--- Comment #1 from burnus at gcc dot gnu dot org 2007-08-22 06:36 ---
Minimal example:
implicit none
real, TARGET :: a(0:100)
real, pointer :: p(:)
p => a
print *, lbound(a), ubound(a)
print *, lbound(p), ubound(p)
end
Prints:
0 100
1
This works:
/usr/local/bin/gcc -O3 -mtune=native -c Weapon.c
This does not:
/usr/local/bin/gcc -march=native Weapon.c -c -O3
Weapon.c:1: error: bad value (native) for -march= switch
Weapon.c:1: error: bad value (native) for -mtune= switch
gcc -v yields:
Using built-in specs.
Target: i386-apple-da
--- Comment #3 from joseph at codesourcery dot com 2007-08-22 01:13 ---
Subject: Re: preprocess should ignore trigraphs in
/* */ comments
On Wed, 22 Aug 2007, pinskia at gcc dot gnu dot org wrote:
> Now in multiple line comments, they should not be warned about.
Unless of course the
I have found an instance of bad code generation. It happens in function
with complex control flow and lots of gotos.
I have isolated the bad code and produced debug output.
The fragment produces this output:
/* First, check if the sequence number of the incoming packet is
what we're expecting
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-08-22 00:37 ---
> Trigraphs in #if 0 ... #endif
No that is correct to warn in that context as text inside #if 0 ... #endif are
supposed to be tokens.
Now in multiple line comments, they should not be warned about.
--
http://gc
--- Comment #1 from mec at google dot com 2007-08-22 00:35 ---
Created an attachment (id=14093)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14093&action=view)
C++ source file with trigraphs in comments and #if 0
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33143
http://gcc.gnu.org/onlinedocs/gcc-4.2.1/gcc/Warning-Options.html#Warning-Options-Wtrigraphs
-Wtrigraphs
Warn if any trigraphs are encountered that might change the meaning of the
program (trigraphs within comments are not warned about). This warning is
enabled by -Wall.
Right now, trigraphs
--- Comment #1 from kargl at gcc dot gnu dot org 2007-08-21 22:31 ---
See PR 20248 before you change -fall-intrinsics behavior.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33141
--- Comment #2 from rguenth at gcc dot gnu dot org 2007-08-21 22:00 ---
Created an attachment (id=14092)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14092&action=view)
untested patch
The problem is we compare for TYPE_MIN_VALUE by pointer, but the conversion
from long to int in
--- Comment #2 from dps at simpson dot demon dot co dot uk 2007-08-21
21:49 ---
Subject: Re: jc1 segault building
lljava/java/util/AbstractMap.java
No CLASSPATH variable exists in the build environment.
The compiler was confiugred with
../gcc/configure --prefix=/usr --enabled-shar
--- Comment #6 from ian at airs dot com 2007-08-21 21:44 ---
Fixed.
--
ian at airs dot com changed:
What|Removed |Added
Status|NEW |RE
--- Comment #1 from rguenth at gcc dot gnu dot org 2007-08-21 21:36 ---
Confirmed.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
CC|
--- Comment #5 from ian at gcc dot gnu dot org 2007-08-21 21:35 ---
Subject: Bug 33134
Author: ian
Date: Tue Aug 21 21:35:06 2007
New Revision: 127679
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127679
Log:
./:
PR tree-optimization/33134
* tree-vrp.c (adjust_r
[ forwarded from http://bugs.debian.org/429657 ]
[EMAIL PROTECTED]:/tmp% cat test.c
int abs(int j);
void abort(void);
__attribute__((noinline)) int lisp_atan2(long dy, long dx) {
if (dx <= 0)
if (dy > 0)
return abs(dx) <= abs(dy);
return 0;
}
int main() {
volat
--- Comment #1 from rakdver at gcc dot gnu dot org 2007-08-21 21:29 ---
This patch fixes the problem:
Index: tree-ssa-loop-niter.c
===
*** tree-ssa-loop-niter.c (revision 127674)
--- tree-ssa-loop-niter.c (worki
--- Comment #8 from patchapp at dberlin dot org 2007-08-21 21:25 ---
Subject: Bug number PR rtl-optimization/32557
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/msg01404.html
--
http://gcc.g
--- Comment #2 from tbm at cyrius dot com 2007-08-21 21:03 ---
typedef long unsigned int size_t;
typedef struct
{
}
ldiv_t;
typedef unsigned long int ulong;
enum
{
t_INT = 1, t_REAL = 2, t_INTMOD = 3, t_FRAC = 4, t_COMPLEX = 6, t_PADIC =
15, t_QFI = 16, t_VEC = 17, t_COL = 18, t_MA
1. There should be a -W* option which gives a warning if a user procedure has
the same name as an intrinsic procedure (in the enabled -std=); e.g.
Warning: 'GAMMA' declared at (1) is also the name of an intrinsic. It can only
be called via an explicit interface or if declared EXTERNAL.
(http://gc
--- Comment #1 from tbm at cyrius dot com 2007-08-21 21:03 ---
Maybe caused by this change?
2008-08-05 Andrew Pinski <[EMAIL PROTECTED]>
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33140
I'm getting the following ICE with current trunk. This appeared between
20070720 and 20070811.
(sid)25830:[EMAIL PROTECTED]: ~] /usr/lib/gcc-snapshot/bin/gcc -c -O2
pari-base1.c
pari-base1.c: In function 'initzeta':
pari-base1.c:42: internal compiler error: in build2_stat, at tree.c:3115
Please
that pointer are
coming out incorrect. Below is a sample code showing the problem. It can be
built without any compiler options. I get this with the most recent version of
gfortran on linux (FC5).
$ gfortran --version
GNU Fortran (GCC) 4.3.0 20070821 (experimental) [trunk revision 127658]
Here is the
--- Comment #15 from gdr at cs dot tamu dot edu 2007-08-21 20:27 ---
Subject: Re: Awful error messages with virtual functions
"reichelt at gcc dot gnu dot org" <[EMAIL PROTECTED]> writes:
| The pointer_plus_exprt stuff has been fixed.
|
| We are now back to error messages like
|
| b
--- Comment #3 from reichelt at gcc dot gnu dot org 2007-08-21 20:06
---
The ICE vanished on mainline between 2007-08-05 and 2007-08-15.
I'd guess this is due to the fix for PR 27211.
--
reichelt at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #14 from reichelt at gcc dot gnu dot org 2007-08-21 19:56
---
The pointer_plus_exprt stuff has been fixed.
We are now back to error messages like
bug.cc: In member function 'void B::bar()':
bug.cc:4: error: could not convert '(((A*)this) + ((unsigned int)(*(int*)(
((B*)thi
--- Comment #13 from reichelt at gcc dot gnu dot org 2007-08-21 19:50
---
Here's an even smaller testcase:
=
int *p;
inline int foo(int *q)
{
if (*q)
*q = 0;
return *q;
}
void bar(int *r, int i)
{
int *s;
while (--i)
*(s = r) = foo(p);
++
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |rguenth at gcc dot gnu dot
|dot org
--- Comment #8 from patchapp at dberlin dot org 2007-08-21 19:06 ---
Subject: Bug number PR32980
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/msg01386.html
--
http://gcc.gnu.org/bugzilla/sh
--- Comment #7 from burnus at gcc dot gnu dot org 2007-08-21 19:04 ---
> * Will we want a -std=f2008 separate from f2003, or do we consider f2003 an
> alias for f2008 (F2008 being, after all, a minor improvement to F2003)?
Who said that it will be a minor change compared to Fortran 2003
--
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 #6 from fxcoudert at gcc dot gnu dot org 2007-08-21 17:52
---
(In reply to comment #5)
> * scaled-erfc will require a library implementation, because it's not widely
> available on systems out there
I should have added this link: http://www.netlib.org/specfun/erf
--
ht
--- Comment #5 from fxcoudert at gcc dot gnu dot org 2007-08-21 17:42
---
Few comments:
* F2008 draft says: "The intrinsic function ATAN is extended so that ATAN (Y,
X) is ATAN2 (Y,X)." We'll need to take care of that too.
* Will we want a -std=f2008 separate from f2003, or do we cons
--- Comment #4 from tbm at cyrius dot com 2007-08-21 17:25 ---
better (C based) testcase:
/* Testcase by Martin Michlmayr <[EMAIL PROTECTED]> */
ui_create_cursor (unsigned char *andmask)
{
unsigned char *cursor, *pcursor;
unsigned char nextbit;
for (nextbit = 0x80; nextbit != 0;
--- Comment #3 from jakub at gcc dot gnu dot org 2007-08-21 17:11 ---
Subject: Bug 32610
Author: jakub
Date: Tue Aug 21 17:11:20 2007
New Revision: 127669
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127669
Log:
PR debug/32610
* dwarf2out.c (gen_decl_die): Don'
--- Comment #4 from burnus at gcc dot gnu dot org 2007-08-21 17:08 ---
Created an attachment (id=14091)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14091&action=view)
invoke.texi patch
Note: I did not mention "Fortran 2008" and I also did not include "LOG_GAMMA"
(F2008 name of "
--- Comment #3 from burnus at gcc dot gnu dot org 2007-08-21 17:06 ---
Created an attachment (id=14090)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14090&action=view)
invoke.texi patch
Note: I did not mention "Fortran 2008" and I also did not include "LOG_GAMMA"
(F2008 name of "
On 8/21/07, Michael Williamson <[EMAIL PROTECTED]> wrote:
> I would expect a message like:
>
> "the second parameter is not a pointer".
Because '\0' can converted to a pointer type as it is a 0 (for a NULL pointer).
-- Pinski
I mistakenly used
memcpy (s, '\0', 5);
instead of
memset (s, '\0', 5);
but the gcc compiler gave no warnings (using -Wall).
I would expect a message like:
"the second parameter is not a pointer".
-Mike
_
--- Comment #3 from hemant dot jaiswal at gmail dot com 2007-08-21 16:24
---
Created an attachment (id=14089)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14089&action=view)
pre-processed source file
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33123
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-08-21 16:21 ---
Do you have the preprocessed source?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33123
--- Comment #6 from pinskia at gcc dot gnu dot org 2007-08-21 16:20 ---
Actually I think this might be related to removing iterative jump threading and
not aliasing.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
---
--- Comment #3 from burnus at gcc dot gnu dot org 2007-08-21 15:46 ---
> I wasn't even aware of their existence!
Me neither. I found it in the NAG f95 release notes.
> I'll do it (unless you want to?), thanks for the doc patch.
Go ahead.
--
http://gcc.gnu.org/bugzilla/show_bug.cg
--- Comment #2 from tbm at cyrius dot com 2007-08-21 15:33 ---
BTW, the segfault goes away when I remove the last 3 lines (function tiger_t).
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33138
--- Comment #1 from tbm at cyrius dot com 2007-08-21 15:32 ---
typedef unsigned long int word64;
extern word64 tiger_table[4 * 256];
tiger_compress (word64 * str, word64 state[3])
{
{
register word64 a, b, c;
word64 aa, bb, cc;
word64 x0, x1, x2, x3, x4, x5, x6, x7;
b =
There are two issues here:
- the inline assembly is rejected at -O. I don't know if the assembler
code is valid, but fwiw it's accepted by gcc 4.1 and 4.2.
- gcc segfaults.
I see this with current trunk and with trunk from 20070720.
(sid)27576:[EMAIL PROTECTED]: ~] /usr/lib/gcc-snapshot/bi
--- Comment #8 from jakub at gcc dot gnu dot org 2007-08-21 15:29 ---
The patch for PR32596 fixes this as well.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
---
--- Comment #2 from tbm at cyrius dot com 2007-08-21 14:55 ---
/* Testcase by Martin Michlmayr <[EMAIL PROTECTED]> */
struct fcs_struct_state_t
{
char freecells[4];
};
typedef struct fcs_struct_state_t fcs_state_t;
struct fcs_struct_state_with_locations_t
{
fcs_state_t s;
};
typedef
--- Comment #1 from tbm at cyrius dot com 2007-08-21 14:55 ---
I wonder if this got introduced by one of Sebastian Pop's patches.
--
tbm at cyrius dot com changed:
What|Removed |Added
bm/tmp/gcc/gcc-4.3-20070821-r127666/gcc/cc1 -O3
~/freecell-solver-state.c
fcs_card_compare freecell_solver_canonize_state
Analyzing compilation unit
Performing interprocedural optimizations
Assembling fu
nctions:
fcs_card_compare freecell_solver_canonize_state
Program received signal SIGSEGV,
--- Comment #7 from tbm at cyrius dot com 2007-08-21 14:39 ---
Can an ia64-maintainer approve one of the patches Jakub has suggested?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32338
--- Comment #6 from jakub at gcc dot gnu dot org 2007-08-21 14:18 ---
Testing a patch.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|una
--- Comment #4 from jakub at gcc dot gnu dot org 2007-08-21 14:13 ---
Testing a patch.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|una
--- Comment #14 from fxcoudert at gcc dot gnu dot org 2007-08-21 14:02
---
I've asked the mingw-users mailing-list about this issue:
http://sourceforge.net/mailarchive/message.php?msg_name=19c433eb0708210654h3b419607t5f2985edfc809348%40mail.gmail.com
(PS: the impact of this bug will be
--- Comment #3 from ian at airs dot com 2007-08-21 13:59 ---
Index: gcc/tree-vrp.c
===
--- gcc/tree-vrp.c (revision 127491)
Testing this patch.
+++ gcc/tree-vrp.c (working copy)
@@ -2641,7 +2641,7 @@ adjust_range_
--- Comment #12 from pixel at mandriva dot com 2007-08-21 13:29 ---
i do know it works nicely with gcc 4.3
but i still get the "memory hog" behaviour using branches/gcc-4_2-branch, ie:
% /usr/lib/gcc/i586-mandriva-linux-gnu/4.2.1/cc1 -O2 fail.c
runs with memory RSS raising up to 1G ma
--- Comment #5 from rguenth at gcc dot gnu dot org 2007-08-21 13:12 ---
trunk regresses even more
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #4 from rguenth at gcc dot gnu dot org 2007-08-21 13:10 ---
4.1 somehow has better alias information, you can verify this by adding
-fno-strict-aliasing to 4.1 and get the same (worse) result as with 4.2.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33136
--- Comment #3 from vda dot linux at googlemail dot com 2007-08-21 12:26
---
Created an attachment (id=14088)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14088&action=view)
assembler output generated by 4.2.1
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33136
--- Comment #2 from vda dot linux at googlemail dot com 2007-08-21 12:26
---
Created an attachment (id=14087)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14087&action=view)
assembler output generated by 4.1.1
Sorry, don't have native gcc 4.1.1 at this machine, but have
i486-lin
--- Comment #1 from vda dot linux at googlemail dot com 2007-08-21 12:23
---
Created an attachment (id=14086)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14086&action=view)
testcase
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33136
When you compile attached testcase with 4.1.1 and 4.2.1, resulting .o modules
have following sizes:
$ size insmod.i-411.o insmod.i-421.o
textdata bss dec hex filename
288 0 0 288 120 insmod.i-411.o
543 0 0 543 21f insmod.i-421.o
T
--- Comment #2 from rguenth at gcc dot gnu dot org 2007-08-21 12:09 ---
Confirmed. Value numbering constant propagates
c1 = 128;
to
c1.1_6 = (signed char)c1;
and creates a constant with overflow flag set which confuses VRP signed
overflow
handling.
--
rguenth at gcc dot gnu do
--- Comment #8 from rguenth at gcc dot gnu dot org 2007-08-21 11:47 ---
Not true either. host_integerp returns false for sizetype -1.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32563
The documentation of -ffinite-math-only states
"This option should never be turned on by any '-O' option"
but for sh's -mieee it is said "the default is set to -ffinite-math-only"
(-mieee is unset by default)
as a result the -ffinite-math-only is turned on by default for the sh for all
optimisati
--
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 #25 from cesarb at cesarb dot net 2007-08-21 10:54 ---
The testcase seems to be missing one case where it should warn:
class H
{
protected:
~H();
public:
virtual void deleteme() = 0;
};
H::~H()
{
}
void
H::deleteme()
{
delete this;
}
class I : publ
--- Comment #1 from tbm at cyrius dot com 2007-08-21 10:23 ---
Testcase:
/* Testcase by Martin Michlmayr <[EMAIL PROTECTED]> */
class FXObject;
class FXStream
{
public:FXStream (const FXObject *cont = __null);
FXStream & operator<< (const unsigned char &v);
};
bool fxsaveGIF (FXSt
I'm seeing the following ICE with current trunk. This goes back at least
to 20070720.
(sid)25719:[EMAIL PROTECTED]: ~] /usr/lib/gcc-snapshot/bin/g++ -c -O2
fox1.6-fxgifio.cc
fox1.6-fxgifio.cc: In function 'bool fxsaveGIF(FXStream&)':
fox1.6-fxgifio.cc:10: internal compiler error: in set_value_ra
--
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 #7 from rguenth at gcc dot gnu dot org 2007-08-21 10:11 ---
Well, not entirely true. The problem is that sizetype is sign extended, and
we convert from unsigned long to unsigned long sizetype, so the representation
changes which causes fit_double_type to say we overflowed.
--- Comment #3 from jakub at gcc dot gnu dot org 2007-08-21 09:22 ---
Subject: Bug 32912
Author: jakub
Date: Tue Aug 21 09:22:14 2007
New Revision: 127661
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127661
Log:
PR middle-end/32912
* fold-const.c (fold_unary):
--- Comment #1 from manu at gcc dot gnu dot org 2007-08-21 09:00 ---
Hi,
could you attach a preprocessed testcase (*.i) generated by adding -save-temps
?
The minimal the testcase, the better.
Thanks.
--
manu at gcc dot gnu dot org changed:
What|Removed
--- Comment #2 from tbm at cyrius dot com 2007-08-21 08:31 ---
Probably caused by:
2007-08-14 Maxim Kuvyrkov <[EMAIL PROTECTED]>
* sched-int.h (struct _dep): Rename field 'kind' to 'type'.
(DEP_KIND): Rename to DEP_TYPE. Update all uses.
...
--
tbm at cyrius dot
--- Comment #1 from tbm at cyrius dot com 2007-08-21 08:29 ---
Testcase:
static char newshuffle[256 + 16] = {
0x0f, 0x08, 0x05, 0x07, 0x0c, 0x02, 0x0e, 0x09, 0x00, 0x01, 0x06, 0x0d,
0x03, 0x04, 0x0b, 0x0a, 0x02, 0x0c, 0x0e, 0x06, 0x0f, 0x00, 0x01, 0x08,
0x0d, 0x03, 0x0a, 0x04,
I'm getting the following ICE at -O3 with current SVN:
(sid)[EMAIL PROTECTED]:~$ /usr/lib/gcc-snapshot/bin/gcc -c -O3 ncpfs-ncplib.c
ncpfs-ncplib.c: In function 'newpassencrypt':
ncpfs-ncplib.c:46: warning: incompatible implicit declaration of built-in
function 'memcpy'
ncpfs-ncplib.c:48: internal
--- Comment #6 from rguenth at gcc dot gnu dot org 2007-08-21 08:24 ---
Subject: Bug 33122
Author: rguenth
Date: Tue Aug 21 08:23:50 2007
New Revision: 127659
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127659
Log:
2007-08-21 Richard Guenther <[EMAIL PROTECTED]>
PR
--- Comment #5 from rguenth at gcc dot gnu dot org 2007-08-21 08:24 ---
Fixed.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
This happens building libssp at revision 127638:
libtool: compile: /home/rask/build/gcc-m32r-unknown-elf/./gcc/xgcc
-B/home/rask/build/gcc-m32r-unknown-elf/./gcc/ -nostdinc
-B/home/rask/build/gcc-m32r-unknown-elf/m32r-unknown-elf/newlib/ -isystem
/home/rask/build/gcc-m32r-unknown-elf/m32r-unknown
77 matches
Mail list logo
