--- Comment #2 from reichelt at gcc dot gnu dot org 2007-08-29 06:55
---
The second testcase is fixed on mainline by Paolo's patch for PR 33209.
It's still open on the branches, though.
Nothing changed for the first testcase.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33210
--- Comment #4 from reichelt at gcc dot gnu dot org 2007-08-29 06:41
---
Fixed, indeed.
--
reichelt at gcc dot gnu dot org changed:
What|Removed |Added
Statu
--- Comment #5 from daney at gcc dot gnu dot org 2007-08-29 06:19 ---
Created an attachment (id=14130)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14130&action=view)
New test.sh
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33218
--- Comment #4 from daney at gcc dot gnu dot org 2007-08-29 06:18 ---
Created an attachment (id=14129)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14129&action=view)
Test case that works.
With the new "Test case that works" and attached test.sh and the original
test.sh I get no
--- Comment #38 from bangerth at dealii dot org 2007-08-29 03:15 ---
(In reply to comment #37)
> No it is not, it is still violating One definition rule as struct X
> will have a different member type for pimpl_ in each TU (this violates
> the whole idea of types being exported).
It is
--- Comment #4 from igodard at pacbell dot net 2007-08-29 02:27 ---
OK, I see. I doubt I'm the only one who is confused by the report of a mis-used
obscure gcc-only "feature" instead of an all-too-common parenthesis error :-)
However, if you are emitting diagnostics on the fly then you
--- Comment #7 from jvdelisle at gcc dot gnu dot org 2007-08-29 02:26
---
Subject: Bug 33055
Author: jvdelisle
Date: Wed Aug 29 02:26:01 2007
New Revision: 127877
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127877
Log:
2007-08-28 Jerry DeLisle <[EMAIL PROTECTED]>
--- Comment #3 from pinskia at gcc dot gnu dot org 2007-08-29 02:10 ---
(In reply to comment #2)
> but it knows that "confirm" is a function template at this point, and can't be
> a future label.
Yes it can.
Try:
template
bool confirm(T t, int i) { return t == i; }
int main() {
int
--- Comment #2 from jvdelisle at gcc dot gnu dot org 2007-08-29 01:27
---
I think the problem has to do with not having a backend_decl associated with
the dummy variable expression created. I will not be able to get to this
quickly, so I will revert the patch until I get this fixed.
T
--- Comment #2 from igodard at pacbell dot net 2007-08-29 01:02 ---
Why does it think a template is a label? I would understand if the code were:
lab: if (i == 0) && lab) ...
but it knows that "confirm" is a function template at this point, and can't be
a future label.
--
http:
--
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
--- Comment #5 from pinskia at gcc dot gnu dot org 2007-08-29 01:00 ---
Was caused by:
2007-08-03 Andrew Pinski <[EMAIL PROTECTED]>
PR middle-end/32399
* tree-ssa-address.c (add_to_parts): Use POINTER_PLUS_EXPR
when adding to the base and convert ELT to sizetyp
--- Comment #5 from pinskia at gmail dot com 2007-08-29 00:56 ---
Subject: Re: GCC outputs invalid assembly when using -O2
On 29 Aug 2007 00:53:35 -, Hans dot Boehm at hp dot com
<[EMAIL PROTECTED]> wrote:
> Based on Book 2, section 1.4 of the architecture manual, I believe aligned
On 29 Aug 2007 00:53:35 -, Hans dot Boehm at hp dot com
<[EMAIL PROTECTED]> wrote:
> Based on Book 2, section 1.4 of the architecture manual, I believe aligned
> word
> access are always atomic on PowerPC.
They are, even cache-inheited memory too.
-- Pinski
--- Comment #4 from Hans dot Boehm at hp dot com 2007-08-29 00:53 ---
I'll make the %X1 change upstream. Are there similar issues with some of the
other routines?
(See
http://bdwgc.cvs.sourceforge.net/bdwgc/bdwgc/libatomic_ops-1.2/src/atomic_ops/sysdeps/gcc/powerpc.h?revision=1.3&view=
--- Comment #4 from pinskia at gcc dot gnu dot org 2007-08-29 00:45 ---
By the way, the ICE in build2_stat are very very target dependent because of
IV-OPTs.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
---
--- Comment #3 from pinskia at gcc dot gnu dot org 2007-08-29 00:37 ---
(In reply to comment #1)
> Maybe caused by this change?
No it could not have been caused by that as it made the assert weaker, not
stronger :).
--
pinskia at gcc dot gnu dot org changed:
What|Rem
--- Comment #2 from patchapp at dberlin dot org 2007-08-29 00:14 ---
Subject: Bug number PR 33215
A patch for this bug has been added to the patch tracker.
The mailing list url for the patch is
http://gcc.gnu.org/ml/gcc-patches/2007-08/msg02019.html
--
http://gcc.gnu.org/bugzilla/s
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|--- |4.3.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33181
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-08-29 00:12 ---
/usr/home/mckelvey/software/gcc-obj/./gcc/cc1plus: relocation error:
/usr/home/mckelvey/software/gcc-obj/./gcc/cc1plus: undefined symbol:
__gmpz_import
Sounds like you have two different versions of GMP installed a
--- Comment #3 from pinskia at gcc dot gnu dot org 2007-08-29 00:06 ---
Fixed.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNCONFIRM
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|--- |4.3.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33184
--- Comment #1 from pinskia at gcc dot gnu dot org 2007-08-29 00:06 ---
Not really since this is &&label which is valid GNU C++.
So this is not really a bad diagnostic for GNU C++.
if you add -pedantic, you get:
t.cc:6: error: taking the address of a label is non-standard
Which tells
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Summary|FAIL: gcc.c-|[4.3 Regression] FAIL:
|torture/execute/20010119-1.
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-08-29 00:00 ---
I just did a bootstrap and got past this point. Can you attach the config.log
from your build directory for the libiberty?
--
pinskia at gcc dot gnu dot org changed:
What|Removed
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|--- |4.3.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33134
--- Comment #3 from pinskia at gcc dot gnu dot org 2007-08-28 23:57 ---
The seg fault was fixed by:
2007-08-28 Jakub Jelinek <[EMAIL PROTECTED]>
PR middle-end/32370
* passes.c (init_optimization_passes): Move pass_df_finish
after pass_postreload sublist.
Also
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Keywords||ice-on-valid-code
Target Milestone|--- |4.3
--- Comment #3 from eweddington at cso dot atmel dot com 2007-08-28 23:31
---
(In reply to comment #2)
> Confirmed bug, and verified patch fixes bug.
>
... For the AVR port only.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33028
--- Comment #2 from eweddington at cso dot atmel dot com 2007-08-28 23:30
---
Confirmed bug, and verified patch fixes bug.
--
eweddington at cso dot atmel dot com changed:
What|Removed |Added
---
On 28 Aug 2007 21:41:05 -, bangerth at dealii dot org
<[EMAIL PROTECTED]> wrote:
> That's what I thought too at first, but the request is valid anyway as long
> as you never use the pointer to the class from anywhere except the one
> implementation file in which the class is defined.
No it is
--- Comment #37 from pinskia at gmail dot com 2007-08-28 22:11 ---
Subject: Re: Unnecessary anonymous namespace warnings
On 28 Aug 2007 21:41:05 -, bangerth at dealii dot org
<[EMAIL PROTECTED]> wrote:
> That's what I thought too at first, but the request is valid anyway as long
>
--- Comment #36 from bangerth at dealii dot org 2007-08-28 21:40 ---
(In reply to comment #35)
> In this case, the warning is correct as Impl will be different in each
> Translation Unit so X can never be the same between two of them.
That's what I thought too at first, but the request
--- Comment #1 from dgregor at gcc dot gnu dot org 2007-08-28 21:12 ---
Patch available here: http://gcc.gnu.org/ml/gcc-patches/2007-08/msg02013.html
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33194
--- Comment #18 from andreast at gcc dot gnu dot org 2007-08-28 21:00
---
Adding -fno-dce at -O1/O2 helps passing the failing test cases mentioned in
comment #8 on ppc-linux and ppc-darwin.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32758
--- Comment #35 from pinskia at gmail dot com 2007-08-28 20:48 ---
Subject: Re: Unnecessary anonymous namespace warnings
On 28 Aug 2007 19:40:14 -, pluto at agmk dot net
<[EMAIL PROTECTED]> wrote:
>
>
> --- Comment #34 from pluto at agmk dot net 2007-08-28 19:40 ---
> (In
On 28 Aug 2007 19:40:14 -, pluto at agmk dot net
<[EMAIL PROTECTED]> wrote:
>
>
> --- Comment #34 from pluto at agmk dot net 2007-08-28 19:40 ---
> (In reply to comment #33)
> > Fixed.
> >
>
> one more testcase:
>
> $ cat X.hpp
> namespace { class Impl; }
> struct X
> {
> ~X();
--
dgregor at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |dgregor at gcc dot gnu dot
|dot org
--- Comment #4 from tbm at cyrius dot com 2007-08-28 20:32 ---
(In reply to comment #3)
> Has anyone been able to reproduce this problem? I'm not able to trigger the
> failure with the preprocessed source.
Well, I noted this problem already in my first message. I see the ICE when
I com
--- Comment #3 from dgregor at gcc dot gnu dot org 2007-08-28 20:26 ---
Has anyone been able to reproduce this problem? I'm not able to trigger the
failure with the preprocessed source.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33185
--- Comment #3 from reichelt at gcc dot gnu dot org 2007-08-28 20:16
---
The message "confused by earlier errors, bailing out" is nothing but a hidden
ICE.
In order not to bother the user with GCC's error recovery problems it replaces
the usual ICE message on release branches, if a regu
--- Comment #2 from reichelt at gcc dot gnu dot org 2007-08-28 20:06
---
It's a duplicate, indeed.
*** This bug has been marked as a duplicate of 33118 ***
--
reichelt at gcc dot gnu dot org changed:
What|Removed |Added
-
--- Comment #1 from reichelt at gcc dot gnu dot org 2007-08-28 20:06
---
*** Bug 33214 has been marked as a duplicate of this bug. ***
--
reichelt at gcc dot gnu dot org changed:
What|Removed |Added
-c X.cpp
In file included from X.cpp:1:
X.hpp:3: warning: 'X' has a field 'X::pimpl_' whose type uses the anonymous
namespace
gcc version 4.2.2 20070828
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29365
--- Comment #17 from jakub at gcc dot gnu dot org 2007-08-28 19:38 ---
I can reproduce this on ppc32-linux, working on a self-contained reproducer.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32758
--- Comment #8 from jason at gcc dot gnu dot org 2007-08-28 19:32 ---
Fixed for 4.3, not going to bother applying to other branches.
--
jason at gcc dot gnu dot org changed:
What|Removed |Added
--
--- Comment #5 from kauer at os dot inf dot tu-dresden dot de 2007-08-28
19:27 ---
(4) abort produce a proper backtrace, if it is not a noreturn function
(5) this optimization could be configurable by something like
-foptimize-noreturns
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?i
--- Comment #18 from rguenth at gcc dot gnu dot org 2007-08-28 19:23
---
Just to add, the reduced testcase fails with r127848. I suppose the failure
might vanish with the inliner change I put in after that.
--
rguenth at gcc dot gnu dot org changed:
What|Removed
--- Comment #4 from kauer at os dot inf dot tu-dresden dot de 2007-08-28
19:07 ---
Some comments from the outside:
(1) It should be possible to exchange the emmited call into a jmp, without
adding the function epilog.
(2) A proper backtrace is not generated on optimized sibling function
--- Comment #16 from andreast at gcc dot gnu dot org 2007-08-28 18:44
---
Volumes/development/gcc/head/objdir/./gcc/xgcc
-B/Volumes/development/gcc/head/objdir/./gcc/
-B/Volumes/development/gcc/head/testbin/powerpc-apple-darwin8.10.0/bin/
-B/Volumes/development/gcc/head/testbin/powerpc-
--- Comment #14 from janis at gcc dot gnu dot org 2007-08-28 18:15 ---
The patch that I tried is the one attached for comment #10, not the one in
comment #9. With that patch a bootstrap of all languages but Ada succeeded.
The test results look reasonable, although I don't have results
--- Comment #1 from pcarlini at suse dot de 2007-08-28 18:10 ---
Oops..
--
pcarlini at suse dot de changed:
What|Removed |Added
AssignedTo|pcarlini at suse dot de
The following is a valid Fortran 2003 program, but if "type t" has no
components, it cannot be imported:
Error: Derived type 't' at (1) is being used before it is defined
type t
! integer :: i
end type t
interface
subroutine fff(a)
import :: t
type(t) :: a
end subroutine fff
end inte
cc -O2 -dr -msoft-float -c f_powf.c
f_powf.c: In function Â_f_powfÂ:
f_powf.c:49: error: impossible constraint in ÂasmÂ
typedef unsigned int __uint32_t;
typedef signed int __int32_t;
typedef union
{
double value;
struct
{
__uint32_t lsw;
__uint32_t msw;
} parts;
} ieee_double_s
--
pcarlini at suse dot de changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |pcarlini at suse dot de
|dot org |
--- Comment #4 from jason at gcc dot gnu dot org 2007-08-28 16:59 ---
Subject: Bug 29000
Author: jason
Date: Tue Aug 28 16:58:59 2007
New Revision: 127868
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127868
Log:
PR c++/29000
* pt.c (build_non_dependent_expr, ty
--- Comment #3 from daney at gcc dot gnu dot org 2007-08-28 16:56 ---
Looking at the current code, it seems that we may have a problem if we
destroy() a process that has already exited. The kill(2) man page suggests
that ESRCH could result, in which case we would throw an InternalError.
--- Comment #5 from andreagrassi at sogeasoft dot com 2007-08-28 16:50
---
Subject: R: Error in compiling when there is a function with a char parameter
called before its declaration with inline parameters.
Yes and the error remains. Nothing changes.
I don't understand the error !! I
--- Comment #2 from cyberflex at mail dot ru 2007-08-28 16:43 ---
(In reply to comment #1)
> Can you post a fully self contained test case? If I can easily reproduce it,
> I
> will try to fix it.
>
Test case is to be following, but reproducing looks like to be a bit tricky :(
gcj (GC
--- Comment #2 from pcarlini at suse dot de 2007-08-28 16:43 ---
Fixed.
--
pcarlini at suse dot de changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #1 from paolo at gcc dot gnu dot org 2007-08-28 16:42 ---
Subject: Bug 33209
Author: paolo
Date: Tue Aug 28 16:41:56 2007
New Revision: 127866
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127866
Log:
/cp
2007-08-28 Paolo Carlini <[EMAIL PROTECTED]>
PR c+
--- Comment #13 from dtemirbulatov at gmail dot com 2007-08-28 16:25
---
reviewed patch posted here
http://gcc.gnu.org/ml/gcc-patches/2007-05/msg01280.html
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31490
--- Comment #17 from rguenth at gcc dot gnu dot org 2007-08-28 16:24
---
Created an attachment (id=14128)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14128&action=view)
reduced testcase
Reduced testcase. Note that even removing an unused class (struct counter)
makes
the testca
--- Comment #1 from jakub at gcc dot gnu dot org 2007-08-28 16:17 ---
Isn't this a dup of PR33118 ?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33214
--- Comment #5 from jakub at gcc dot gnu dot org 2007-08-28 16:10 ---
Fixed.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #10 from jakub at gcc dot gnu dot org 2007-08-28 16:09 ---
Fixed.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #7 from jakub at gcc dot gnu dot org 2007-08-28 16:09 ---
Subject: Bug 22244
Author: jakub
Date: Tue Aug 28 16:08:57 2007
New Revision: 127864
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127864
Log:
PR fortran/22244
* Make-lang.in (fortran/trans-ty
--- Comment #9 from jakub at gcc dot gnu dot org 2007-08-28 16:08 ---
Subject: Bug 32400
Author: jakub
Date: Tue Aug 28 16:08:02 2007
New Revision: 127863
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127863
Log:
PR c++/32596
PR c++/32400
* pt.c (check_e
--- Comment #4 from jakub at gcc dot gnu dot org 2007-08-28 16:08 ---
Subject: Bug 32596
Author: jakub
Date: Tue Aug 28 16:08:02 2007
New Revision: 127863
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127863
Log:
PR c++/32596
PR c++/32400
* pt.c (check_e
--- Comment #1 from daney at gcc dot gnu dot org 2007-08-28 15:59 ---
Can you post a fully self contained test case? If I can easily reproduce it, I
will try to fix it.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33218
--- Comment #6 from jakub at gcc dot gnu dot org 2007-08-28 15:58 ---
if (a == 0) a = bar (); isn't necessary either.
salias has:
# BLOCK 2 freq:1
# PRED: ENTRY [100.0%] (fallthru,exec)
# VUSE { qD.2026 SMT.25D.2079 }
D.2027_3 = foo ();
pD.2025_4 = (struct S *) D.2027_3
--- Comment #16 from rguenth at gcc dot gnu dot org 2007-08-28 15:56
---
The differences in alias info are due to different memory partitioning that
happens (regardless of same IL, same number of referenced vars and same BB
frequencies) as we have different "Memory symbol references bef
--- Comment #4 from pcarlini at suse dot de 2007-08-28 15:56 ---
Out of curiosity, do you ever add -Wall to the command line?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33219
--- Comment #3 from andreagrassi at sogeasoft dot com 2007-08-28 15:42
---
Sorry I confirmed for mistake ... Continue the bug report...
The code is very simple and you can read it in the souce code a.c.
Command line of the compile is simple too: "gcc a.c".
--- Comment #2 from andreagrassi at sogeasoft dot com 2007-08-28 15:25
---
Created an attachment (id=14127)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14127&action=view)
preprocessed file
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33219
--- Comment #1 from andreagrassi at sogeasoft dot com 2007-08-28 15:24
---
Created an attachment (id=14126)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14126&action=view)
Source
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33219
--
Summary: Error in compiling when there is a function with a char
parameter called before its declaration with inline
parameters.
Product: gcc
Version: 4.1.2
Status: UNCONFIRMED
Severity: normal
--- Comment #5 from jakub at gcc dot gnu dot org 2007-08-28 15:16 ---
Even more simplified testcase:
extern void abort (void);
struct S
{
void *s1, *s2;
unsigned char s3, s4, s5;
};
__attribute__((noinline))
void *foo (void)
{
static struct S s;
return &s;
}
__attribute__((no
When creating process (Process p = ...) which do not respond to Ctrl+C
then behavior of destroy or waitFor or both is incorrect.
Process blocking/discarding signal sent by Ctrl+C is not killed by destroy().
(The Process.destroy() supposed (IMHO) to kill the child process forcibly.)
After calling
--- Comment #4 from jakub at gcc dot gnu dot org 2007-08-28 14:55 ---
On the trunk I'm just seeing
*** in database main ***
Page 2 is never used
That seems to be because of miscompiled sqlite3SelectNew function with
-O2 -fstrict-aliasing, with -O2 -fno-strict-aliasing that works.
Here i
--- Comment #5 from vda dot linux at googlemail dot com 2007-08-28 14:49
---
Created an attachment (id=14125)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14125&action=view)
An example of code compiled differently by 4.2.1
It still exists in 4.2.1
testcase-421 contains preporce
--
fxcoudert at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNCONFIRMED |NEW
Ever Confirmed|0 |1
Last recon
--- Comment #1 from fxcoudert at gcc dot gnu dot org 2007-08-28 14:45
---
I think the data member should have the type. If it doesn't, it's one of the
many debug-info gfortran bugs :( We could really use a hand in that area.
--
fxcoudert at gcc dot gnu dot org changed:
--- Comment #1 from fxcoudert at gcc dot gnu dot org 2007-08-28 14:41
---
This appeared at some point between 20070821 [rev 127667] and 20070827 [rev
127830]. Most likely culprit at this point is:
2007-08-26 Jerry DeLisle <[EMAIL PROTECTED]>
PR fortran/33055
* trans-
--
fxcoudert at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNCONFIRMED |NEW
Ever Confirmed|0 |1
Last recon
--- Comment #7 from jason at gcc dot gnu dot org 2007-08-28 14:00 ---
Fixed in 4.2 and 4.3.
--
jason at gcc dot gnu dot org changed:
What|Removed |Added
Statu
--- Comment #7 from jason at gcc dot gnu dot org 2007-08-28 13:57 ---
Subject: Bug 28558
Author: jason
Date: Tue Aug 28 13:57:09 2007
New Revision: 127859
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127859
Log:
PR c++/28558
* cp/decl.c (groktypename): Ignore a
--- Comment #15 from rguenth at gcc dot gnu dot org 2007-08-28 13:54
---
The first difference (between -g and non--g build) is after DSE where for the
-g
build it doesn't delete a redundant store. Which is because we have different
alias information with/without -g (in fact this effect
When I compile the module listed below using the version of gfortran for Linux
dated August 27 (downloaded from http://gcc.gnu.org/wiki/GFortranBinaries), I
get the following error:
p.f90: In function 'i':
p.f90:5: internal compiler error: in make_decl_rtl, at varasm.c:1263
Please submit a full bu
--- Comment #5 from chris dot fairles at gmail dot com 2007-08-28 12:32
---
The following code snippet also causes an ICE (g++43 (GCC) 4.3.0 20070828
(experimental) with -Wall, -g and -std=c++0x)
test4.cpp: In instantiation of A:
test4.cpp:11: instantiated from here
test4.cpp:7
--- Comment #15 from rguenth at gcc dot gnu dot org 2007-08-28 12:16
---
-O1 doesn't enable -fstrict-aliasing, only -O2 does.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32758
--- Comment #3 from ubizjak at gmail dot com 2007-08-28 12:02 ---
Current mainline [GCC: (GNU) 4.3.0 20070828] generates:
test:
.LFB2:
xorl%eax, %eax
xorl%edx, %edx
.align 16
.L2:
addbtable(%rdx), %al
addq$1, %rdx
cmpq
--- Comment #14 from aph at gcc dot gnu dot org 2007-08-28 11:44 ---
Is this with -fno-strict-aliasing?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32758
--- Comment #3 from xk-corpse at hotmail dot com 2007-08-28 11:25 ---
(In reply to comment #2)
> This works for me on a x86_64 target.
>
For me it works on i686-pc-linux (with cross-compiler, configured with the same
method as above - crosstool-0.43 + glibc-2.3.2), works on i386-mingw3
--- Comment #14 from rguenth at gcc dot gnu dot org 2007-08-28 11:05
---
I'm reducing the testcase now.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33199
--- Comment #3 from rguenth at gcc dot gnu dot org 2007-08-28 10:58 ---
long foo(float x)
{
return __builtin_lfloorf(x);
}
generates
foo:
.LFB2:
cvttss2siq %xmm0, %rax
cvtsi2ssq %rax, %xmm1
leaq-1(%rax), %rdx
comiss %xmm0, %xmm1
--
pcarlini at suse dot de changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |pcarlini at suse dot de
|dot org |
--- Comment #1 from rguenth at gcc dot gnu dot org 2007-08-28 10:21 ---
The ICE/complain goes away if I remove either the section attribute or the
const
qualification of __CTOR_LIST__.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33216
--- Comment #8 from ubizjak at gmail dot com 2007-08-28 10:08 ---
FIxed by http://gcc.gnu.org/ml/gcc-patches/2007-07/msg00354.html
--
ubizjak at gmail dot com changed:
What|Removed |Added
--- Comment #7 from ubizjak at gmail dot com 2007-08-28 10:04 ---
This is fixed in current mainline, gcc version 4.3.0 20070827 produces:
foo:
pushl %ebx
movl8(%esp), %ebx
xorl%edx, %edx
movl12(%esp), %ecx
.p2align 4,,7
.p2al
1 - 100 of 119 matches
Mail list logo
