--- Comment #7 from dorit at gcc dot gnu dot org 2007-05-01 07:59 ---
Subject: Bug 31589
Author: dorit
Date: Tue May 1 07:58:59 2007
New Revision: 124315
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=124315
Log:
PR testsuite/31589
* gcc.dg/vect/vect-iv-9.c:
--- Comment #15 from pluto at agmk dot net 2007-05-01 08:58 ---
(In reply to comment #14)
typed_rep points to:
N4sigc8internal14typed_slot_repINS_12bind_functorILin1ENS_16pointer_functor1IPvS4_EEPl
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30252
--- Comment #5 from rob1weld at aol dot com 2007-05-01 09:16 ---
This make -i check report shows Java compiled with ZERO errors:
http://gcc.gnu.org/ml/gcc-testresults/2007-04/msg01490.html
gcc version 4.2.0 20070427 (prerelease)
---
=== libjava tests ===
Running target unix
===
--- Comment #2 from rob1weld at aol dot com 2007-05-01 10:19 ---
Successfully _forced_ the compile on Cygwin platform and posted a seperate
report about the segmentation fault issue at:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31761
After running make -i check on Cygwin compile I
--
dfranke at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |dfranke at gcc dot gnu dot
|dot org
/home/richard/src/trunk2/gcc/testsuite/g++.dg/gomp/pr26913.C: In function 'void
_Z3bazv.omp_fn.0(void*)':^M
/home/richard/src/trunk2/gcc/testsuite/g++.dg/gomp/pr26913.C:14: error:
unnecessary EH edge 5-6^M
/home/richard/src/trunk2/gcc/testsuite/g++.dg/gomp/pr26913.C:14: internal
compiler error:
--- Comment #1 from rguenth at gcc dot gnu dot org 2007-05-01 11:17 ---
Zdenek, this may be yours.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #9 from dgregor at gcc dot gnu dot org 2007-05-01 13:32 ---
The easy fix is to SET_TYPE_STRUCTURAL_EQUALITY (fulltype), which will force
structural equality checks rather than use canonical types. I usually don't
like this solution, but it's better than duplicating all of
--- Comment #16 from rguenth at gcc dot gnu dot org 2007-05-01 13:35
---
Created an attachment (id=13468)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13468action=view)
slightly simplified testcase
I don't see anything wrong with the testcase. I changed it to not take the
--- Comment #17 from rguenth at gcc dot gnu dot org 2007-05-01 14:00
---
Without -fstrict-aliasing we mark this_20-functor_ = D.2946; obviously
necessary
because it is is_hidden_global_store ():
;; Function sigc::slot0 A::bar() (_ZN1A3barEv)
-Marking useful stmt: this_20-functor_ =
--- Comment #18 from rguenth at gcc dot gnu dot org 2007-05-01 14:05
---
So my guess is this is related to PR22488 which is rotting since some time...
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
--
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
--- Comment #1 from jakub at gcc dot gnu dot org 2007-05-01 14:10 ---
Zdenek, this seems to be also related to your cfg cleanup changes.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #8 from tkoenig at gcc dot gnu dot org 2007-05-01 14:11 ---
Subject: Bug 31732
Author: tkoenig
Date: Tue May 1 14:11:36 2007
New Revision: 124326
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=124326
Log:
2007-05-01 Thomas Koenig [EMAIL PROTECTED]
PR
--- Comment #9 from tkoenig at gcc dot gnu dot org 2007-05-01 14:13 ---
Fixed. Closing.
--
tkoenig at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #2 from tkoenig at gcc dot gnu dot org 2007-05-01 14:18 ---
Closely related to PR 31251.
--
tkoenig at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #4 from jvdelisle at gcc dot gnu dot org 2007-05-01 14:47
---
With IMPLICIT NONE the double code path goes away and we get just one correct
error message.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31251
--- Comment #3 from jvdelisle at gcc dot gnu dot org 2007-05-01 14:55
---
As with pr31251, I do not see the segfault here.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31716
When I try compile a small program I get an error.
[EMAIL PROTECTED]:~ gcc -Wall prog.c -O2 -S
prog.c: In function main:
prog.c:21: error: Attempt to delete prologue/epilogue insn:
(insn/f 92 91 93 0 (set (mem:SI (plus:SI (reg/f:SI 6 bp)
(const_int -16 [0xfff0])) [0 S4
--- Comment #1 from pinskia at gcc dot gnu dot org 2007-05-01 15:36 ---
When I try compile a small program I get an error.
small undefined code at that :).
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #2 from bangerth at dealii dot org 2007-05-01 17:01 ---
int x[4];
for(n=0;n5;n++)
x[n]=10;
You're asking for trouble here.
That said, we shouldn't produce a compiler error. I can't seem to reproduce
this, what platform is this on?
W.
--
bangerth at dealii
I had a typo in my code, id = + 1i that baffled me as to why it compiled.
Once I realized that it was legal, I noticed that it was producing the wrong
result (aside from the obvious typo).
It appears that numberi always becomes 0 inside gcc. The cut-and-paste
below is from my computer at my
--- Comment #1 from pinskia at gcc dot gnu dot org 2007-05-01 17:58 ---
1i is complex integer where the imangary part is 1 and the real part is 0.
See http://gcc.gnu.org/onlinedocs/gcc-4.1.2/gcc/Complex.html
This is a gcc extension.
--
pinskia at gcc dot gnu dot org changed:
--- Comment #9 from mstein at phenix dot rootshell dot be 2007-05-01 17:59
---
Created an attachment (id=13469)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13469action=view)
from src/sim/frv
This attachment triggers a ICE in set_value_range, at tree-vrp.c:278
with a host gcc
--- Comment #2 from schlake+gccbug at nmt dot edu 2007-05-01 18:05 ---
Ok, thanks. The other compiler we use here at work uses the i suffix to mean
integer, just like l/L for long.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31773
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-05-01 18:14 ---
Reopening this one as it is not a dup.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Severity|normal |blocker
Component|target |middle-end
--- Comment #3 from ian at gcc dot gnu dot org 2007-05-01 18:52 ---
Subject: Bug 31739
Author: ian
Date: Tue May 1 18:51:56 2007
New Revision: 124334
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=124334
Log:
PR tree-optimization/31739
* tree-vrp.c
--- Comment #10 from ian at airs dot com 2007-05-01 18:58 ---
Please let me know if this is still a problem after revision 124334.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31710
--- Comment #4 from ian at airs dot com 2007-05-01 18:59 ---
Fixed on mainline.
--
ian at airs dot com changed:
What|Removed |Added
CC|ian at gcc dot gnu
--- Comment #3 from pinskia at gcc dot gnu dot org 2007-05-01 19:15 ---
Reduced testcase:
typedef signed int signed16 __attribute__ ((__mode__ (__HI__)));
typedef unsigned int unsigned16 __attribute__ ((__mode__ (__HI__)));
typedef signed16 HI;
typedef unsigned16 UHI;
unsigned short
--- Comment #3 from bangerth at dealii dot org 2007-05-01 19:16 ---
(In reply to comment #2)
Ok, thanks. The other compiler we use here at work uses the i suffix to mean
integer, just like l/L for long.
This is definitely non-standard. The 'i' suffix is for the imaginary part of
--- Comment #4 from pinskia at gcc dot gnu dot org 2007-05-01 19:26 ---
Yes this was fixed after:
2007-05-01 Ian Lance Taylor [EMAIL PROTECTED]
PR tree-optimization/31739
* tree-vrp.c (vrp_val_is_max): New static function.
(vrp_val_is_min): New static
middle-end/31740
* gcc.c-torture/compile/20070501-1.c: New testcase.
Added:
trunk/gcc/testsuite/gcc.c-torture/compile/20070501-1.c
Modified:
trunk/gcc/testsuite/ChangeLog
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31740
I have problems linking executable with C++ library with --as-needed ld flag
while linking with gcc succeeds. During investigation I found inconsistency in
gcc and g++ calls concerning --as-needed. First I will show you what I do next
I'll summarize what is this bug about:
* What I do:
First I
In [basic.link] paragraph 6, there's an example that shows that (unlike in C)
it is permissible to define an object 'static' in a namespace scope and then
have another object which is 'extern', and reference both in the same
translation unit.
The compiler optimises that example so that there's no
--- Comment #1 from geoffk at gcc dot gnu dot org 2007-05-01 19:56 ---
This testcase is the same principle, but might use a different code path in the
compiler:
extern C void abort();
extern int *p;
int main()
{
extern int i;
i = 1;
*p = 2;
if (i == 2)
abort ();
return
--- Comment #5 from ian at gcc dot gnu dot org 2007-05-01 20:23 ---
Subject: Bug 31739
Author: ian
Date: Tue May 1 20:23:47 2007
New Revision: 124338
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=124338
Log:
PR tree-optimization/31739
* tree-vrp.c
--- Comment #2 from rguenth at gcc dot gnu dot org 2007-05-01 20:44 ---
How do you define main::i? That is, how'd you make the testcase work from a
second translation unit if it would fail now?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31775
--- Comment #1 from pinskia at gcc dot gnu dot org 2007-05-01 21:03 ---
I think this is really a libtool bug passing -nostdlib to g++, there is no
reason why it should be doing that anyways.
Also --as-needed is a binutils option, since the pthreads library is not needed
by the main
--- Comment #1 from sje at cup dot hp dot com 2007-05-01 21:22 ---
This does not look like a GCC bug, you just need to link in the YACC library on
the link line. The lex library (-ll) has calls to the yacc library (-ly) and
so that library needs to be linked in.
--
sje at cup dot
The current svn repository does not bootstrap on my machine.
$ svn info
Path: .
URL: svn://gcc.gnu.org/svn/gcc/trunk
Repository Root: svn://gcc.gnu.org/svn/gcc
Repository UUID: 138bc75d-0d04-0410-961f-82ee72b054a4
Revision: 124341
Node Kind: directory
Schedule: normal
Last Changed Author: dwarak
--- Comment #1 from pinskia at gcc dot gnu dot org 2007-05-01 22:03 ---
How did you configure?
This is most likely the same as reported in
http://gcc.gnu.org/ml/gcc-patches/2007-05/msg00033.html
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31776
--- Comment #3 from rakdver at gcc dot gnu dot org 2007-05-01 22:24 ---
I am checking this; so far it looks like some memory corruption.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31762
in revision 83355 of libstdc++-v3/include/ext/pool_allocator.h the efficient
thread support code for GLIBCXX_FORCE_NEW was broken by changing the test
(_S_force_new0) to (_S_force_new==1). now there is a race condition.
--
Summary: GLIBCXX_FORCE_NEW doesn't always work in
--- Comment #2 from schnetter at aei dot mpg dot de 2007-05-01 22:37
---
I configured with the options
~/src/gcc/configure --prefix=$HOME/gcc --with-gmp=/opt/local
--with-mpfr=/opt/local
(outside the source tree).
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31776
--- Comment #3 from rguenth at gcc dot gnu dot org 2007-05-01 22:57 ---
It looks like the same cause but different symptom. The configure
check for strsignal for darwin doesn't seem to work.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31776
--- Comment #4 from rakdver at gcc dot gnu dot org 2007-05-01 23:02 ---
To tree-data-ref.c:add_multivariate_self_dist, we get with
c_2 = {{0, +, 30}_1, +, 1}_2
but
ddr-loop_nest contains only loops 2 and 3.
x_1 is then set to 2, which is outside of the bounds of the dist_v
array.
genattrtab.c has
struct insn_def
{
struct insn_def *next;/* Next insn in chain. */
rtx def; /* The DEFINE_... */
int insn_code;/* Instruction number. */
int insn_index; /* Expression numer in file, for errors. */
int lineno;
One more potential conspiracy item against gcc-4.2.0.
The following three items were obtained on three different machines sporting
late gcc-4.2.0
versions. (less than one week old)
cmake: relocation error: cmake: symbol _ZNSo9_M_insertEPKci, version
GLIBCXX_3.4.9 not defined in file
--- Comment #1 from pcarlini at suse dot de 2007-05-01 23:36 ---
Obviously, you are linking a binary compiled with 4.2.0 to a 4.1.2 library.
That kind of backward compatibility is not provided and never was in the
past.
--
pcarlini at suse dot de changed:
What
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-05-01 23:39 ---
(In reply to comment #1)
Obviously, you are linking a binary compiled with 4.2.0 to a 4.1.2 library.
That kind of backward compatibility is not provided and never was in the
past.
That is called forwards
--- Comment #3 from pcarlini at suse dot de 2007-05-01 23:41 ---
(In reply to comment #2)
That is called forwards compatibility...
The name depends on the point of view ;) No, seriusly, thanks about the name, I
always get this one wrong.
--
--- Comment #1 from pcarlini at suse dot de 2007-05-01 23:57 ---
Gosh you are right, will fix it asap. Thanks.
--
pcarlini at suse dot de changed:
What|Removed |Added
--
pcarlini at suse dot de changed:
What|Removed |Added
Target Milestone|--- |4.2.1
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31777
--- Comment #4 from malitzke at metronets dot com 2007-05-02 00:20 ---
I accept that there is something wrong on my side. Be it forward or
backward.
However, there are some things that I still do not understand.
Cmake was compiled several times with different bootstrapped gcc-4.2.0
The following invalid code snippet triggers an ICE on mainline:
===
bool foo();
void bar()
{
__builtin_expect(foo(), 1) ? 0i : 0;
}
===
bug.cc: In function 'void bar()':
bug.cc:5: error: could not convert '0' to
--
reichelt at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|--- |4.3.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31780
--- Comment #3 from geoffk at gcc dot gnu dot org 2007-05-02 00:54 ---
You would add a translation unit that says
int i;
or similar. It's not main::i, it's ::i, because of [basic.link] paragraph
7:
When a block scope declaration of an entity with linkage is not found to refer
to
--- Comment #4 from geoffk at gcc dot gnu dot org 2007-05-02 01:46 ---
I just happen to have a patch which fixes this.
--
geoffk at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #1 from pinskia at gcc dot gnu dot org 2007-05-02 01:46 ---
bug.cc:5: error: could not convert '0' to 'int __complex__'
Actually it should be able to convert that and I think this is the same reason
why we are rejecting PR 30209.
--
pinskia at gcc dot gnu dot org
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-05-02 01:47 ---
Also I think this was caused by PR 31338.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #12 from crowl at google dot com 2007-05-02 02:24 ---
(In reply to comment #10)
I am not convinced that the code in Comment #8 is valid.
Although the operand of sizeof is not in fact evaluated, it seems odd to
permit an operand which cannot, even in principle, be
if you --disable-checking when you build, you get additional fortran
regressions. This is true on the current trunk and has most likely been true
for at least two weeks.
FAIL: gfortran.dg/repeat_2.f90 -O0 execution test
FAIL: gfortran.dg/repeat_2.f90 -O1 execution test
FAIL:
Hi,
This code compiles fine with gcc:
void a$() {}
int main() { a$(); }
This generates invalid assembly code:
void $() {}
int main() { $(); }
To gas, an initial $ is not allowed in an identifier, but to gcc, it is, so gcc
can't simply copy the function name to the assembly output, unless $
--- Comment #1 from truedfx at gentoo dot org 2007-05-02 03:21 ---
Sorry, forgot to mention, I'm using gcc 4.1.2 with binutils 2.17, and I also
tried gcc 3.3.6 with binutils 2.16.1, which behaves the same for me.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31782
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-05-02 03:33 ---
unless $ also becomes an invalid identifier in C.
Well read:
http://gcc.gnu.org/onlinedocs/gcc-4.1.2/gcc/Dollar-Signs.html
it is can be but does not have to be.
This works for me targetting spu-elf and
--- Comment #1 from pinskia at gcc dot gnu dot org 2007-05-02 03:34 ---
Most likely someone has a gcc_assert which has side effects.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
gcc 4.1.2 and gcc 4.1.1 shows the following bogus warning message.
[EMAIL PROTECTED] ~/tests] arm-rtems-gcc -Os -Wall -S test.cpp
test.cpp: In member function 'int attr::create(int)':
test.cpp:17: warning: control reaches end of non-void function
the minimal testcase:
class attr
{
attr() {}
--- Comment #5 from jvdelisle at gcc dot gnu dot org 2007-05-02 04:13
---
Created an attachment (id=13470)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13470action=view)
Possible patch for this problem
The double error comes from trying to match_type_spec for a variable
--- Comment #4 from jvdelisle at gcc dot gnu dot org 2007-05-02 04:19
---
I attached a patch to pr31251, can someone try that and see what effect it has
on this one.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31716
--
jvdelisle at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNCONFIRMED |WAITING
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31467
--- Comment #3 from jvdelisle at gcc dot gnu dot org 2007-05-02 04:57
---
backtrace is uggly, repeats the following for a long long time
#3554 0x0043c63e in mio_expr (ep=value optimized out)
at ../../gcc43/gcc/fortran/module.c:2133
#3555 0x0043c779 in mio_expr
--- Comment #5 from rob1weld at aol dot com 2007-05-02 05:08 ---
Thanks Andreas,
The annoying un-quoted second occurance of the languages is now fixed.
Somehow AutoConf decided that it was going to rebuild ./configure .
I renamed my current configure and then did a new SVN to get the
--- Comment #3 from truedfx at gentoo dot org 2007-05-02 05:58 ---
Thanks for the link. I don't see how GAS could be fixed, though. How would the
assembler tell the difference between $0 the constant and $0 the identifier?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31782
--- Comment #2 from pva at gentoo dot org 2007-05-02 06:56 ---
libtool bug was discussed earlier:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25460
But this bug is not about libtool. Libtool just expose problem, although it
should hide but that is another bug.
I may wish to have
76 matches
Mail list logo
