Compiling
const int n = -2147483648;
results in the warning
minus-max.cc:1: warning: this decimal constant is unsigned only in ISO C90
which is wrong: The constant is exactly -2^31, which fits into a (32 bit) int,
thus no automatic conversions to larger types should occur.
--
--
What|Removed |Added
CC||niemayer at isg dot de
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18460
--- Additional Comments From bernhard dot walle at gmx dot de 2004-11-13
10:47 ---
What does this mean? You cannot reproduce this?
I configured gcc 3.4.3 with
./configure \
--prefix=%{prefix} \
--target=m68k-elf
--- Additional Comments From paulthomas2 at wanadoo dot fr 2004-11-13
11:02 ---
Subject: Re: new/delete much slower than malloc/free because of sjlj exceptions
Here's a test case for you...
-Ken
That's interesting
Using your test case:
(i) gcc 3.2 20020927 ( prerelease)
--
Bug 16994 depends on bug 12807, which changed state.
Bug 12807 Summary: Failure to diagnose VLA in template
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12807
What|Old Value |New Value
--- Additional Comments From reichelt at gcc dot gnu dot org 2004-11-13
13:08 ---
Fixed by Mark's patch for PR18429.
In fact PR12807 is more or less a duplicate of PR18429.
*** This bug has been marked as a duplicate of 18429 ***
--
What|Removed
--- Additional Comments From reichelt at gcc dot gnu dot org 2004-11-13
13:08 ---
*** Bug 12807 has been marked as a duplicate of this bug. ***
--
What|Removed |Added
--- Additional Comments From reichelt at gcc dot gnu dot org 2004-11-13
13:23 ---
The testcase from comment #12 ICEs again:
PR17635.c: In function 'foo':
PR17635.c:2: internal compiler error: in clean_predicate_lists, at
tree-if-conv.c:661
Devang, your patch
--- Additional Comments From cvs-commit at gcc dot gnu dot org 2004-11-13
13:25 ---
Subject: Bug 18230
CVSROOT:/cvs/gcc
Module name:gcc
Changes by: [EMAIL PROTECTED] 2004-11-13 13:25:10
Modified files:
gcc: ChangeLog
gcc/testsuite :
--- Additional Comments From schwab at suse dot de 2004-11-13 13:41 ---
There are no negative constants/literals in C and C++. 2147483648 is not
representable in (32-bit) int or long, so its type is unsigned in C90, but in
C++98 the behaviour is undefined.
--
What
--- Additional Comments From cvs-commit at gcc dot gnu dot org 2004-11-13
13:43 ---
Subject: Bug 18230
CVSROOT:/cvs/gcc
Module name:gcc
Changes by: [EMAIL PROTECTED] 2004-11-13 13:42:45
Modified files:
gcc/testsuite : ChangeLog
--- Additional Comments From reichelt at gcc dot gnu dot org 2004-11-13
13:54 ---
Devang, your patch
http://gcc.gnu.org/ml/gcc-cvs/2004-11/msg00591.html
is responsible for the new ICE.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635
--- Additional Comments From giovannibajo at libero dot it 2004-11-13
14:31 ---
A regression wrt the old frontend.
--
What|Removed |Added
Summary|Alignment
--- Additional Comments From giovannibajo at libero dot it 2004-11-13
14:36 ---
It would be good to get updated timings for this.
--
What|Removed |Added
CC|
--- Additional Comments From ebotcazou at gcc dot gnu dot org 2004-11-13
15:37 ---
Present on all versions of Solaris up to (and including) version 9.
--
What|Removed |Added
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
15:57 ---
No just bugzilla has no way to go from waiting to unconfirmed, I just have not
have time to try to
reproduce and reduce it yet. Maybe someone else will.
--
after a successful gmake bootstrap on IRIX 5.3 a consequent gmake check ends up
with:
gmake[2]: Entering directory
`/usr/people/schwarz/gccb/mips-sgi-irix5.3/libf2c/libU77'
../../../gcc/g77 -B../../../gcc/ -L.. -g
../../../../gcc-3.4.3/libf2c/libU77/u77-test.f \
-lg2c -lm -o a.out ./a.out
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
16:04 ---
*** Bug 18461 has been marked as a duplicate of this bug. ***
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15746
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
16:04 ---
This was already as will not fix (in fact it was your bug which was filed).
This is a will not fix for all of
3.4.x releases.
*** This bug has been marked as a duplicate of 15746 ***
--
--
What|Removed |Added
CC||berndtrog at yahoo dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17994
--
What|Removed |Added
CC||berndtrog at yahoo dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17993
The compiler segfaults when compiling a class declaration that has a very large
array member of class type. This works fine with an array of int and also works
with gcc 3.3.5 and earlier, but fails with all gcc 3.4 releases.
--
Summary: [3.4 Regression] Segfault on declaration of
--- Additional Comments From peturr02 at ru dot is 2004-11-13 16:26 ---
Created an attachment (id=7539)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=7539action=view)
Preprocessed source code
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18462
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
16:45 ---
Either this is a dup of PR 16681 or there is another bug. The splay_tree is
very long which causes the
stack to be overflowed. Maybe we should fix the recursive nature of
splay_tree_delete.
--
The item Moving floating point through integer registers
on http://gcc.gnu.org/projects/optimize.html shows how GCC
can move a float array element via an integer register.
Consider the following test case:
void
fcpy(float *restrict a, float *restrict b,
float *restrict aa, float
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
17:22 ---
Confirmed, the problem is that DOM does:
D.1192 = (unsigned int) i;
D.1194 = (float * restrict) D.1192 * 4B;
*(aa2 + D.1194) = *(a2 + D.1194);
*(bb2 + D.1194) = *(b2 + D.1194);
Note how we use
--- Additional Comments From pierre42d at 9online dot fr 2004-11-13 17:40
---
Ok it was a problem with my glibc
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18132
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
17:42 ---
Though I should note that PPC is much better on the mainline than before:
gcc 4.0.0:
L4:
lfsx f0,r3,r2
stfsx f0,r5,r2
lfsx f13,r4,r2
stfsx f13,r6,r2
addi r2,r2,4
--- Additional Comments From lpadovan at cs dot unibo dot it 2004-11-13
17:52 ---
I'm experiencing a problem with
g++-3.4 (GCC) 3.4.2 (Debian 3.4.2-3)
which seems related to this bug, at least because the error message is similar
and the operator ?: is involved. Here is the program
--- Additional Comments From steven at gcc dot gnu dot org 2004-11-13
17:52 ---
At least x86 and ARM have {reg + reg OP const} addressing
modes. Unfortunately we rip such expressions apart already
in the gimplifier. This is something we canot fix properly
on trees. TER could
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
17:54 ---
For PPC at least IV-OPTS should note that we have post increment and decrement
the pointers before
the loop and then increment all of them inside the loop, aka:
void
fcpy(float *restrict a, float
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
18:14 ---
Here is the reduced testcase for the problem, it has nothing to do with loops
at all:
void
fcpy(float *restrict a, float *restrict b,
float *restrict aa, float *restrict bb, unsigned n)
{
The following code is rejected since gcc 3.4.0:
struct A
{
A(int);
operator void*() const;
};
templateint void foo(const A x) { 0 ? x : (x ? x : 0); }
bug.cc: In
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
18:46 ---
This is a RTL problem as it works correctly on ARM which has it ...
I should note that arm's instruction has nothing special in its .md file:
(define_insn *arm_movsf_soft_insn
[(set (match_operand:SF 0
--- Additional Comments From pierre42d at 9online dot fr 2004-11-13 18:46
---
Ok it was a problem with my glibc
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18356
--- Additional Comments From reichelt at gcc dot gnu dot org 2004-11-13
18:47 ---
The problem in comment #12 is indeed very similar, but not the same,
since it's not fixed ;-)
I opened PR 18464 to track the new problem.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9440
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
18:51 ---
: Search converges between 2003-08-19-trunk (#328) and 2003-08-20-trunk (#329).
Confirmed.
--
What|Removed |Added
With this code:
static void f() __asm__(ff) __attribute__((used));
static void __attribute__((used)) f() { }
gcc 3.3 doesn't emit any code for ff when optimizing (-O3). gcc 3.4 and 4 do.
This breaks mozilla on alpha-linux.
--
Summary: static function with asm(name) and
The C++ frontend accepts the following (IMHO invalid) variable declaration
since gcc 3.0:
=
int ::i;
=
--
Summary: [3.3/3.4/4.0 regression] int ::i; accepted
Product: gcc
Version: 4.0.0
Status:
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
19:02 ---
Confirmed been failing since at least 2000-12-31.
--
What|Removed |Added
--- Additional Comments From reichelt at gcc dot gnu dot org 2004-11-13
19:03 ---
Nathan, your patch
http://gcc.gnu.org/ml/gcc-cvs/2003-08/msg00634.html
introduced the regression.
Could you please have a look?
--
What|Removed |Added
--
What|Removed |Added
Keywords||wrong-code
Summary|static function with|[3.3 only] static function
--- Additional Comments From ovidr at users dot sourceforge dot net
2004-11-13 19:18 ---
The attached patch works - install succeeds.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18458
(...)
/home/users/pluto/rpm/BUILD/gcc-3.4.3/obj-arm-pld-linux/gcc/xgcc
-B/home/users/pluto/rpm/BUILD/gcc-3.4.3/obj-arm-pld-linux/gcc/
-B/usr/arm-pld-linux/bin/ -B/usr/arm-pld-linux/lib/
-isystem /usr/arm-pld-linux/include -isystem /usr/arm-pld-linux/sys-include
-O2 -DIN_GCC -DCROSS_COMPILE
--- Additional Comments From pluto at pld-linux dot org 2004-11-13 19:19
---
Created an attachment (id=7540)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=7540action=view)
buildlog
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18467
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
19:21 ---
For linux you have to use glibc as that is only one supported for unwinding.
--
What|Removed |Added
--- Additional Comments From ovidr at users dot sourceforge dot net
2004-11-13 19:22 ---
I don't think this is a duplicate - it has nothing to do with LANG settings.
The stacktrace may look similar, and end on _Jv_FindClass, but it does not go
through PrintStream/UnicodeToBytes.
I
--- Additional Comments From reichelt at gcc dot gnu dot org 2004-11-13
19:23 ---
Mark, your patch
http://gcc.gnu.org/ml/gcc-cvs/2003-07/msg00378.html
seems to be responsible for the regression.
Could you please have a look?
--
What|Removed |Added
--- Additional Comments From steven at gcc dot gnu dot org 2004-11-13
19:34 ---
CSE is trying to reconstruct the addressing mode, but it has
(plus:SI (ashift:SI (reg/v:SI 61 [ n ]) (const_int 2 [0x2]))
(reg/v/f:SI 59 [ a ]))
According to hp, the canonical form would be
--- Additional Comments From cvs-commit at gcc dot gnu dot org 2004-11-13
19:53 ---
Subject: Bug 18347
CVSROOT:/cvs/gcc
Module name:gcc
Changes by: [EMAIL PROTECTED] 2004-11-13 19:52:55
Modified files:
gcc: ChangeLog
gcc/config/mmix:
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
19:56 ---
Hmm, then JCF sections are not support.
--
What|Removed |Added
--- Additional Comments From hp at gcc dot gnu dot org 2004-11-13 20:00
---
Keeping this open until I've checked the 3.4 branch.
--
What|Removed |Added
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
20:14 ---
Actually I was wrong about IV-OPTS being the problem. This is the same problem
as PR 18463, the
problem is that CSE does not recombine to form the addressing mode at all.
--
What|Removed
--
Bug 18359 depends on bug 18326, which changed state.
Bug 18326 Summary: [4.0 regression] mmix-knuth-mmixware testsuite failure:
gcc.c-torture/execute/20040625-1.c
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18326
What|Old Value |New Value
--- Additional Comments From hp at gcc dot gnu dot org 2004-11-13 20:17
---
See comment in URL:http://gcc.gnu.org/ml/gcc-patches/2004-11/msg0.html
regarding the TImode libcall. I'm closing this anyway; it's not a testsuite
regression anymore.
--
What|Removed
--- Additional Comments From dannysmith at users dot sourceforge dot net
2004-11-13 20:18 ---
This excerpt from java/class.c appears relavant:
void
emit_register_classes (tree *list_p)
{
if (registered_class == NULL)
return;
/* ??? This isn't quite the correct test. We also
A template function (which swaps two values)
is called in a loop in an inlined method of a template class. The number
of loop iterations is a constant, so the compiler might be trying to unroll
the loop.
The final assignment of the swap function is not executed after the
first iteration, causing
--- Additional Comments From james_avera at yahoo dot com 2004-11-13 20:32
---
Created an attachment (id=7542)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=7542action=view)
test case
Please compile with -O2
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18468
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
20:37 ---
You are violating the C++ aliasing rules:
class GetPtrtokeyFunctor {
public:
const PtrInt * operator()(const Objptr objptr)
{ return reinterpret_castconst PtrInt *(objptr); }
};
--
--- Additional Comments From hp at gcc dot gnu dot org 2004-11-13 20:53
---
The part of this bug that was execution failure was another incarnation of the
TImode-non-support bug mentioned in PR target/18347. There, it caused a SEGV,
while here, it picked up the inner QImode in the
--
What|Removed |Added
Target Milestone|--- |4.0.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18332
The following valid code snippet is rejected by mainline:
templatetypename struct A
{
static const int i=1;
};
templatetypename T struct B : AT
{
using AT::i;
char s[i]; // fails
char t[AT::i]; // compiles
};
Configured with: ../../src/cvs/head/gcc/configure --
prefix=/aaronwl/cs/env/solaris-head --target=sparc-sun-solaris2.8 --disable-
nls --enable-languages=c --with-sysroot=/aaronwl/cs/env/solaris-head/root
Thread model: single
gcc version 4.0.0 20041113 (experimental)
--
Summary: configure
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
21:03 ---
This worked with 2004-11-12-161002 so it has to be recent.
--
What|Removed |Added
Target
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
21:06 ---
auto-host.h should not be included in crtstuff.c at all.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18469
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-13
21:08 ---
Confirmed.
--
What|Removed |Added
Status|UNCONFIRMED |NEW
--- Additional Comments From cvs-commit at gcc dot gnu dot org 2004-11-13
23:09 ---
Subject: Bug 18300
CVSROOT:/cvs/gcc
Module name:gcc
Changes by: [EMAIL PROTECTED] 2004-11-13 23:09:08
Modified files:
gcc: ChangeLog
gcc/config/i386:
--- Additional Comments From giovannibajo at libero dot it 2004-11-13
23:31 ---
Fixed.
--
What|Removed |Added
Status|NEW |RESOLVED
The following invalid code causes an ICE on mainline:
=
typedef int X;
templateX struct X {};
=
bug.cc:2: error: 'templateint anonymous struct X' redeclared as different
kind of symbol
bug.cc:1: error: previous declaration of 'typedef int
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-14
01:45 ---
: Search converges between 2004-02-01-trunk (#445) and 2004-03-01-trunk (#446).
--
What|Removed |Added
--- Additional Comments From pinskia at gcc dot gnu dot org 2004-11-14
01:56 ---
Patch here: http://gcc.gnu.org/ml/gcc-patches/2004-11/msg01133.html.
--
What|Removed |Added
--- Additional Comments From cvs-commit at gcc dot gnu dot org 2004-11-14
04:53 ---
Subject: Bug 18347
CVSROOT:/cvs/gcc
Module name:gcc
Branch: gcc-3_4-branch
Changes by: [EMAIL PROTECTED] 2004-11-14 04:53:23
Modified files:
gcc:
--
Bug 18332 depends on bug 18347, which changed state.
Bug 18347 Summary: [3.4/4.0 regression] mmix-knuth-mmixware testsuite failure:
gcc.dg/uninit-C.c
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18347
What|Old Value |New Value
--- Additional Comments From hp at gcc dot gnu dot org 2004-11-14 05:15
---
See URL:http://gcc.gnu.org/ml/gcc-patches/2004-11/msg0.html
and URL:http://gcc.gnu.org/ml/gcc-patches/2004-11/msg01141.html.
--
What|Removed |Added
--
What|Removed |Added
Target Milestone|4.0.0 |3.4.4
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18332
--
What|Removed |Added
Target Milestone|3.4.4 |4.0.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18332
tree-if-conv.c:tree_if_convert_stmt uses TREE_OPERAND (t, 1)
for t being GOTO_EXPR like so:
case GOTO_EXPR:
/* Unconditional goto */
add_to_predicate_list (bb_for_stmt (TREE_OPERAND (t, 1)), cond);
bsi_remove (bsi);
cond = NULL_TREE;
break;
GOTO_EXPR has only
--
What|Removed |Added
CC||dpatel at apple dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18472
78 matches
Mail list logo
