get:
f ()
{
complex int t;
int D.1551;
bb 2:
# SFT.1_12 = VDEF SFT.1_11
REALPART_EXPR t = 2;
# SFT.0_13 = VDEF SFT.0_10
IMAGPART_EXPR t = 2;
# SFT.0_14 = VDEF SFT.0_13
# SFT.1_15 = VDEF SFT.1_12
D.1551_5 = g (t);
return D.1551_5;
}
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33562
[Bug fortran/33568] ICE with ANINT (with KIND and an array)
--- Comment #1 from pault at gcc dot gnu dot org 2007-09-27 10:39 ---
Confirmed. Thanks, Ignacio.
This is the fix (the absence of the KIND working and the regression wrt 4.2
were giveaways):
Index: gcc/fortran/trans-intrinsic.c
===
*** gcc/fortran/trans-intrinsic.c (révision 128323)
--- gcc/fortran/trans-intrinsic.c (copie de travail)
*** gfc_conv_intrinsic_aint (gfc_se * se, gf
*** 393,399
{
tree type;
tree itype;
! tree arg;
tree tmp;
tree cond;
mpfr_t huge;
--- 393,399
{
tree type;
tree itype;
! tree arg[2];
tree tmp;
tree cond;
mpfr_t huge;
*** gfc_conv_intrinsic_aint (gfc_se * se, gf
*** 448,467
/* Evaluate the argument. */
gcc_assert (expr-value.function.actual-expr);
! gfc_conv_intrinsic_function_args (se, expr, arg, 1);
/* Use a builtin function if one exists. */
if (n != END_BUILTINS)
{
tmp = built_in_decls[n];
! se-expr = build_call_expr (tmp, 1, arg);
return;
}
/* This code is probably redundant, but we'll keep it lying around just
in case. */
type = gfc_typenode_for_spec (expr-ts);
! arg = gfc_evaluate_now (arg, se-pre);
/* Test if the value is too large to handle sensibly. */
gfc_set_model_kind (kind);
--- 448,467
/* Evaluate the argument. */
gcc_assert (expr-value.function.actual-expr);
! gfc_conv_intrinsic_function_args (se, expr, arg, 2);
/* Use a builtin function if one exists. */
if (n != END_BUILTINS)
{
tmp = built_in_decls[n];
! se-expr = build_call_expr (tmp, 1, arg[0]);
return;
}
/* This code is probably redundant, but we'll keep it lying around just
in case. */
type = gfc_typenode_for_spec (expr-ts);
! arg[0] = gfc_evaluate_now (arg[0], se-pre);
/* Test if the value is too large to handle sensibly. */
gfc_set_model_kind (kind);
*** gfc_conv_intrinsic_aint (gfc_se * se, gf
*** 469,485
n = gfc_validate_kind (BT_INTEGER, kind, false);
mpfr_set_z (huge, gfc_integer_kinds[n].huge, GFC_RND_MODE);
tmp = gfc_conv_mpfr_to_tree (huge, kind);
! cond = build2 (LT_EXPR, boolean_type_node, arg, tmp);
mpfr_neg (huge, huge, GFC_RND_MODE);
tmp = gfc_conv_mpfr_to_tree (huge, kind);
! tmp = build2 (GT_EXPR, boolean_type_node, arg, tmp);
cond = build2 (TRUTH_AND_EXPR, boolean_type_node, cond, tmp);
itype = gfc_get_int_type (kind);
! tmp = build_fix_expr (se-pre, arg, itype, op);
tmp = convert (type, tmp);
! se-expr = build3 (COND_EXPR, type, cond, tmp, arg);
mpfr_clear (huge);
}
--- 469,485
n = gfc_validate_kind (BT_INTEGER, kind, false);
mpfr_set_z (huge, gfc_integer_kinds[n].huge, GFC_RND_MODE);
tmp = gfc_conv_mpfr_to_tree (huge, kind);
! cond = build2 (LT_EXPR, boolean_type_node, arg[0], tmp);
mpfr_neg (huge, huge, GFC_RND_MODE);
tmp = gfc_conv_mpfr_to_tree (huge, kind);
! tmp = build2 (GT_EXPR, boolean_type_node, arg[0], tmp);
cond = build2 (TRUTH_AND_EXPR, boolean_type_node, cond, tmp);
itype = gfc_get_int_type (kind);
! tmp = build_fix_expr (se-pre, arg[0], itype, op);
tmp = convert (type, tmp);
! se-expr = build3 (COND_EXPR, type, cond, tmp, arg[0]);
mpfr_clear (huge);
}
Cheers
Paul
--
pault at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNCONFIRMED |NEW
Ever Confirmed|0 |1
Last reconfirmed|-00-00 00:00:00 |2007-09-27 10:39:02
date||
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33568
[Bug java/33570] New: Just tried to compile a java source code with gcc java
[EMAIL PROTECTED]:[Receiver]$ gcc -v
Using built-in specs.
Target: i486-linux-gnu
Configured with: ../src/configure -v
--enable-languages=c,c++,fortran,objc,obj-c++,treelang --prefix=/usr
--enable-shared --with-system-zlib --libexecdir=/usr/lib
--without-included-gettext --enable-threads=posix --enable-nls
--program-suffix=-4.1 --enable-__cxa_atexit --enable-clocale=gnu
--enable-libstdcxx-debug --enable-mpfr --with-tune=i686
--enable-checking=release i486-linux-gnu
Thread model: posix
gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)
[EMAIL PROTECTED]:[Receiver]$ uname -a
Linux localhost 2.6.18-4-686 #1 SMP Mon Mar 26 17:17:36 UTC 2007 i686 GNU/Linux
[EMAIL PROTECTED]:[Receiver]$ gcc Receiver.java
Receiver.java:5: internal compiler error: in uses_jv_markobj_p, at
java/boehm.c:245
Please submit a full bug report,
with preprocessed source if appropriate.
See URL:http://gcc.gnu.org/bugs.html for instructions.
For Debian GNU/Linux specific bug reporting instructions,
see URL:file:///usr/share/doc/gcc-4.1/README.Bugs.
import java.io.*;
import java.util.*;
import java.net.*;
class DataFrame {
byte type; // 1 - data, 2 - ack.
byte num; // 0, 1.
byte dataLen;
byte[] data;
static byte[] frameDataToByte(byte type, byte num, byte[] data) {
byte[] tmp = new byte[3 + data.length];
tmp[0] = type;
tmp[1] = num;
tmp[2] = (byte)data.length;
for(int i = 0; i data.length; i++)
tmp[3 + i] = data[i];
return tmp;
}
static DataFrame byteToFrame(byte[] data) {
DataFrame d = new DataFrame();
d.type = data[0];
d.num = data[1];
d.dataLen = data[2];
d.data = new byte[d.dataLen];
for(int i = 0; i data.length; i++)
d.data[i] = data[3 + i];
return d;
}
}
public class Receiver {
static int cp = 1;
static BufferedReader ir = new BufferedReader(new
InputStreamReader(System.in));
static int ourPort = 1789;
public static void main (String[] args) throws Exception {
receiver();
}
static void receiver() throws Exception {
System.out.println(Receiver started.);
DatagramSocket ds = new DatagramSocket(ourPort);
int n = 0;
byte[] b1 = new byte[200];
while(true) {
DatagramPacket dp1 = new DatagramPacket(b1, b1.length);
try {
ds.receive(dp1);
}
catch(IOException e) {
continue;
}
System.out.println(Receiver: Packet received.);
DataFrame d1 = DataFrame.byteToFrame(b1);
if(d1.num != n)
continue; // Invalid frame number.
toNetworkLayer(b1);
byte[] data = new String(Acknowledge +
cp).getBytes();
byte[] b2 = DataFrame.frameDataToByte((byte)2, (byte)n,
data);
DatagramPacket dp2 = new DatagramPacket(b2, b2.length,
dp1.getAddress(), dp1.getPort());
try {
ds.send(dp2);
}
catch(IOException e) {
continue;
}
n = (n == 1) ? 0 : 1;
cp++;
}
}
static void toNetworkLayer(byte[] data) throws Exception {
System.out.println(Received message + cp + :);
System.out.println(new String(data));
}
}
--
Summary: Just tried to compile a java source code with gcc java
Product: gcc
Version: unknown
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: java
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: black dot uprise at gmail dot com
GCC host triplet: All in msg body
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33570
[Bug fortran/33568] ICE with ANINT (with KIND and an array)
--- Comment #2 from pault at gcc dot gnu dot org 2007-09-27 11:46 --- I might as well take it:-) Paul -- pault at gcc dot gnu dot org changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |pault at gcc dot gnu dot org |dot org | Status|NEW |ASSIGNED Last reconfirmed|2007-09-27 10:39:02 |2007-09-27 11:46:50 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33568
[Bug fortran/33568] ICE with ANINT (with KIND and an array)
--- Comment #3 from tobi at gcc dot gnu dot org 2007-09-27 12:09 --- Weird, I tried essentially the same patch (where I modeled the code after gfc_conv_intrinsic_nint), but the failure persisted. I wonder what's different now. -- tobi at gcc dot gnu dot org changed: What|Removed |Added CC||tobi at gcc dot gnu dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33568
[Bug c++/32565] [4.3 regression] ICE with specialization of variadic template
--- Comment #1 from jakub at gcc dot gnu dot org 2007-09-27 12:12 --- Created an attachment (id=14255) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14255action=view) gcc43-pr32565.patch WIP patch to fix this. As shown in the testcase, I believe trying to handle ARGUMENT_PACK_P (parm) !ARGUMENT_PACK_P (arg) in unify in unify is bad, as parm may need more than one argument. So IMHO this has to be handled one level higher where we have access to all the arguments. I don't know what exactly to do with argument packs which have PACK_EXPANSION_P at the end (can you come up with a testcase for that) and especially if such ARGUMENT_PACK_P is not the last template argument (that is only possible for function templates, right)? -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32565
[Bug c/25309] [4.1/4.2 Regression] ICE on initialization of a huge array
--- Comment #16 from jsm28 at gcc dot gnu dot org 2007-09-27 12:19 --- Fixed for 4.3.0. -- jsm28 at gcc dot gnu dot org changed: What|Removed |Added Summary|[4.0/4.1/4.2/4.3 Regression]|[4.1/4.2 Regression] ICE on |ICE on initialization of a |initialization of a huge |huge array |array http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25309
[Bug c/32295] [4.1/4.2 Regression] An invalid c code causes an ICE in create_tmp_var, at gimplify.c:489
--- Comment #7 from jsm28 at gcc dot gnu dot org 2007-09-27 12:20 --- Fixed for 4.3.0. -- jsm28 at gcc dot gnu dot org changed: What|Removed |Added Summary|[4.1/4.2/4.3 Regression] An |[4.1/4.2 Regression] An |invalid c code causes an ICE|invalid c code causes an ICE |in create_tmp_var, at |in create_tmp_var, at |gimplify.c:489 |gimplify.c:489 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32295
[Bug c++/33461] [4.3 regression] ICE with invalid specialization involving parameter packs
-- pcarlini at suse dot de changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |pcarlini at suse dot de |dot org | Status|UNCONFIRMED |ASSIGNED Ever Confirmed|0 |1 Last reconfirmed|-00-00 00:00:00 |2007-09-27 12:20:43 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33461
[Bug fortran/33544] [4.3 only] Warning in TRANSFER intrinsic should be made optional
--- Comment #9 from fxcoudert at gcc dot gnu dot org 2007-09-27 12:27 --- (In reply to comment #7) It doesn't matter if it's bad coding (which I can agree). AFAIK, the standard says the code is legal. After reading the standard, I concur. I think a warning option should be added, maybe -Wshort-transfer, maybe -Wsurprising... -- fxcoudert at gcc dot gnu dot org changed: What|Removed |Added CC||fxcoudert at gcc dot gnu dot ||org Status|UNCONFIRMED |NEW Ever Confirmed|0 |1 Keywords||diagnostic Last reconfirmed|-00-00 00:00:00 |2007-09-27 12:27:35 date|| Summary|Spurious warning in TRANSFER|[4.3 only] Warning in |intrinsic in Sept 24|TRANSFER intrinsic should be |snapshot of gfortran|made optional Target Milestone|--- |4.3.0 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33544
[Bug fortran/33550] ICE (segfault) when USEing ambiguous symbols
-- fxcoudert at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |NEW Ever Confirmed|0 |1 Last reconfirmed|-00-00 00:00:00 |2007-09-27 13:26:54 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33550
[Bug c++/32121] [4.1/4.2/4.3 Regression] C++ front-end accepts invalid __label__ declarations
-- jakub at gcc dot gnu dot org changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |jakub at gcc dot gnu dot org |dot org | Status|UNCONFIRMED |ASSIGNED Ever Confirmed|0 |1 Last reconfirmed|-00-00 00:00:00 |2007-09-27 13:30:54 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32121
[Bug tree-optimization/33562] [4.3 Regression] aggregate DSE disabled
--- Comment #5 from rguenth at gcc dot gnu dot org 2007-09-27 13:38 --- Created an attachment (id=14256) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14256action=view) more complete patch to resture DCE of killing defs It still breaks in some cases. With the unfortunate fact that we need the VMAYUSEs even for killing defs we might end up to need to do a two-phase DCE for those :/ (See the special casing of PHI_NODEs in the patch, it seems some more special casing is missing) -- rguenth at gcc dot gnu dot org changed: What|Removed |Added Attachment #14254|0 |1 is obsolete|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33562
[Bug tree-optimization/33562] [4.3 Regression] aggregate DSE disabled
--- Comment #6 from rguenth at gcc dot gnu dot org 2007-09-27 13:42 --- Diego, it sucks that we need to jump through hoops to get V_MUST_DEF back. -- rguenth at gcc dot gnu dot org changed: What|Removed |Added CC||dnovillo at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33562
[Bug tree-optimization/33562] [4.3 Regression] aggregate DSE disabled
--- Comment #7 from dnovillo at google dot com 2007-09-27 13:48 --- Subject: Re: [4.3 Regression] aggregate DSE disabled On 27 Sep 2007 13:42:11 -, rguenth at gcc dot gnu dot org [EMAIL PROTECTED] wrote: Diego, it sucks that we need to jump through hoops to get V_MUST_DEF back. Unless we can prove that it is impossible to implement DSE any other way, I would prefer to keep virtual SSA as simple as possible. It's meant as a safety net and a good enough UD web for passes that do not care for being too aggressive. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33562
[Bug fortran/33568] ICE with ANINT (with KIND and an array)
--- Comment #4 from dominiq at lps dot ens dot fr 2007-09-27 13:59 --- The patch fixes the test case on this PR, but gives ICE on several of my tests. The simplest is: program aint_anint_1 implicit none real(8) :: s = 42.7D0, s1, s2 s1 = aint(s) ! s2 = aint(s, kind=4) end program aint_anint_1 aint_anint_1_red.f90: In function 'MAIN__': aint_anint_1_red.f90:1: internal compiler error: in gfc_conv_intrinsic_function_args, at fortran/trans-intrinsic.c:188 Is not there some tests about the shape of the argument(s) missing? Currently regtesting. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33568
[Bug tree-optimization/33562] [4.3 Regression] aggregate DSE disabled
--- Comment #8 from rguenther at suse dot de 2007-09-27 14:01 --- Subject: Re: [4.3 Regression] aggregate DSE disabled On Thu, 27 Sep 2007, dnovillo at google dot com wrote: --- Comment #7 from dnovillo at google dot com 2007-09-27 13:48 --- Subject: Re: [4.3 Regression] aggregate DSE disabled On 27 Sep 2007 13:42:11 -, rguenth at gcc dot gnu dot org [EMAIL PROTECTED] wrote: Diego, it sucks that we need to jump through hoops to get V_MUST_DEF back. Unless we can prove that it is impossible to implement DSE any other way, I would prefer to keep virtual SSA as simple as possible. It's meant as a safety net and a good enough UD web for passes that do not care for being too aggressive. I sort-of agree. Still DCE was able to handle tree-ssa/complex-4.c before we removed V_MUST_DEF. Which is what I'm trying to get back. As good enough UD web it would be nice to have only single VDEFs on stores (I don't care for clobbers at call sites). Though finding the optimal static partitioning to ensure this is probably hard? Richard. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33562
[Bug tree-optimization/33562] [4.3 Regression] aggregate DSE disabled
--- Comment #9 from dnovillo at google dot com 2007-09-27 14:12 --- Subject: Re: [4.3 Regression] aggregate DSE disabled On 27 Sep 2007 14:01:18 -, rguenther at suse dot de [EMAIL PROTECTED] wrote: I sort-of agree. Still DCE was able to handle tree-ssa/complex-4.c before we removed V_MUST_DEF. Which is what I'm trying to get back. Yeah, it is somewhat tempting to make the infrastructure more powerful because you suddenly get more out of seemingly innocent passes. However, a more powerful infrastructure creates problems of its own, it needs to be maintained and it causes slowdowns even in passes that do not need all the expressive power. As good enough UD web it would be nice to have only single VDEFs on stores (I don't care for clobbers at call sites). Though finding the optimal static partitioning to ensure this is probably hard? Yeah, that is the whole motivation behind the dynamic aspects of mem-ssa, but getting it right has proven tricky. Unfortunately, I have not had time to come back to that idea in some time. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33562
[Bug c++/33571] New: [4.3 Regression] ICEs on unknown attributes on template functions
struct foo {
template class T
void __attribute__((leafify)) bar() {}
};
void bar(void)
{
foo f;
f.barint();
}
--
Summary: [4.3 Regression] ICEs on unknown attributes on template
functions
Product: gcc
Version: 4.3.0
Status: UNCONFIRMED
Keywords: ice-on-valid-code
Severity: normal
Priority: P3
Component: c++
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: rguenth at gcc dot gnu dot org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33571
[Bug c++/33571] [4.3 Regression] ICEs on unknown attributes on template functions
--- Comment #1 from rguenth at gcc dot gnu dot org 2007-09-27 14:39 --- Program received signal SIGSEGV, Segmentation fault. 0x00545c5a in is_late_template_attribute (attr=0x2b2f2b957b70, decl=0x2b2f2f438a00) at /space/rguenther/src/svn/pointer_plus/gcc/cp/decl2.c:994 994 else if (TREE_CODE (decl) == TYPE_DECL || spec-type_required) (gdb) print spec $1 = (const struct attribute_spec *) 0x0 (gdb) call debug_tree (decl) function_decl 0x2b2f2f438a00 evaluate type function_type 0x2b2f2bd620d0 -- rguenth at gcc dot gnu dot org changed: What|Removed |Added Target Milestone|--- |4.3.0 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33571
[Bug rtl-optimization/33552] [4.3 Regression] wrong code for multiple output asm, wrong df?
--- Comment #19 from matz at gcc dot gnu dot org 2007-09-27 14:52 ---
Have a patch on http://gcc.gnu.org/ml/gcc-patches/2007-09/msg01968.html .
Fixes also the reload failure on x86 -O2 -fPIC on this testcase (which hits
glibc):
/* { dg-do compile } */
/* { dg-options -O2 -fPIC } */
typedef unsigned int size_t;
char *
__mempcpy_by2 (char *__dest, __const char *__src, size_t __srclen)
{
register char *__tmp = __dest;
register unsigned long int __d0, __d1;
__asm__ __volatile__
(shrl $1,%3\n\t
jz2f\n
1:\n\t
movl (%2),%0\n\t
leal 4(%2),%2\n\t
movl %0,(%1)\n\t
leal 4(%1),%1\n\t
decl %3\n\t
jnz 1b\n
2:\n\t
movw (%2),%w0\n\t
movw %w0,(%1)
: =q (__d0), =r (__tmp), =r (__src), =r (__d1),
=m ( *(struct { __extension__ char __x[__srclen]; } *)__dest)
: 1 (__tmp), 2 (__src), 3 (__srclen / 2),
m ( *(struct { __extension__ char __x[__srclen]; } *)__src)
: cc);
return __tmp + 2;
}
--
matz at gcc dot gnu dot org changed:
What|Removed |Added
CC||matz at gcc dot gnu dot org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33552
[Bug rtl-optimization/33552] [4.3 Regression] wrong code for multiple output asm, wrong df?
--- Comment #20 from bonzini at gnu dot org 2007-09-27 15:01 --- Thanks. -- bonzini at gnu dot org changed: What|Removed |Added AssignedTo|bonzini at gnu dot org |unassigned at gcc dot gnu ||dot org Status|ASSIGNED|NEW http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33552
[Bug target/33479] SyncTest Intermittent failing on MIPS
--- Comment #6 from daney at gcc dot gnu dot org 2007-09-27 15:53 --- Fixed now. -- daney at gcc dot gnu dot org changed: What|Removed |Added Status|NEW |RESOLVED Resolution||FIXED Target Milestone|--- |4.3.0 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33479
[Bug c++/33572] New: wrong code with -O
cat bug.cc
#include vector
#include memory
struct Foo { virtual void f() {} };
int main(int argc, char**)
{
std::auto_ptrFoo foo;
if (argc) {
foo.reset(new Foo());
} else {
std::vectorint v;
}
Foo* p = foo.release();
p-f();
}
g++ -O bug.cc
a.out
Segmentation fault (core dumped)
I'm using SVN revision 128838 on linux-x86_64.
Without -O or with g++ version 4.2.1 this program works correctly.
--
Summary: wrong code with -O
Product: gcc
Version: 4.3.0
Status: UNCONFIRMED
Severity: major
Priority: P3
Component: c++
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: wouter dot vermaelen at scarlet dot be
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33572
[Bug bootstrap/32272] make exit because build/genmodes.exe doesn't exist
--- Comment #4 from haubi at gentoo dot org 2007-09-27 16:09 --- Same here on aix5.3 - it seems to be a make problem, not finding the implicit rule build/gen%$(build_exeext): for target 'build/genmodes', defined in gcc-4.2.0/gcc/Makefile.in line 3026. Which version of GNU make do you use (prerequisites say = 3.79.1) ? Does it make a difference when you use GNU make 3.81 (or 3.80) ? -- haubi at gentoo dot org changed: What|Removed |Added CC||haubi at gentoo dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32272
[Bug fortran/33568] ICE with ANINT (with KIND and an array)
--- Comment #5 from dominiq at lps dot ens dot fr 2007-09-27 16:18 --- With the new patch I still have an ICE on: reala real*8 c print *, (nearest(0.5,-1.0)+0.5)-1.0 a = 8388609.0 print '(3(1PG26.9))', a, anint(a), anint(8388609.0) a = 8388610.0 print '(3(1PG26.9))', a, anint(a), anint(8388610.0) a = 0.4992 print '(3(1PG26.9))', a, anint(a), anint(0.4992) a = 0.4997 print '(3(1PG26.9))', a, anint(a), anint(0.4997) print *, a-nearest(0.5,-1.0), (a+0.5)-1.0 c = 4503599627370498.0d0 print '(3(1PG26.18))', c, dnint(c), dnint(4503599627370498.0d0) c = 4503599627370497.0d0 print '(3(1PG26.18))', c, dnint(c), dnint(4503599627370497.0d0) c = 0.4999d0 print '(3(1PG26.18))', c, dnint(c), dnint(0.4999d0) c = 0.49994d0 print '(3(1PG26.18))', c, dnint(c), dnint(0.49994d0) end nint_tst.f90:6: internal compiler error: in gfc_conv_intrinsic_function_args, at fortran/trans-intrinsic.c:188 I also failures for gfortran.dg/PR19754_2.f90 -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33568
[Bug c++/31434] [4.3 regression] ICE with invalid use of parameter pack in function arg
--- Comment #4 from jakub at gcc dot gnu dot org 2007-09-27 16:33 --- This is a special case of PR33509. regenerate_decl_from_template doesn't handle error_mark_node returned from tsubst_pack_expansion. As tsubst_pack_expansion doesn't issue any error in this case (it hits the /* We could not find any argument packs that work. */ if (len 0) return error_mark_node; case), I believe it is up to regenerate_decl_from_template to issue some diagnostics. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31434
[Bug c++/33572] wrong code with -O
--- Comment #1 from pcarlini at suse dot de 2007-09-27 16:36 --- I can't reproduce... -- pcarlini at suse dot de changed: What|Removed |Added Summary|wrong code with -O |wrong code with -O http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33572
[Bug c++/33573] New: over-optimization in 4.1.2
Compiling the following code with -O2 or higher in g++ 4.1.2 produces incorrect
results. The problem doesn't occur in version 4.2.1.
I compile like this:
g++ -W -Wall -Werror-O2 -fPIC -o gccbug gccbug.cpp
The last line of output should show the number 27, but it shows 0 when
optimizing with -O2.
A tar archive containing the source file and a Makefile to build it is here:
http://dooglus.rincevent.net/random/gccbug.1190912601.tgz
Defining any of FIX0, FIX1 or FIX2 fixes the problem.
Here's the source file:
-
#include iostream
#include string
#include vector
using namespace std;
struct Vector {
float _x, _y;
Vector(): _x(0), _y(0) {};
float m(const int i) { return i?_y:_x; }
};
struct Node {
vectorNode* nl;
Node() {nl.push_back(this);};
string get_name() const {return x;};
vectorNode* get_children() { return nl;};
};
int six() { return 6; }
Vector parse_vector(Node *element)
{
Vector vect;
Node *node = element-get_children()[0];
if(node-get_name()==x) { // true
#ifdef FIX0
cout 1. vect.m(0) = 27 endl;
#else
printf(1. vect.m(0) = 27\n);
#endif
vect.m(0)=27;
#ifdef FIX1 // printing the value stops the
over-optimization
cout 2. vect.m(0) == vect.m(0) endl;
#endif
} else if(six() == 6 node-get_children().empty()) { // false
exit(0);
return Vector();
}
#ifdef FIX2 // checking the value also stops the over optimization
if (vect.m(0) + vect.m(1) == 24234.657) cout har endl;
#endif
return vect;
}
int main() {
cout 3. vect.m(0) == parse_vector(new Node).m(0) --
should be 27 endl;
}
-
--
Summary: over-optimization in 4.1.2
Product: gcc
Version: 4.1.2
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c++
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: dooglus at gmail dot com
GCC build triplet: i486-linux-gnu
GCC host triplet: i486-linux-gnu
GCC target triplet: i486-linux-gnu
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33573
[Bug bootstrap/33567] Bootstrapping not working in b128824
--- Comment #1 from karthikkumar at gmail dot com 2007-09-27 17:29 --- my bad. gcc be good. -- karthikkumar at gmail dot com changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||INVALID http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33567
[Bug tree-optimization/33565] [4.3 regression] spurious warning: assuming signed overflow does not occur when assuming that (X + c) = X is always true
--- Comment #3 from ian at gcc dot gnu dot org 2007-09-27 17:31 --- Subject: Bug 33565 Author: ian Date: Thu Sep 27 17:31:34 2007 New Revision: 128840 URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=128840 Log: ./: PR tree-optimization/33565 * tree-ssa-loop-ch.c (copy_loop_headers): Set TREE_NO_WARNING on assignments of comparisons. * tree-ssa-sccvn.c (simplify_binary_expression): Add stmt parameter. Change caller. Defer overflow warnings around call to fold_binary. * fold-const.c (fold_undefer_overflow_warnings): Don't warn if TREE_NO_WARNING is set on the statement. * tree-ssa-forwprop.c (tree_ssa_forward_propagate_single_use_vars): Don't test TREE_NO_WARNING when calling fold_undefer_overflow_warnings. * tree-cfg.c (fold_cond_expr_cond): Likewise. testsuite/: PR tree-optimization/33565 * gcc.dg/Wstrict-overflow-20.c: New test. Added: trunk/gcc/testsuite/gcc.dg/Wstrict-overflow-20.c Modified: trunk/gcc/ChangeLog trunk/gcc/fold-const.c trunk/gcc/testsuite/ChangeLog trunk/gcc/tree-cfg.c trunk/gcc/tree-ssa-forwprop.c trunk/gcc/tree-ssa-loop-ch.c trunk/gcc/tree-ssa-sccvn.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33565
[Bug tree-optimization/33565] [4.3 regression] spurious warning: assuming signed overflow does not occur when assuming that (X + c) = X is always true
--- Comment #4 from ian at airs dot com 2007-09-27 17:37 --- Fixed. -- ian at airs dot com changed: What|Removed |Added Status|NEW |RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33565
[Bug fortran/33574] New: getarg and iargc seem to be required by mpich
Hi, I compiled gcc and gfortran from svn trunk, revision: 128668, date:20070922. I am compiling mpich and it complains unavailability of getarg and iargc. This problem seems to have been patched in older version. Did this problem reappear or am I doing something wrong? thanks, Krishna. -- Summary: getarg and iargc seem to be required by mpich Product: gcc Version: 4.3.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: fortran AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: gkmohan at gmail dot com GCC build triplet: ia64-unknown-linux-gnu GCC host triplet: ia64-unknown-linux-gnu GCC target triplet: ia64-unknown-linux-gnu http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33574
[Bug c++/31434] [4.3 regression] ICE with invalid use of parameter pack in function arg
--- Comment #5 from jakub at gcc dot gnu dot org 2007-09-27 18:23 ---
Actually, why do you think this is invalid?
templatetypename... T void foo(const T ...) {}
templatetypename... T void foo(T *const ...) {}
etc. are all accepted and I couldn't find in n2152.pdf any word that would
say that const T... is invalid.
Will post a patch that cures this if that is not invalid.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31434
[Bug fortran/33568] ICE with ANINT (with KIND and an array)
--- Comment #6 from pault at gcc dot gnu dot org 2007-09-27 18:40 --- Subject: Bug 33568 Author: pault Date: Thu Sep 27 18:39:55 2007 New Revision: 128843 URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=128843 Log: 2007-09-27 Paul Thomas [EMAIL PROTECTED] PR fortran/33568 * trans-intrinsic.c (gfc_conv_intrinsic_aint): Allow for the possibility of the optional KIND argument by making arg an array, counting the number of arguments and using arg[0]. 2007-09-27 Paul Thomas [EMAIL PROTECTED] PR fortran/33568 * gfortran.dg/anint_1.f90: New test. Added: trunk/gcc/testsuite/gfortran.dg/anint_1.f90 Modified: trunk/gcc/fortran/ChangeLog trunk/gcc/fortran/trans-intrinsic.c trunk/gcc/testsuite/ChangeLog -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33568
[Bug fortran/33574] getarg and iargc seem to be required by mpich
--- Comment #1 from pault at gcc dot gnu dot org 2007-09-27 18:49 --- After a bit of messing around, this is fixed on trunk. Cheers Paul -- pault at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33574
[Bug fortran/33574] getarg and iargc seem to be required by mpich
--- Comment #2 from pault at gcc dot gnu dot org 2007-09-27 18:50 --- Woops - wrong bug! Paul -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33574
[Bug fortran/33574] getarg and iargc seem to be required by mpich
--- Comment #3 from pault at gcc dot gnu dot org 2007-09-27 18:51 --- Woops - wrong bug! Paul -- pault at gcc dot gnu dot org changed: What|Removed |Added Status|RESOLVED|UNCONFIRMED Resolution|FIXED | http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33574
[Bug fortran/33568] ICE with ANINT (with KIND and an array)
--- Comment #7 from pault at gcc dot gnu dot org 2007-09-27 18:52 --- After a bit of messing around, this is fixed on trunk. Paul -- pault at gcc dot gnu dot org changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33568
[Bug c++/31434] [4.3 regression] ICE with invalid use of parameter pack in function arg
-- jakub at gcc dot gnu dot org changed: What|Removed |Added AssignedTo|dgregor at gcc dot gnu dot |jakub at gcc dot gnu dot org |org | URL||http://gcc.gnu.org/ml/gcc- ||patches/2007- ||09/msg01987.html Status|REOPENED|ASSIGNED Keywords|accepts-invalid, ice-on-|ice-on-valid-code |invalid-code| Last reconfirmed|2007-04-25 17:59:04 |2007-09-27 18:57:34 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31434
[Bug c/33575] New: segfault while building libgcc
Hi. an error occured while compiling to _muldi3.o of libgcc
A verbose/gdb shows a null reference somewhere in fold_builtin_1 in
gcc/builtins.c, line 9768
--
GNU gdb 6.6
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type show copying to see the conditions.
There is absolutely no warranty for GDB. Type show warranty for details.
This GDB was configured as i486-slackware-linux...
Using host libthread_db library /lib/libthread_db.so.1.
(gdb) run
Starting program: /home/karthik/WS/act/gcc-mainline-dp-build/gcc/cc1 -quiet -v
-I. -I. -I../.././gcc -I../../../gcc-mainline-dp/libgcc
-I../../../gcc-mainline-dp/libgcc/. -I../../../gcc-mainline-dp/libgcc/../gcc
-I../../../gcc-mainline-dp/libgcc/../include
-I../../../gcc-mainline-dp/libgcc/config/libbid -iprefix
/home/karthik/WS/act/gcc-mainline-dp-build/gcc/../lib/gcc/i686-pc-linux-gnu/4.3.0/
-isystem /home/karthik/WS/act/gcc-mainline-dp-build/./gcc/include -isystem
/home/karthik/WS/act/gcc-mainline-dp-build/./gcc/include-fixed -MD _muldi3.d
-MF _muldi3.dep -MP -MT _muldi3.o -DIN_GCC -DHAVE_GTHR_DEFAULT -DIN_LIBGCC2
-D__GCC_FLOAT_NOT_NEEDED -DENABLE_DECIMAL_BID_FORMAT -DHAVE_CC_TLS -DUSE_TLS
-DL_muldi3 -DHIDE_EXPORTS -isystem
/opt/gcc-mainline-dp/i686-pc-linux-gnu/include -isystem
/opt/gcc-mainline-dp/i686-pc-linux-gnu/sys-include -isystem ./include
../../../gcc-mainline-dp/libgcc/../gcc/libgcc2.c -quiet -dumpbase libgcc2.c
-mtune=generic -auxbase-strip _muldi3.o -g -g -g -O2 -O2 -O2 -O2 -W -Wall
-Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition
-version -fomit-frame-pointer -fPIC -fvisibility=hidden -o /tmp/cc77FHjt.s
ignoring nonexistent directory /opt/gcc-mainline-dp/i686-pc-linux-gnu/include
ignoring nonexistent directory
/opt/gcc-mainline-dp/i686-pc-linux-gnu/sys-include
ignoring nonexistent directory ./include
ignoring nonexistent directory
/home/karthik/WS/act/gcc-mainline-dp-build/gcc/../lib/gcc/i686-pc-linux-gnu/4.3.0/include
ignoring nonexistent directory
/home/karthik/WS/act/gcc-mainline-dp-build/gcc/../lib/gcc/i686-pc-linux-gnu/4.3.0/include-fixed
ignoring nonexistent directory
/home/karthik/WS/act/gcc-mainline-dp-build/gcc/../lib/gcc/i686-pc-linux-gnu/4.3.0/../../../../i686-pc-linux-gnu/include
ignoring nonexistent directory /opt/gcc-mainline-dp/include
ignoring nonexistent directory
/opt/gcc-mainline-dp/lib/gcc/i686-pc-linux-gnu/4.3.0/include
ignoring nonexistent directory
/opt/gcc-mainline-dp/lib/gcc/i686-pc-linux-gnu/4.3.0/include-fixed
ignoring nonexistent directory /opt/gcc-mainline-dp/i686-pc-linux-gnu/include
ignoring duplicate directory .
ignoring duplicate directory ../../../gcc-mainline-dp/libgcc/.
#include ... search starts here:
#include ... search starts here:
.
../.././gcc
../../../gcc-mainline-dp/libgcc
../../../gcc-mainline-dp/libgcc/../gcc
../../../gcc-mainline-dp/libgcc/../include
../../../gcc-mainline-dp/libgcc/config/libbid
/home/karthik/WS/act/gcc-mainline-dp-build/./gcc/include
/home/karthik/WS/act/gcc-mainline-dp-build/./gcc/include-fixed
/usr/local/include
/usr/include
End of search list.
GNU C (GCC) version 4.3.0 20070927 (experimental) (i686-pc-linux-gnu)
compiled by GNU C version 4.3.0 20070927 (experimental), GMP version
4.2.1, MPFR version 2.2.1-p1.
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096
Compiler executable checksum: 1f5c3ab45e0ab524f80d84058122b450
Program received signal SIGSEGV, Segmentation fault.
0x08108292 in fold_builtin_1 (fndecl=0xb7ceb680, arg0=0xb7bb9bc0, ignore=0
'\0')
at ../../gcc-mainline-dp/gcc/builtins.c:9768
9768{
--
Summary: segfault while building libgcc
Product: gcc
Version: 4.3.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: karthikkumar at gmail dot com
GCC build triplet: i686-pc-linux-gnu
GCC host triplet: i686-pc-linux-gnu
GCC target triplet: i686-pc-linux-gnu
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33575
[Bug rtl-optimization/33552] [4.3 Regression] wrong code for multiple output asm, wrong df?
--- Comment #21 from ubizjak at gmail dot com 2007-09-27 19:59 --- (In reply to comment #19) Have a patch on http://gcc.gnu.org/ml/gcc-patches/2007-09/msg01968.html . Fixes also the reload failure on x86 -O2 -fPIC on this testcase (which hits glibc): Do we need a solution for %. decorations of matched operands? strtoul() will not return 0 for %0 operands, as described in Comment 11. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33552
[Bug c++/31434] [4.3 regression] ICE with invalid use of parameter pack in function arg
--- Comment #6 from reichelt at gcc dot gnu dot org 2007-09-27 20:19 --- Actually, why do you think this is invalid? Hmmm. The code looks valid to me now. Must have been a temporary lack of coffee... -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31434
[Bug c++/33572] wrong code with -O
--- Comment #2 from belyshev at depni dot sinp dot msu dot ru 2007-09-27 20:20 --- Could you please attach preprocessed source (.ii file ) and assembly output (.s) (use -save-temps option to generate them) -- belyshev at depni dot sinp dot msu dot ru changed: What|Removed |Added CC||belyshev at depni dot sinp ||dot msu dot ru Severity|major |normal Status|UNCONFIRMED |WAITING http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33572
[Bug c++/33572] wrong code with -O
--- Comment #3 from wouter dot vermaelen at scarlet dot be 2007-09-27 20:30 --- Created an attachment (id=14257) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14257action=view) preprocessed source (reduced) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33572
[Bug c++/33572] wrong code with -O
--- Comment #4 from wouter dot vermaelen at scarlet dot be 2007-09-27 20:31 --- Created an attachment (id=14258) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14258action=view) assembly output (for reduced bug.ii) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33572
[Bug c++/33571] [4.3 Regression] ICEs on unknown attributes on template functions
-- jason at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |WAITING http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33571
[Bug c++/33571] [4.3 Regression] ICEs on unknown attributes on template functions
--- Comment #2 from jason at gcc dot gnu dot org 2007-09-27 20:58 --- Fixed. -- jason at gcc dot gnu dot org changed: What|Removed |Added Status|WAITING |RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33571
[Bug tree-optimization/33572] [4.3 Regression] wrong code with -O
--- Comment #5 from rguenth at gcc dot gnu dot org 2007-09-27 21:03 ---
Confirmed. The reduced testcase gets optimized to
bb 3:
D.2145 = operator new (8);
bb 4:
((struct Foo *) D.2145)-_vptr.Foo = _ZTV3Foo[2];
bb 5:
D.2147 ={v} 0B-_vptr.Foo;
OBJ_TYPE_REF(*D.2147;0B-0) (0B);
where dereferencing NULL obviously is bogus. After 038t.sdse2 everything
looks ok:
bb 4:
D.2124_3 = (struct Foo *) D.2145_2;
D.2124_3-_vptr.Foo ={v} _ZTV3Foo[2];
goto bb 6;
bb 5:
deallocate (v, 0B);
bb 6:
# foo$ptr_16 = PHI foo$ptr_18(ab)(5), D.2124_3(4)
foo$ptr_19(ab) = 0B;
D.2147_6 = foo$ptr_16-_vptr.Foo;
D.2148_7 = *D.2147_6;
but after apply inline it's wrong!?
bb 3:
D.2145_2 = operator new (8);
bb 4:
D.2124_3 = (struct Foo *) D.2145_2;
D.2124_3-_vptr.Foo ={v} _ZTV3Foo[2];
bb 5:
# foo$ptr_16 = PHI foo$ptr_18(ab)(2), D.2124_3(4)
foo$ptr_19(ab) = 0B;
D.2147_6 = foo$ptr_19(ab)-_vptr.Foo;
D.2148_7 = *D.2147_6;
OBJ_TYPE_REF(D.2148_7;foo$ptr_19(ab)-0) (foo$ptr_19(ab));
goto bb 7;
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
CC||rguenth at gcc dot gnu dot
||org
Status|WAITING |NEW
Component|c++ |tree-optimization
Ever Confirmed|0 |1
Keywords||wrong-code
Last reconfirmed|-00-00 00:00:00 |2007-09-27 21:03:18
date||
Summary|wrong code with -O |[4.3 Regression] wrong code
||with -O
Target Milestone|--- |4.3.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33572
[Bug tree-optimization/33576] New: segfault in extract_muldiv for cpu2006 benchmark
The following trimmed testcase from 464.h264ref is segfaulting.
int a1[6][4][4];
short b1[16];
int c1;
void CalculateQuantParam(void)
{
int i, j, k, temp;
for(k=0; k6; k++)
for(j=0; j4; j++)
for(i=0; i4; i++)
{
temp = (i2)+j;
a1[k][j][i] = c1/b1[temp];
}
}
$ /home/pthaugen/install/gcc/trunk/bin/gcc -c -m32 -O2 -ftree-loop-linear
junk.c
junk.c: In function #CalculateQuantParam#:
junk.c:6: internal compiler error: Segmentation fault
Please submit a full bug report,
Program received signal SIGSEGV, Segmentation fault.
extract_muldiv (t=0xf6efa1c0, c=0xf7de0f60, code=MULT_EXPR,
wide_type=0xf6e60150,
strict_overflow_p=0xffaaee08 )
at /home/pthaugen/src/gcc/trunk/gcc/gcc/fold-const.c:6008
6008GET_MODE_SIZE (TYPE_MODE (type)))
(gdb) bt 10
#0 extract_muldiv (t=0xf6efa1c0, c=0xf7de0f60, code=MULT_EXPR,
wide_type=0xf6e60150,
strict_overflow_p=0xffaaee08 )
at /home/pthaugen/src/gcc/trunk/gcc/gcc/fold-const.c:6008
#1 0x42084484 in ?? ()
#2 0x10212c6c in extract_muldiv (t=value optimized out, c=0xf7de0f60,
code=MULT_EXPR, wide_type=0xf6e60460, strict_overflow_p=0xffaaeee0 )
at /home/pthaugen/src/gcc/trunk/gcc/gcc/fold-const.c:6150
#3 0x22084442 in ?? ()
#4 0x10212f58 in extract_muldiv (t=0xf6f2adc0, c=0xf7de0480, code=MULT_EXPR,
wide_type=0xf6e60150, strict_overflow_p=0xffaaeee0 )
at /home/pthaugen/src/gcc/trunk/gcc/gcc/fold-const.c:6066
#5 0x42084442 in ?? ()
#6 0x101f1e34 in fold_binary (code=MULT_EXPR, type=0xf6e60150, op0=0xf6f2adc0,
op1=0xf7de0480) at /home/pthaugen/src/gcc/trunk/gcc/gcc/fold-const.c:10359
#7 0x42008482 in ?? ()
#8 0x101fcb40 in fold_build2_stat (code=ERROR_MARK, type=0xf7de0f60, op0=0x43,
op1=0xf6e60150) at /home/pthaugen/src/gcc/trunk/gcc/gcc/fold-const.c:13560
#9 0x106d5730 in chrec_fold_multiply (type=0xf6e60150, op0=0xf6f2adc0,
op1=0xf7de0480)
at /home/pthaugen/src/gcc/trunk/gcc/gcc/tree-chrec.c:426
--
Summary: segfault in extract_muldiv for cpu2006 benchmark
Product: gcc
Version: 4.3.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: tree-optimization
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: pthaugen at gcc dot gnu dot org
GCC build triplet: powerpc64-linux
GCC host triplet: powerpc64-linux
GCC target triplet: powerpc64-linux
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33576
[Bug tree-optimization/33572] [4.3 Regression] wrong code with -O
--- Comment #6 from pcarlini at suse dot de 2007-09-27 21:19 --- I can see now from the reduced testcase that the library uses mt_alloc, not the default allocator. Next time, just say it... ;) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33572
[Bug c++/10179] alignment attributes are not inherited correctly with empty classes
-- jason at gcc dot gnu dot org changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |jason at gcc dot gnu dot org |dot org | Status|NEW |ASSIGNED Last reconfirmed|2005-12-18 20:20:22 |2007-09-27 21:21:39 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10179
[Bug fortran/33554] [4.3 regression] Seg.fault: Default initialization of derived type uses uninitialized values
--- Comment #9 from pault at gcc dot gnu dot org 2007-09-27 21:31 --- This probably caused by: http://gcc.gnu.org/ml/gcc-cvs/2007-07/msg00745.html r126885 | pault | 2007-07-24 21:15:27 +0200 (Di, 24 Jul 2007) | 36 lines 2007-07-24 Paul Thomas [EMAIL PROTECTED] PR 31205 PR 32842 * trans-expr.c (gfc_conv_function_call): Remove the default initialization of intent(out) derived types. I'd better look into it then hadn't I? Perhaps, Tobias, we had better start a meta-PR of regressions that I have caused? :-) Paul -- pault at gcc dot gnu dot org changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |pault at gcc dot gnu dot org |dot org | Status|NEW |ASSIGNED Last reconfirmed|2007-09-25 19:16:45 |2007-09-27 21:31:46 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33554
[Bug fortran/33268] read ('(f3.3)'), a rejected due to the extra (...)
--- Comment #5 from tobi at gcc dot gnu dot org 2007-09-27 21:41 --- I'm unassigning myself for the time being, as I couldn't think of a fix that does the right thing and doesn't involve overhauling parts of the compiler. -- tobi at gcc dot gnu dot org changed: What|Removed |Added AssignedTo|tobi at gcc dot gnu dot org |unassigned at gcc dot gnu ||dot org Status|ASSIGNED|NEW http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33268
[Bug fortran/33574] getarg and iargc seem to be required by mpich
--- Comment #4 from kargl at gcc dot gnu dot org 2007-09-27 21:50 --- (In reply to comment #0) Hi, I compiled gcc and gfortran from svn trunk, revision: 128668, date:20070922. I am compiling mpich and it complains unavailability of getarg and iargc. This problem seems to have been patched in older version. Did this problem reappear or am I doing something wrong? You need to fix the mpich source code. It most likely contains an EXTERNAL statement listing getarg and iargc. Remove that line. Also, instead of interpreting for us what the compiler complains about, you should include the actual output, so we don't have to guess. -- kargl at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||INVALID http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33574
[Bug c/29172] --combine can't handle #pragma once
--- Comment #5 from tromey at gcc dot gnu dot org 2007-09-27 22:04 --- Confirmed. The problem is that the loop in c_common_parse_file calls cpp_undef_all -- which undefines all symbols but does not mark _cpp_file objects as not seen. We probably need a new libcpp API for this. -- tromey at gcc dot gnu dot org changed: What|Removed |Added CC||tromey at gcc dot gnu dot ||org Status|UNCONFIRMED |NEW Ever Confirmed|0 |1 Last reconfirmed|-00-00 00:00:00 |2007-09-27 22:04:24 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29172
[Bug driver/33577] New: Compile failing for xpdf 3.02
I am trying to install xpdf 3.02 on a AIX 5.3.0.5 machine. The following is the information I have at hand: the exact version of GCC; Reading specs from /opt/freeware/lib/gcc-lib/powerpc-ibm-aix5.2.0.0/3.3.2/specs Configured with: ../configure --with-as=/usr/bin/as --with-ld=/usr/bin/ld --disable-nls --enable-languages=c,c++ --prefix=/opt/free ware --enable-threads --enable-version-specific-runtime-libs --host=powerpc-ibm-aix5.2.0.0 Thread model: aix gcc version 3.3.2 the system type; AIX 5.3.0.5 the options given when GCC was configured/built; rpm -Uvv gcc-3.3.2-5.aix5.2.ppc.rpm the complete command line that triggers the bug; make or gmake the compiler output (error messages, warnings, etc.); [EMAIL PROTECTED]:/fis70_patches/xpdf/xpdf-3.02 ==make cd goo; make Target all is up to date. cd fofi; make Target all is up to date. cd splash; make Target all is up to date. cd xpdf; make all g++ -g -O2 -DHAVE_CONFIG_H -I.. -I./../goo -I./../fofi -I./../splash -I. -I/usr/local/include/freetype2 -c CoreOut putDev.cc g++ -g -O2 -DHAVE_CONFIG_H -I.. -I./../goo -I./../fofi -I./../splash -I. -I/usr/local/include/freetype2 -c Gfx.cc g++: Internal error: Segmentation fault (program as) Please submit a full bug report. See URL:http://gcc.gnu.org/bugs.html for instructions. make: 1254-004 The error code from the last command is 1. Stop. make: 1254-004 The error code from the last command is 2. Stop. [EMAIL PROTECTED]:/fis70_patches/xpdf/xpdf-3.02 ==which gmake With 'gmake' --- [EMAIL PROTECTED]:/fis70_patches/xpdf/xpdf-3.02 ==gmake cd goo; gmake gmake[1]: Entering directory `/fis70_patches/xpdf/xpdf-3.02/goo' gmake[1]: Nothing to be done for `all'. gmake[1]: Leaving directory `/fis70_patches/xpdf/xpdf-3.02/goo' cd fofi; gmake gmake[1]: Entering directory `/fis70_patches/xpdf/xpdf-3.02/fofi' gmake[1]: Nothing to be done for `all'. gmake[1]: Leaving directory `/fis70_patches/xpdf/xpdf-3.02/fofi' cd splash; gmake gmake[1]: Entering directory `/fis70_patches/xpdf/xpdf-3.02/splash' gmake[1]: Nothing to be done for `all'. gmake[1]: Leaving directory `/fis70_patches/xpdf/xpdf-3.02/splash' cd xpdf; gmake all gmake[1]: Entering directory `/fis70_patches/xpdf/xpdf-3.02/xpdf' g++ -g -O2 -DHAVE_CONFIG_H -I.. -I./../goo -I./../fofi -I./../splash -I. -I/usr/local/include/freetype2 -c Gfx.cc g++: Internal error: Segmentation fault (program as) Please submit a full bug report. See URL:http://gcc.gnu.org/bugs.html for instructions. gmake[1]: *** [Gfx.o] Error 1 gmake[1]: Leaving directory `/fis70_patches/xpdf/xpdf-3.02/xpdf' gmake: *** [all] Error 2 -- Summary: Compile failing for xpdf 3.02 Product: gcc Version: unknown Status: UNCONFIRMED Severity: normal Priority: P3 Component: driver AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: cabanasg at metro dot net http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33577
[Bug c++/28743] [4.1/4.2/4.3 regression] ICE with invalid specialization
--- Comment #4 from janis at gcc dot gnu dot org 2007-09-27 22:11 --- A regression hunt on mainline using the submitter's testcase identified the following patch: http://gcc.gnu.org/viewcvs?view=revrev=114023 r114023 | mmitchel | 2006-05-23 20:45:44 + (Tue, 23 May 2006) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||mmitchel at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28743
[Bug c++/33573] over-optimization in 4.1.2
--- Comment #2 from dooglus at gmail dot com 2007-09-27 23:02 --- Created an attachment (id=14260) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14260action=view) the Makefile -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33573
[Bug middle-end/33575] segfault while building libgcc
--- Comment #1 from pinskia at gcc dot gnu dot org 2007-09-27 22:46 --- How did you invoke configure? -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33575
[Bug driver/33577] Compile failing for xpdf 3.02
--- Comment #1 from pinskia at gcc dot gnu dot org 2007-09-27 22:46 --- g++: Internal error: Segmentation fault (program as) as is crashing so I am assuming you don't have an updated as from IBM. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33577
[Bug c++/33573] over-optimization in 4.1.2
--- Comment #1 from dooglus at gmail dot com 2007-09-27 23:02 --- Created an attachment (id=14259) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14259action=view) the source file -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33573
[Bug driver/33577] Compile failing for xpdf 3.02
--- Comment #2 from cabanasg at metro dot net 2007-09-27 23:35 --- (In reply to comment #1) g++: Internal error: Segmentation fault (program as) as is crashing so I am assuming you don't have an updated as from IBM. You mean I should upgrade my GNU++ compiler? I have version 3.3.2-5 -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33577
[Bug libstdc++/33578] New: libstdc++ parallel mode broken on mingw32
/develop/svn/trunk/build/mingw32/libstdc++-v3/include/parallel/compatibility.h: In function 'void __gnu_parallel::yield()': /develop/svn/trunk/build/mingw32/libstdc++-v3/include/parallel/compatibility.h:331: error: 'Sleep' was not declared in this scope make[4]: *** [parallel_list.lo] Error 1 make[3]: *** [all-recursive] Error 1 make[2]: *** [all] Error 2 make[1]: *** [all-target-libstdc++-v3] Error 2 make: *** [all] Error 2 The 'obvious' fix is to just include windows.h for __MINGW32__ as well as _MSC_VER to get the prototype for Sleep. However, I expect that would cause many complaints because it would pollute the global namespace with all the Win32api names. These do cause problems in practice , especially when porting non-Win32 apps to mingw32. The _GTHREAD_HIDE_W32API business in gthr-win32.h was implemented precisely to avoid having to expose these names in libstdc++. From a mingw point of view it would be better if __gnu_parallel::yield could be hidden away either in a gthread_sched_yield function in libgcc or in a libstdc++ object file. Secondly, I don't think Sleep is actually the correct function to call on NT4 and later The SwitchToThread() function which Causes the calling thread to yield execution to another thread that is ready to run on the current processor. The operating system selects the next thread to be executed. may be better. Danny -- Summary: libstdc++ parallel mode broken on mingw32 Product: gcc Version: 4.3.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: libstdc++ AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: dannysmith at users dot sourceforge dot net GCC build triplet: i686-pc-mingw32 GCC host triplet: i686-pc-mingw32 GCC target triplet: i686-pc-mingw32 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33578
[Bug middle-end/33436] Bad constant output with TARGET_ASM_ALIGNED_DI_OP
--- Comment #3 from danglin at gcc dot gnu dot org 2007-09-28 00:57 --- Fixed on trunk. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33436
[Bug c++/33579] INIT_PRIORITY is broken
--- Comment #1 from dave at hiauly1 dot hia dot nrc dot ca 2007-09-28 01:41 --- Subject: Re: New: INIT_PRIORITY is broken On Fri, Sep 28, 2007 at 01:35:04AM -, danglin at gcc dot gnu dot org wrote: FAIL: gcc.dg/initpri1.c execution test I've attached the assembler output and the collect2/ld log. Dave --- Comment #2 from dave at hiauly1 dot hia dot nrc dot ca 2007-09-28 01:41 --- Created an attachment (id=14261) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14261action=view) --- Comment #3 from dave at hiauly1 dot hia dot nrc dot ca 2007-09-28 01:41 --- Created an attachment (id=14262) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14262action=view) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33579
[Bug bootstrap/32272] make exit because build/genmodes.exe doesn't exist
--- Comment #5 from jeff at jeffunit dot com 2007-09-28 01:37 --- I am using gnu make-3.81 I just verified that the problem is still present. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32272
[Bug c++/33579] New: INIT_PRIORITY is broken
FAIL: gcc.dg/initpri1.c execution test FAIL: g++.dg/special/conpr-2.C execution test FAIL: g++.dg/special/conpr-4.C execution test FAIL: g++.dg/special/initp1.C execution test FAIL: g++.dg/special/initpri1.C execution test On January 4, 2003, I installed a patch which in included support for init_priority on hppa*-*-hpux11*, http://gcc.gnu.org/ml/gcc-patches/2003-01/msg00208.html. This used collect2 to support constructors/destructors. Things started to go wrong with Mark's February 24, 2007 patch, http://gcc.gnu.org/ml/gcc-patches/2007-02/msg01950.html. However, the number of init_priority fails has varied somewhat with time. At the time of Mark's change, I believe on gcc.dg/initpri1.c started failing. -- Summary: INIT_PRIORITY is broken Product: gcc Version: 4.3.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: c++ AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: danglin at gcc dot gnu dot org GCC build triplet: hppa2.0w-hp-hpux11.11 GCC host triplet: hppa2.0w-hp-hpux11.11 GCC target triplet: hppa2.0w-hp-hpux11.11 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33579
[Bug middle-end/33436] Bad constant output with TARGET_ASM_ALIGNED_DI_OP
--- Comment #2 from danglin at gcc dot gnu dot org 2007-09-28 00:54 --- Subject: Bug 33436 Author: danglin Date: Fri Sep 28 00:54:29 2007 New Revision: 128855 URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=128855 Log: PR middle-end/33436 * expr.c (emit_group_load_1): Split constant double when destination length is half source length. Modified: trunk/gcc/ChangeLog trunk/gcc/expr.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33436
[Bug driver/33577] Compile failing for xpdf 3.02
--- Comment #3 from pcarlini at suse dot de 2007-09-28 00:10 --- (In reply to comment #2) (In reply to comment #1) g++: Internal error: Segmentation fault (program as) as is crashing so I am assuming you don't have an updated as from IBM. You mean I should upgrade my GNU++ compiler? I have version 3.3.2-5 No, he means that the error message says that the program *as*, *not* the compiler proper, is crashing. To be clear, that program is not part of GCC, you took it from IBM, together with the operating system (and you can of course collect more info about it via as -v or something very similar) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33577
[Bug rtl-optimization/7003] [ppc] address of array loaded int register twice without any need
--- Comment #7 from pinskia at gcc dot gnu dot org 2007-09-28 01:03 --- Subject: Bug 7003 Author: pinskia Date: Fri Sep 28 01:02:30 2007 New Revision: 128856 URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=128856 Log: 2007-09-27 Andrew Pinski [EMAIL PROTECTED] PR middle-end/7003 * gcc.target/powerpc/gcse-1.c: New test. Added: trunk/gcc/testsuite/gcc.target/powerpc/gcse-1.c Modified: trunk/gcc/testsuite/ChangeLog -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=7003
[Bug libfortran/33421] [4.3 Regression] Weird quotation of namelist output of character arrays
--- Comment #6 from jvdelisle at gcc dot gnu dot org 2007-09-28 03:10 --- I have reverted the offending patch and will reopen pr33253, the least of the two evils. This is one of those bugs where the fix for one breaks the other. I am still trying to puzzle it out. -- jvdelisle at gcc dot gnu dot org changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33421
[Bug testsuite/31828] FAIL: gcc.dg/float-range-[3-5].c (test for excess errors)
--- Comment #4 from danglin at gcc dot gnu dot org 2007-09-28 03:01 --- On the otherhand, maybe the tests should use __builtin_inff. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31828
[Bug c++/33580] New: ICE with -g on legal code: Boost lambda library test control_structures.cpp
It may be a duplicate of 31899. Compiling Boost lambda library test control_structures.cpp with -g results in internal compiler error: in reference_to_unused, at dwarf2out.c:10010. Without -g, it compiles cleanly. This is on HP-UX ia64 with gcc 4.2.1 How to reproduce: download Boost 1.34.1 and run x.sh below with BOOST_BUILD_PATH env. variable pointing to the Boost tree. As of this writing, the problem is also reproducible with sources from Boost SVN trunk x.sh #!/bin/sh -x export BOOST_BUILD_PATH=/proj/acxx/borisg/boost-v27/regression cd $BOOST_BUILD_PATH/boost/libs g++ -g -c -I.. ../libs/lambda/test/control_structures.cpp Snapshot: - bash-2.03$ uname -a HP-UX hpadl901 B.11.23 U ia64 2728209989 unlimited-user license bash-2.03$ g++ --version g++ (GCC) 4.2.1 Copyright (C) 2007 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. bash-2.03$ x.sh + export BOOST_BUILD_PATH=/proj/acxx/borisg/boost-v27/regression + cd /proj/acxx/borisg/boost-v27/regression/boost/libs + g++ -g -c -I.. ../libs/lambda/test/control_structures.cpp ../libs/lambda/test/control_structures.cpp:27: internal compiler error: in reference_to_unused, at dwarf2out.c:10010 Please submit a full bug report, with preprocessed source if appropriate. See URL:http://gcc.gnu.org/bugs.html for instructions. bash-2.03$ cat x.sh #!/bin/sh -x export BOOST_BUILD_PATH=/proj/acxx/borisg/boost-v27/regression cd $BOOST_BUILD_PATH/boost/libs g++ -g -c -I.. ../libs/lambda/test/control_structures.cpp bash-2.03$ -- Summary: ICE with -g on legal code: Boost lambda library test control_structures.cpp Product: gcc Version: 4.2.1 Status: UNCONFIRMED Severity: normal Priority: P3 Component: c++ AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: boris dot gubenko at hp dot com http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33580
[Bug testsuite/31828] FAIL: gcc.dg/float-range-[3-5].c (test for excess errors)
--- Comment #3 from danglin at gcc dot gnu dot org 2007-09-28 02:45 --- Is the use of macro FP_INFINITE in these tests a typo? It seems to me that the tests should be using the C99 macro INFINITY. In which case, the tests should be conditional on c99_runtime. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31828
[Bug libfortran/33421] [4.3 Regression] Weird quotation of namelist output of character arrays
--- Comment #5 from jvdelisle at gcc dot gnu dot org 2007-09-28 02:57 --- Subject: Bug 33421 Author: jvdelisle Date: Fri Sep 28 02:57:41 2007 New Revision: 128857 URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=128857 Log: 2007-09-27 Jerry DeLisle [EMAIL PROTECTED] PR libfortran/33421 * io/list_read.c (read_character): Revert r128057. Modified: trunk/libgfortran/ChangeLog trunk/libgfortran/io/list_read.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33421
[Bug libfortran/33253] namelist: reading back a string with apostrophe
--- Comment #10 from jvdelisle at gcc dot gnu dot org 2007-09-28 03:16 --- Reverted the patch to close pr33421. Reopening this to find the right solution. -- jvdelisle at gcc dot gnu dot org changed: What|Removed |Added Status|RESOLVED|REOPENED Resolution|FIXED | http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33253
[Bug ada/31108] [4.2 regression] ACATS C35507M fails
-- mmitchel at gcc dot gnu dot org changed: What|Removed |Added Priority|P3 |P5 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31108
[Bug ada/31174] [4.2 regression] ACATS C380004 fails
-- mmitchel at gcc dot gnu dot org changed: What|Removed |Added Priority|P3 |P5 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31174
[Bug ada/32452] [4.1//4.2 Regression] Incorrect type debugging information for variables in other compilation units
-- mmitchel at gcc dot gnu dot org changed: What|Removed |Added Priority|P3 |P5 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32452
[Bug c++/32470] [4.2/4.3 regression] fvisibility=hidden without effect in some cases
--- Comment #6 from mmitchel at gcc dot gnu dot org 2007-09-28 03:55 --- Andrew, thanks for the analysis. But, I'm not very happy with the patch. Can we change the compiler so that we set has_visibility on a binding level when the binding level is created -- and never reset it, ever? -- mmitchel at gcc dot gnu dot org changed: What|Removed |Added Keywords||wrong-code Priority|P3 |P1 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32470
[Bug libgomp/33131] [4.2 regression] libgomp/env.c:60: warning: implicit declaration of function 'strncasecmp'
-- mmitchel at gcc dot gnu dot org changed: What|Removed |Added Priority|P3 |P2 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33131
[Bug middle-end/33088] [4.1/4.2/4.3 Regression] spurious exceptions with -ffloat-store
--- Comment #3 from mmitchel at gcc dot gnu dot org 2007-09-28 03:59 --- I think Joseph's code is valid. All he's doing is initializing various parts of complex numbers and then using them. If using a fully initialized value results in a floating-point exception, that's a bug. -- mmitchel at gcc dot gnu dot org changed: What|Removed |Added Priority|P3 |P1 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33088
[Bug target/33008] [4.1/4.2 regression] code pessimization after -fforce-mem removal.
-- mmitchel at gcc dot gnu dot org changed: What|Removed |Added Priority|P3 |P2 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33008
[Bug tree-optimization/32901] [4.1/4.2/4.3 regression] bitfield constants with multiple bitfields take up space in .data
-- mmitchel at gcc dot gnu dot org changed: What|Removed |Added Priority|P3 |P2 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32901
[Bug bootstrap/25672] [4.1/4.2/4.3 regression] cross build's libgcc picks up CFLAGS
-- mmitchel at gcc dot gnu dot org changed: What|Removed |Added Priority|P3 |P2 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25672
