Configuring with: ../gcc-4.3.2/configure --prefix=/usr --bindir=/bin
--sbindir=/sbin --sysconfdir=/etc --libdir=/lib --localstatedir=/var
--enable-languages=c,c++ --enable-threads --enable-tls --with-system-zlib
results in wrong search path for c++ include files.
I can see in
--- Comment #2 from mikpe at it dot uu dot se 2008-10-09 07:04 ---
The program below illustrates the issue. It declares a private C99-compliant
snprintf() implementation and invokes it with %zu and %llx formats. This
triggers the following bogus warnings on x86_64-pc-mingw32:
Found at
http://groups.google.com/group/comp.lang.fortran/browse_thread/thread/70dd268bac2af9c9
The following program should be rejected as NAG f95,
Error: line 4: Subroutine FOO is not RECURSIVE
, and ifort,
error #6437: A subroutine or function is calling itself recursively. [FOO]
, do.
Follow up to PR 37635.
From trans-intrinsic.c:
/* LEADZ (i) = (i == 0) ? BIT_SIZE (i)
: __builtin_clz(i) - (BIT_SIZE('int') - BIT_SIZE(i))
The conditional expression is necessary because the result of LEADZ(0)
is defined, but the result of __builtin_clz(0) is
--- Comment #23 from cnstar9988 at gmail dot com 2008-10-09 07:16 ---
to Rainer Orth.
when apply your patch on gcc 4.3.2 release, cause PR3.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33100
--- Comment #3 from domob at gcc dot gnu dot org 2008-10-09 07:29 ---
Subject: Bug 35723
Author: domob
Date: Thu Oct 9 07:28:22 2008
New Revision: 141001
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=141001
Log:
2008-10-09 Daniel Kraft [EMAIL PROTECTED]
PR
--- Comment #4 from domob at gcc dot gnu dot org 2008-10-09 07:33 ---
Fixed.
--
domob at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #5 from jakub at gcc dot gnu dot org 2008-10-09 08:18 ---
Subject: Bug 37774
Author: jakub
Date: Thu Oct 9 08:17:08 2008
New Revision: 141003
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=141003
Log:
PR middle-end/37774
* tree.h
--- Comment #5 from paolo dot carlini at oracle dot com 2008-10-09 08:19
---
Yes, all those inserters, per 27.6.2.5.4/3 Require a non-null s.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37765
--- Comment #3 from cnstar9988 at gmail dot com 2008-10-09 11:25 ---
sorry, closed.
--
cnstar9988 at gmail dot com changed:
What|Removed |Added
--- Comment #7 from dominiq at lps dot ens dot fr 2008-10-09 09:27 ---
(In reply to comment #6)
On i686-apple-darwin9 I get:
[ibook-dhum] f90/bug% gfc pr37707.f90
[ibook-dhum] f90/bug% a.out
87 88 89
97 98 99
NAMLIS
A(1)%M=
--- Comment #6 from jakub at gcc dot gnu dot org 2008-10-09 11:29 ---
Fixed.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #8 from burnus at gcc dot gnu dot org 2008-10-09 13:31 ---
(In reply to comment #6)
Tobias, should A(3)%M not be 89??
Yes. I realized too late that I pasted the output of a slightly different
program, which I used during testing. In any case today's build gives the same
Boost header is not procceeded correctly and ended with error:
/usr/local/include/boost/mpl/apply_wrap.hpp:81:31: error: missing binary
operator before token (
The code is:
#elif BOOST_PP_ITERATION_DEPTH() == 1
where BOOST_PP_ITERATION_DEPTH() is defined as (for example):
#define
--- Comment #1 from jakub at gcc dot gnu dot org 2008-10-09 14:23 ---
Created an attachment (id=16481)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16481action=view)
gcc44-pr35760.patch
I think attached patch should fix it, it is after all what ELF does too
and
--- Comment #25 from ebotcazou at gcc dot gnu dot org 2008-10-09 11:22
---
to Rainer Orth.
when apply your patch on gcc 4.3.2 release, cause PR3.
Are you really sure? Please retry and close the PR otherwise.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33100
--- Comment #6 from schwab at suse dot de 2008-10-09 14:46 ---
*** Bug 37781 has been marked as a duplicate of this bug. ***
--
schwab at suse dot de changed:
What|Removed |Added
--- Comment #1 from schwab at suse dot de 2008-10-09 14:46 ---
*** This bug has been marked as a duplicate of 36453 ***
--
schwab at suse dot de changed:
What|Removed |Added
--- Comment #6 from pault at gcc dot gnu dot org 2008-10-09 09:15 ---
(In reply to comment #1)
NAMLIS
A(1)%M = 1,
A(1)%N = 5,
A(2)%M = 2,
A(2)%N = 6,
A(3)%M = 5,
A(3)%N = 99
/
--- Comment #24 from cnstar9988 at gmail dot com 2008-10-09 11:00 ---
works ok on gcc 4.3.2
--- config.gcc.4.3.2.oldThu Oct 9 18:51:26 2008
+++ config.gcc.4.3.2.newThu Oct 9 18:51:26 2008
@@ -1289,7 +1289,7 @@
*-*-solaris2.1[0-9]*)
tm_file=${tm_file}
The patch for PR37451 causes the stage2 ada compiler to be miscompiled.
$ ../prev-gcc/xgcc -B../prev-gcc/ -B../root/powerpc64-suse-linux/bin/ -c -g -O2
-gnatpg -gnata -gnatwns -nostdinc -I- -I. -Iada -I../../gcc/ada
-I../../gcc/ada/gcc-interface ../../gcc/ada/ada.ads -o ada/ada.o
fatal error:
the following code
double aaminusbb (const double a, const double b) { return (a+a)-(b+b); }
yields:
addsd %xmm0, %xmm0
mulsd .LC0(%rip), %xmm1
addsd %xmm1, %xmm0
ret
Where LC0 holds -2.0.
There's a perfectly good subtract instruction. Using it will avoid
--- Comment #1 from rrh at google dot com 2008-10-09 16:38 ---
I should have added that this is for -O, -O2, -O3.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37783
--- Comment #26 from ebotcazou at gcc dot gnu dot org 2008-10-09 17:35
---
We have filed case #65952072 with Sun to get this backported to Solaris 10.
Do you have any news about this?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33100
The function:
double dfunc_001 ( const double a, const double b) { return a b; }
yields the x86_64 code:
cmpltsd %xmm0, %xmm1
movapd %xmm1, %xmm0
movsd .LC1(%rip), %xmm1
andpd %xmm0, %xmm1
xorpd %xmm2, %xmm2
andnpd %xmm2, %xmm0
orpd
PLATFORM
PC i386 + GNU/Linux Ubuntu Hardy 8.04 + GCC 4.2.3
Compiling and Linking options: none.
Excuse my poor English. (By the way, what is Host triplet, Target triplet,
Build triplet?)
BUG:
printf really prints at terminal when finds '\n' character. So if I have a
simple printf iteration
--- Comment #1 from paolo dot carlini at oracle dot com 2008-10-09 18:36
---
In any case, sorry but this has nothing to do with the compiler.
--
paolo dot carlini at oracle dot com changed:
What|Removed |Added
--- Comment #2 from pinskia at gcc dot gnu dot org 2008-10-09 18:46 ---
FILEs only fushes its internal buffer when it encounters a newline or when done
explicitly. Do it explicitly, use fflush (stdout);.
Anyways this is not a compiler or a library issue.
--
gcc has flags -Bsymbolic and -Bsymbolic-functions
which are documented in the usage message.
$ gcc -v --help 21 | grep symbolic
-Bsymbolic Bind global references locally
-Bsymbolic-functionsBind global function references locally
However, the info pages are
--- Comment #1 from pinskia at gcc dot gnu dot org 2008-10-09 18:57 ---
The -Bsymbolic* options are linker options so they don't get documented in
GCC's documentation at all.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37786
--- Comment #2 from pinskia at gcc dot gnu dot org 2008-10-09 19:00 ---
Only a few systems support this option.
x86-linux-gnu is one of those targets that don't support this option :).
There is no changes needed here for the documentation as far as I can tell so
closing as
--- Comment #2 from pinskia at gcc dot gnu dot org 2008-10-09 19:03 ---
*** This bug has been marked as a duplicate of 19988 ***
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #13 from pinskia at gcc dot gnu dot org 2008-10-09 19:03
---
*** Bug 37783 has been marked as a duplicate of this bug. ***
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #3 from martinrb at google dot com 2008-10-09 19:26 ---
(In reply to comment #2)
Only a few systems support this option.
x86-linux-gnu is one of those targets that don't support this option :).
There is no changes needed here for the documentation as far as I can
--- Comment #3 from tkoenig at gcc dot gnu dot org 2008-10-09 19:30 ---
Subject: Bug 37753
Author: tkoenig
Date: Thu Oct 9 19:28:58 2008
New Revision: 141008
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=141008
Log:
2008-10-09 Thomas Koenig [EMAIL PROTECTED]
PR
The following program fails to recognize that the equivalence between qla1 and
qla2 causes a dependency in the assignment. Using qla1 on the right (ie
resolving the dependency by hand) makes the problem go away. It occurs for
both explicit constant subscripts and variable subscripts.
Dick
--- Comment #4 from jakub at gcc dot gnu dot org 2008-10-09 20:16 ---
Subject: Bug 37568
Author: jakub
Date: Thu Oct 9 20:15:17 2008
New Revision: 141010
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=141010
Log:
PR c++/37568
* semantics.c (finalize_nrv_r): Clear
--- Comment #5 from jakub at gcc dot gnu dot org 2008-10-09 20:20 ---
Fixed.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #3 from alexcastan at gmail dot com 2008-10-09 20:24 ---
Then the second problem it is also the normal behavior. The fork()
makes child inherit the unprinted parent buffer, then two similar buffers get
print later.
10 9 8 7 6 5 4 3 2 1 0 I'm the child
10 9 8 7 6 5 4 3
--- Comment #4 from pinskia at gcc dot gnu dot org 2008-10-09 20:34 ---
If you want to use a specific linker option, use -Wl,XYZ or -Xlinker XYZ .
Since those are the documented way to pass a linker option via gcc.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37786
The following valid code snippet triggers an ICE on mainline when compiled
with -fmudflap -O -g:
==
struct A
{
A();
};
void foo()
{
while (true)
A a;
}
==
bug.cc: In function 'void foo()':
bug.cc:10: internal compiler error: tree check: expected block,
--
reichelt at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|--- |4.4.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37788
--- Comment #7 from sam at gcc dot gnu dot org 2008-10-09 20:39 ---
Note that this has been fixed in Boost SVN repository recently
(http://svn.boost.org/trac/boost/ticket/2069).
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36453
--- Comment #9 from reichelt at gcc dot gnu dot org 2008-10-09 20:51
---
I just stumbled over the same bug and a similar ICE in vectorizable_store:
The following valid code snippet triggers an ICE on mainline when compiled
with -march=pentium4 -O2 -ftree-vectorize on
The following invalid code snippet triggers an ICE on mainline:
===
void foo():
{
__FUNCTION__;
}
===
bug.cc: In function 'void foo()':
bug.cc:2: error: only constructors take base initializers
bug.cc:2: error: expected identifier before '{' token
bug.cc:2:
--
reichelt at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|--- |4.4.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37789
--
burnus at gcc dot gnu dot org changed:
What|Removed |Added
CC||burnus at gcc dot gnu dot
|
--- Comment #4 from tkoenig at gcc dot gnu dot org 2008-10-09 21:57 ---
Fixed. Closing.
--
tkoenig at gcc dot gnu dot org changed:
What|Removed |Added
The test case gcc-c-torture/compile/limits-fnargs.c takes multiple hours to
compile at -O2 or above optimization on IA64. It looks like most of the time
is spent in IRA. The IRA pass calls reload which calls
ira_sort_regnos_for_alter_reg with n = 28977. This in turn calls
coalesce_spill_slots
--- Comment #6 from tdragon at tdragon dot net 2008-10-09 22:57 ---
Ping! Any ideas on this bug?
--
tdragon at tdragon dot net changed:
What|Removed |Added
Known to
--- Comment #1 from jakub at gcc dot gnu dot org 2008-10-09 23:09 ---
That has been fixed already.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #5 from martinrb at google dot com 2008-10-09 23:10 ---
(In reply to comment #4)
If you want to use a specific linker option, use -Wl,XYZ or -Xlinker XYZ .
Since those are the documented way to pass a linker option via gcc.
I know about -Wl, and -Xlinker.
I am trying to
--- Comment #6 from pinskia at gcc dot gnu dot org 2008-10-09 23:15 ---
-B is documented on
http://gcc.gnu.org/onlinedocs/gcc-4.3.2/gcc/Directory-Options.html:
-Bprefix
This option specifies where to find the executables, libraries, include files,
and data files of the compiler itself.
--- Comment #7 from martinrb at google dot com 2008-10-09 23:16 ---
The flag `-Bsymbolic' works (or at least is accepted)
despite the fact that it is not documented that such a
flag is passed to the linker.
Ops.
Please ignore my previous msg.
I have misunderstood for
A few Mauve tests now throw an OutOfMemory error following the merge up to
Classpath 0.98:
TEST: java.lang.Character.Blocks15
uncaught exception:
java.lang.OutOfMemoryError
java.lang.OutOfMemoryError
No stacktrace available
Strangely, the bug seems to be unreproducible outside the test
--- Comment #27 from cnstar9988 at gmail dot com 2008-10-10 00:42 ---
works well on gcc 4.3.2 + patch.
I don't build libgcc_s.so.
gas 2.18 + gmp 4.2.4 + mpfr 2.3.2.
Configured with: ../src/configure --prefix=/opt/gcc-4.3.2
--with-gmp=/opt/gcc-4.3.2 --with-mpfr=/opt/gcc-4.3.2
--- Comment #21 from cnstar9988 at gmail dot com 2008-10-10 00:56 ---
Does this patch works well on 4.3?
Thanks!
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37102
--- Comment #9 from jvdelisle at gcc dot gnu dot org 2008-10-10 02:11
---
Subject: Bug 37707
Author: jvdelisle
Date: Fri Oct 10 02:10:14 2008
New Revision: 141016
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=141016
Log:
2008-10-08 Jerry DeLisle [EMAIL PROTECTED]
PR
--- Comment #10 from jvdelisle at gcc dot gnu dot org 2008-10-10 02:14
---
This is fixed on trunk. I think I will back port this to 4.3 since the bug
does result in failing in valid namelists.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37707
--- Comment #28 from ebotcazou at gcc dot gnu dot org 2008-10-10 05:14
---
works well on gcc 4.3.2 + patch.
Thanks but this patch is not suitable for the branch, only Rainer's is. And
you didn't say whether you can reproduce PR 3 so I'm closing it.
--
The following program gives an ICE:
internal compiler error: in gfc_conv_array_initializer, at
fortran/trans-array.c:3976
However, it works when one uses -fno-range-check
See: http://gcc.gnu.org/ml/fortran/2008-10/msg00078.html
--
Summary: ICE in gfc_conv_array_initializer; works
61 matches
Mail list logo
