--- Comment #2 from echristo at apple dot com 2005-11-30 06:29 ---
Patch committed.
--
echristo at apple dot com changed:
What|Removed |Added
Status|ASSIGNED
gcc.target/powerpc/pr18096-1.c has not been updated for the changing of the
stack overflowing from a warning to an error.
I don't have access to ppc-darwin to test a fix which is why I am filing a bug.
--
Summary: gcc.target/powerpc/pr18096-1.c fails on PPC
Product: gcc
--- Comment #10 from jw203198 at hotmail dot com 2005-11-30 04:16 ---
>
> If you use -E -H -g -fno-working-directory, you will not see the directory
> name.
Thanks, I can use that.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25175
--- Comment #6 from joseph at codesourcery dot com 2005-11-30 04:14 ---
Subject: Re: [4.2 Regression] FAIL: gcc.c-torture/execute/conversion.c
compilation
On Wed, 30 Nov 2005, dave at hiauly1 dot hia dot nrc dot ca wrote:
> I'm testing a fix for the current reload problems on the PA.
--- Comment #9 from ian at airs dot com 2005-11-30 04:06 ---
This is documented behaviour.
The -g option enables the -fworking-directory option, as described in the
documentation of -fworking-directory. It is the -fworking-directory option
which is printing the directory name.
If you
--- Comment #8 from jw203198 at hotmail dot com 2005-11-30 03:52 ---
(In reply to comment #7)
> (In reply to comment #6)
> > > Your orginal example does not show a difference for -H at all.
> Wait a minute, -E outputs the preprocessed source. -H outputs the files which
> are included.
--- Comment #5 from dave at hiauly1 dot hia dot nrc dot ca 2005-11-30
03:51 ---
Subject: Re: [4.2 Regression] FAIL: gcc.c-torture/execute/conversion.c
compilation
> As explained in bug 24998, I can't test on PA at present but the fix is
> probably similar to that for IA64.
I'm testi
--- Comment #7 from pinskia at gcc dot gnu dot org 2005-11-30 03:42 ---
(In reply to comment #6)
> > Your orginal example does not show a difference for -H at all.
Wait a minute, -E outputs the preprocessed source. -H outputs the files which
are included.
Again this is not a bug.
-
--- Comment #2 from dave at hiauly1 dot hia dot nrc dot ca 2005-11-30
03:38 ---
Subject: Re: FAIL: gcc.dg/weak/weak-14.c
> Isn't this just bug 24478?
Yes. However, 24478 didn't have any host/target/build and I didn't
see it.
Dave
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=
--- Comment #6 from jw203198 at hotmail dot com 2005-11-30 03:37 ---
> Your orginal example does not show a difference for -H at all.
Does for me:
pc:~/work/stats $ cc -E -H x.c
# 1 "x.c"
# 1 ""
# 1 ""
# 1 "x.c"
main(){}
and:
pc:~/work/stats $ cc -E -H -g
--- Comment #2 from dave at hiauly1 dot hia dot nrc dot ca 2005-11-30
03:32 ---
Subject: Re: FAIL: gfortran.dg/module_equivalence_1.f90
> This is obvious an issue, this is what I get on x86_64-linux-gnu:
> .comm test_equiv.eq.1_,16,16
> .comm test_equiv.eq.1_,16,16
--
ghazi at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|--- |4.0.3
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25169
Kornel wrote:
g++ crashes for a very long function name ;)
Bugs should be filed into bugzilla, rather than mailed to the gcc-bugs
list. We won't track bugs mailed to the list.
I took a quick look. I couldn't reproduce on my system. However, it
occured to me that gcc is probably trying to
--
ghazi at gcc dot gnu dot org changed:
What|Removed |Added
URL||http://gcc.gnu.org/ml/gcc-
|
--- Comment #5 from pinskia at gcc dot gnu dot org 2005-11-30 03:08 ---
(In reply to comment #4)
> Don't you think that forcing people to cast pointers is much much more
> dangerous than allowing simple equivalence of pointers when they aren't even
> being derefenced?
No. Read my comme
--- Comment #5 from pinskia at gcc dot gnu dot org 2005-11-30 03:06 ---
(In reply to comment #4)
> > Can you show where -H will give you the CWD?
> Use my original example and add -H.
Your orginal example does not show a difference for -H at all.
pc64:~> gcc -H t.c -g
pc64:~> gcc -H t.c
--- Comment #4 from jw203198 at hotmail dot com 2005-11-30 02:55 ---
> Can you show where -H will give you the CWD?
Use my original example and add -H.
I *never* had "-save-temps" in my example so I don't understand why you are
corrupting my example so. And I *never* used a -S flag ei
--- Comment #7 from jvdelisle at gcc dot gnu dot org 2005-11-30 02:53
---
Subject: Bug 25109
Author: jvdelisle
Date: Wed Nov 30 02:53:18 2005
New Revision: 107699
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=107699
Log:
2005-11-29 Jerry DeLisle <[EMAIL PROTECTED]>
--- Comment #4 from jw203198 at hotmail dot com 2005-11-30 02:52 ---
Don't you think that forcing people to cast pointers is much much more
dangerous than allowing simple equivalence of pointers when they aren't even
being derefenced?
What about this:
struct x *f() { return ptr;}
--- Comment #1 from pinskia at gcc dot gnu dot org 2005-11-30 02:32 ---
This is obvious an issue, this is what I get on x86_64-linux-gnu:
.comm test_equiv.eq.1_,16,16
.comm test_equiv.eq.1_,16,16
.comm test_equiv.eq.1_,16,16
Though one could say that this i
--- Comment #1 from pinskia at gcc dot gnu dot org 2005-11-30 02:29 ---
Hmm, I really want to say this is a bug with respect of -WSystem-headers (or
what ever the spelling for that option is) as -pedantic-errors should not
warn/error for system headers. It might also be a driver issue t
--- Comment #1 from pinskia at gcc dot gnu dot org 2005-11-30 02:26 ---
Confirmed, also happens on ppc-darwin. (Note this is most likely on other
targets too, there is a bug about x86_64 and this IIRC to get libuwind to pass
its tests).
--
pinskia at gcc dot gnu dot org changed:
--- Comment #3 from pinskia at gcc dot gnu dot org 2005-11-30 02:24 ---
(In reply to comment #2)
> That is rather unfortunate.
> Because if you use the -H option then you will get different output depending
> on whether you use "-g" or not. So while you might be helping the
> "-save-tem
--- Comment #2 from jw203198 at hotmail dot com 2005-11-30 02:18 ---
(In reply to comment #1)
> The current directory is outputted so that when compiling with -save-temps -g
> is no different from -g. The double // is not really a problem at all.
>
That is rather unfortunate.
Because
libjava/testsuite/libjava.lang/Array_3.java has failed on powerpc64-linux for
quite a while. See
http://gcc.gnu.org/ml/gcc-testresults/2005-03/msg01215.html. The problem is
that in
static int baz ()
{
int[] x = (int[])null;
int nn = x.length;
return 5;
}
...
try
{
--- Comment #1 from pinskia at gcc dot gnu dot org 2005-11-30 02:14 ---
The current directory is outputted so that when compiling with -save-temps -g
is no different from -g. The double // is not really a problem at all.
--
pinskia at gcc dot gnu dot org changed:
What
--- Comment #3 from pinskia at gcc dot gnu dot org 2005-11-30 02:13 ---
(In reply to comment #2)
> But surely that example isn't evidence of a problem.
yes it is, since it is invalid code, I repeat it is invalid code. How many
more times do I need to repeat that?
Just add a cast to ge
Create this file, x.c:
main(){}
Then pre-process with, "cc -E -g x.c" (the -g is significant).
The output will contain a line such as:
# 1 "/home/me/testdir//"
This seems to be at odds with all other such annotations.
Why is the current directory output? And why the double /?
--
--- Comment #6 from jvdelisle at gcc dot gnu dot org 2005-11-30 02:09
---
Subject: Bug 25109
Author: jvdelisle
Date: Wed Nov 30 02:09:13 2005
New Revision: 107697
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=107697
Log:
2005-11-29 Jerry DeLisle <[EMAIL PROTECTED]>
--- Comment #2 from jw203198 at hotmail dot com 2005-11-30 02:07 ---
But surely that example isn't evidence of a problem.
Because if somebody codes:
int *p = f();
does it really matter what the signedness of f()'s return value is?
The caller has a pointer which is most definitely
gfortran seems to ignore explicit return types of functions.
Example follows:
-bug.for
PROGRAM BUG
external real function Mfunc
real*4 X,Y
X = Xfunc()
Y = Mfunc()
write(6,*) X,Y
stop
end
--- Comment #1 from pinskia at gcc dot gnu dot org 2005-11-30 01:36 ---
Because the code is really invalid C.
that is:
int *a;
unsigned int *f(void)
{
return a;
}
Is invalid C. We used to just output a warning with -pedantic but now the
warning is always on unless you add -Wno-pointe
--- Comment #19 from pinskia at gcc dot gnu dot org 2005-11-30 01:33
---
*** Bug 25170 has been marked as a duplicate of this bug. ***
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
-
--- Comment #1 from pinskia at gcc dot gnu dot org 2005-11-30 01:33 ---
This has been an error since 3.0.x.
And it is an error in your code.
*** This bug has been marked as a duplicate of 7976 ***
--
pinskia at gcc dot gnu dot org changed:
What|Removed
I refer to the very frequent warnings about signed/unsigned mismatches.
Can I please say this:
What does it matter if the signedness of pointers don't match? Surely it only
*really* matters when a pointer is actually dereferenced. That is the only
time when signedness really matters. The rest o
Executing on host: /mnt/gnu/gcc-3.3/objdir/gcc/testsuite/../gfortran
-B/mnt/gnu/
gcc-3.3/objdir/gcc/testsuite/../
/mnt/gnu/gcc-3.3/gcc/gcc/testsuite/gfortran.dg/
module_equivalence_1.f90 -O0 -pedantic-errors
-L/mnt/gnu/gcc-3.3/objdir/hpp
a2.0w-hp-hpux11.11/./libgfortran/.libs
-L/mnt/gnu/gcc-3.
Executing on host: /mnt/gnu/gcc-3.3/objdir/gcc/testsuite/../gfortran
-B/mnt/gnu/
gcc-3.3/objdir/gcc/testsuite/../
/mnt/gnu/gcc-3.3/gcc/gcc/testsuite/gfortran.dg/
mixed_io_1.f90 -O0 -pedantic-errors
/mnt/gnu/gcc-3.3/gcc/gcc/testsuite/gfor
tran.dg/mixed_io_1.c
-L/mnt/gnu/gcc-3.3/objdir/hppa2.0w
Here is the test code:
typedef struct
{
int ele;
} tstruct;
#define TST(r,t) r##t = 1
void test()
{
tstruct ts;
TST(ts,.ele);
}
It generates the error:
"error: pasting "ts" and "." does not give a valid preprocessing token"
In previous
--- Comment #5 from geoffk at geoffk dot org 2005-11-30 01:01 ---
Subject: Re: aliases, including weakref, break alias analysis
On 29/11/2005, at 5:55 AM, aoliva at gcc dot gnu dot org wrote:
> Make it a weak alias, then.
A weak alias is still an alias and still not supported by Mac
--- Comment #1 from ghazi at gcc dot gnu dot org 2005-11-30 00:44 ---
Based on the date it started failing, I'm guessing it was this patch that
triggered it:
2005-11-07 Paolo Bonzini <[EMAIL PROTECTED]>
PR c/24599
* c-typeck.c (build_c_cast): Try using a shared consta
On i686-unknown-linux-gnu with the 4.0.x branch, I'm getting tree checking
failures in gcc.dg/20040203-1.c, cast-1.c, cast-2.c, cast-3.c. From the
logfile, the errors in 20040203-1.c are of this form:
20040203-1.c:17: internal compiler error: tree check: expected class
'constant', have 'unary' (n
--- Comment #4 from joseph at codesourcery dot com 2005-11-30 00:35 ---
Subject: Re: New: FAIL: gcc.c-torture/execute/conversion.c
compilation
As explained in bug 24998, I can't test on PA at present but the fix is
probably similar to that for IA64.
--
http://gcc.gnu.org/bugzil
--- Comment #1 from joseph at codesourcery dot com 2005-11-30 00:34 ---
Subject: Re: New: FAIL: gcc.dg/weak/weak-14.c
Isn't this just bug 24478?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25167
Executing on host: /mnt/gnu/gcc-3.3/objdir/gcc/testsuite/../g++
-B/mnt/gnu/gcc-3
.3/objdir/gcc/testsuite/../
/mnt/gnu/gcc-3.3/gcc/gcc/testsuite/g++.old-deja/g++.
abi/cxa_vec.C -nostdinc++
-I/mnt/gnu/gcc-3.3/objdir/hppa2.0w-hp-hpux11.11/libst
dc++-v3/include/hppa2.0w-hp-hpux11.11
-I/mnt/gnu/gcc-3.3
Executing on host: /mnt/gnu/gcc-3.3/objdir/gcc/xgcc
-B/mnt/gnu/gcc-3.3/objdir/gc
c/ /mnt/gnu/gcc-3.3/gcc/gcc/testsuite/gcc.dg/weak/weak-14.c -O2 -fno-common
-f
no-show-column -lm -o ./weak-14.exe(timeout = 300)
/mnt/gnu/gcc-3.3/gcc/gcc/testsuite/gcc.dg/weak/weak-14.c:20: error: alias
defin
--- Comment #3 from pinskia at gcc dot gnu dot org 2005-11-30 00:00 ---
This is a regression.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Su
--- Comment #2 from danglin at gcc dot gnu dot org 2005-11-30 00:00 ---
And gcc.dg/torture/fp-int-convert-timode.c.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25166
--- Comment #1 from danglin at gcc dot gnu dot org 2005-11-29 23:58 ---
The same error also causes the failure of
gcc.dg/torture/fp-int-convert-long-double.c:
Executing on host: /mnt/gnu/gcc-3.3/objdir/gcc/xgcc
-B/mnt/gnu/gcc-3.3/objdir/gc
c/
/mnt/gnu/gcc-3.3/gcc/gcc/testsuite/gcc.dg/tor
--- Comment #7 from danglin at gcc dot gnu dot org 2005-11-29 23:55 ---
The same errors occur on hppa2.0w-hp-hpux11.11 except that ld also dumps
core:
collect2: ld terminated with signal 10 [Bus error], core dumped
/usr/bin/ld: Unsatisfied symbols:
wf1 (first referenced in /var/tmp//
--- Comment #11 from ed at catmur dot co dot uk 2005-11-29 23:55 ---
This is not invalid. Yes, reiser4 is broken, but so is gcc.
gcc (3.4.4, I haven't used 4.x yet) dies when it encounters an inaccessible
directory as a subdirectory of a member of the include path. See comment 6.
--
Executing on host: /mnt/gnu/gcc-3.3/objdir/gcc/xgcc
-B/mnt/gnu/gcc-3.3/objdir/gc
c/ /mnt/gnu/gcc-3.3/gcc/gcc/testsuite/gcc.c-torture/execute/conversion.c -w
-O
0 -fno-show-column -lm -o
/mnt/gnu/gcc-3.3/objdir/gcc/testsuite/conversion.x
0(timeout = 300)
/usr/bin/ld: Unsatisfied symbols:
--- Comment #10 from pinskia at gcc dot gnu dot org 2005-11-29 23:23
---
(In reply to comment #1)
> This is a duplicate of 11242 which was (incorrectly) closed.
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11242
That is because it was not reported against mingw32 but a GNU/Linux OS.
A
--- Comment #2 from gdr at integrable-solutions dot net 2005-11-29 23:20
---
Subject: Re: New: [3.4 Regression] g++.dg/abi/vtt1.C failure with
"-funit-at-a-time"
"reichelt at gcc dot gnu dot org" <[EMAIL PROTECTED]> writes:
| The test g++.dg/abi/vtt1.C is failing with "-funit-at-a-t
--- Comment #6 from pinskia at gcc dot gnu dot org 2005-11-29 23:08 ---
Shorter testcase:
template struct F;
template F &operator<<(F&, const char*);
template
struct F
{
friend F& operator<< <>(F&, const char*);
};
typedef F F1;
template struct X;
typedef X X2;
F1& operator<< (F1&, c
--- Comment #6 from tromey at gcc dot gnu dot org 2005-11-29 22:37 ---
I tried the included test case with current svn head.
I had to add '.' to classpath, then it worked for me:
opsy. CLASSPATH=test.jar:. gij test1
Microsoft is crap
Compiling as in comment #4 also worked fine.
I also
--- Comment #6 from dave at hiauly1 dot hia dot nrc dot ca 2005-11-29
22:00 ---
Subject: Re: Failure to build, :1:2: error: missing '(' after
predicate
> On Sun, Nov 27, 2005 at 06:57:05PM -, danglin at gcc dot gnu dot org
> wrote:
> > The "-Aa" option is likely the problem. It'
--- Comment #2 from ghazi at gcc dot gnu dot org 2005-11-29 21:46 ---
Hmm this is convoluted, but I think I know what's going on:
We're running the builtin fprintf check. I recently added a small sanity check
to ensure that fprintf_unlocked also works. Now we're getting an unresolved
--- Comment #10 from tromey at gcc dot gnu dot org 2005-11-29 21:14 ---
Fix checked in.
--
tromey at gcc dot gnu dot org changed:
What|Removed |Added
Status|A
--- Comment #5 from pda at freeshell dot org 2005-11-29 21:03 ---
Subject: Re: Failure to build, :1:2: error: missing '(' after
predicate
On Fri, Nov 25, 2005 at 08:27:34PM -, pinskia at gcc dot gnu dot org wrote:
> How did you configure GCC?
> >From using google it sounds like you
--- Comment #1 from pinskia at gcc dot gnu dot org 2005-11-29 20:57 ---
Confirmed.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
CC|
--- Comment #1 from reichelt at gcc dot gnu dot org 2005-11-29 20:53
---
> The test g++.dg/abi/vtt1.C is failing with "-funit-at-a-time"
> (or optimization "-O" or higher since this enables "-funit-at-a-time"
Sorry, this should read "-O2" instead of just "-O".
--
http://gcc.gnu.or
The test g++.dg/abi/vtt1.C is failing with "-funit-at-a-time"
(or optimization "-O" or higher since this enables "-funit-at-a-time"
by default).
GCC 4.x.y doesn't seem to be affected.
Since this behaviour can be triggered with just "-O2" I rate this as
a regression (although "-funit-at-a-time" was
export OMP_NUM_THREADS=2
gfortran-gomp -v Using built-in specs.
Target: i686-pc-linux-gnu
Configured with: ../gcc/configure --prefix=/usr/local/gomp
--program-suffix=-gomp --enable-threads=posix --enable-languages=c,c++,fortran
Thread model: posix
gcc version 4.2.0-gomp-20050608-branch 20051126 (e
--- Comment #2 from andreast at gcc dot gnu dot org 2005-11-29 20:45
---
Compilation succeeded on sparc-solaris8 this morning, just after bje ci'ed I
had to use the same patch.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25157
--- Comment #1 from pinskia at gcc dot gnu dot org 2005-11-29 20:44 ---
I would assume this was caused by the patch which fixed PR 25022.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
---
--- Comment #4 from pda at freeshell dot org 2005-11-29 20:41 ---
Subject: Re: Failure to build, :1:2: error: missing '(' after
predicate
On Sun, Nov 27, 2005 at 06:57:05PM -, danglin at gcc dot gnu dot org wrote:
> The "-Aa" option is likely the problem. It's probably set in
> CF
--- Comment #1 from pinskia at gcc dot gnu dot org 2005-11-29 20:41 ---
The patch here:
http://gcc.gnu.org/ml/gcc-patches/2005-11/msg02046.html
Should fix it (but I have no warranties).
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Adde
--- Comment #5 from gdr at gcc dot gnu dot org 2005-11-29 20:37 ---
(In reply to comment #4)
> | instead of the FUNCTION_TYPE's type which has the correct type, I might fix
> | this later tonight.
>
> Try this
this patch is incomplete.
--
gdr at gcc dot gnu dot org changed:
GCC version :
4.0.2
Complete C file :
int a;
char buf[(int)&a];
command line :
gcc-4.0 -c file.c
Error message :
t.c:3: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See http://gcc.gnu.org/bugs.html>
--- Comment #5 from reichelt at gcc dot gnu dot org 2005-11-29 20:33
---
This is not a bug in GCC. The code is invalid.
Let's consider the following simplified example which is just a
part of the diamond:
A0
|
A
/
B
==
struct A0
{
A0(int);
};
struc
--- Comment #9 from toon at moene dot indiv dot nluug dot nl 2005-11-29
19:20 ---
FX,
Your patch solved the problem for me.
AFAICS, this patch is indeed the correct fix for this problem.
Please apply it to (at least) 4.1 and trunk.
It might be appropriate for the 4.0 branch, too.
--
-linux-gnuspe/head-glibc-2.3.3/powerpc-linux-gnuspe
--disable-multilib --with-newlib --without-headers --disable-nls
--enable-threads=no --enable-symvers=gnu --enable-__cxa_atexit
--enable-languages=c --disable-shared
Thread model: single
gcc version 4.2.0 20051129 (experimental)
According to the
--- Comment #5 from reichelt at gcc dot gnu dot org 2005-11-29 19:03
---
Confirmed.
Here's a reduced testcase.
Compile with "g++ --param ggc-min-expand=0 --param ggc-min-heapsize=0".
template struct Dummy;
void dummy();
--- Comment #9 from tromey at gcc dot gnu dot org 2005-11-29 18:58 ---
Subject: Bug 18278
Author: tromey
Date: Tue Nov 29 18:58:23 2005
New Revision: 107677
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=107677
Log:
gcc/java:
PR java/18278:
* expr.c (build_jni_st
--- Comment #6 from reichelt at igpm dot rwth-aachen dot de 2005-11-29
18:44 ---
Subject: Re: [4.0/4.1/4.2 Regression] ICE on throw code
On 29 Nov, janis at gcc dot gnu dot org wrote:
> Andrew, are you sure about 3.4.0 crashing for this testcase? I tried mainline
> as far back as 200
--- Comment #4 from jb at gcc dot gnu dot org 2005-11-29 18:37 ---
It doesn't have anything to do with array transfers. The following test program
is equivalent to the one in #3, but uses the scalar transfer_integer instead of
transfer_array:
program test3
integer dat(5)
dat = (/ 1,
--- Comment #5 from janis at gcc dot gnu dot org 2005-11-29 18:37 ---
The question in my previous comment should have been to Volker, not Andrew.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24996
--- Comment #4 from janis at gcc dot gnu dot org 2005-11-29 18:36 ---
A regression hunt identified the following patch (not terribly useful):
r81764 | dnovillo | 2004-05-13 06:41:07 + (Thu, 13 May 2004) | 3 lines
Merge tree-ssa-20020619-branch into mainline.
Andrew, are you su
--- Comment #8 from tromey at gcc dot gnu dot org 2005-11-29 18:35 ---
Subject: Bug 18278
Author: tromey
Date: Tue Nov 29 18:34:58 2005
New Revision: 107676
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=107676
Log:
gcc/java:
PR java/18278:
* expr.c (build_jni_st
--- Comment #2 from tromey at gcc dot gnu dot org 2005-11-29 18:31 ---
In the version of ant I have (1.6.2) the class is
org.apache.tools.ant.launch.Launcher
not
org.apache.tools.ant.launcher.Launcher
So I think this is just a typo in the --main argument.
--
tromey at gcc dot gnu d
At -O1, -O2 and -O3:
Executing on host: /mnt/gnu/gcc-3.3/objdir/gcc/xgcc
-B/mnt/gnu/gcc-3.3/objdir/gc
c/ /mnt/gnu/gcc-3.3/gcc/gcc/testsuite/gcc.c-torture/execute/builtins/fputs.c
/mn
t/gnu/gcc-3.3/gcc/gcc/testsuite/gcc.c-torture/execute/builtins/fputs-lib.c
/mnt/
gnu/gcc-3.3/gcc/gcc/testsuite/gcc.
At all optimizations:
Executing on host: /mnt/gnu/gcc-3.3/objdir/gcc/xgcc
-B/mnt/gnu/gcc-3.3/objdir/gc
c/ /mnt/gnu/gcc-3.3/gcc/gcc/testsuite/gcc.c-torture/execute/builtins/fprintf.c
/
mnt/gnu/gcc-3.3/gcc/gcc/testsuite/gcc.c-torture/execute/builtins/fprintf-lib.c
/
mnt/gnu/gcc-3.3/gcc/gcc/testsuite
make[1]: Entering directory `/xxx/gnu/gcc-3.3/objdir/libdecnumber'
source='../../gcc/libdecnumber/decNumber.c' object='decNumber.o' libtool=no gcc
-I../../gcc/libdecnumber -I. -g -O2 -W -Wall -Wwrite-strings
-Wstrict-prototyp
es -Wmissing-prototypes -Wold-style-definition -Wmissing-format-attribu
--- Comment #4 from gdr at integrable-solutions dot net 2005-11-29 18:02
---
Subject: Re: [3.4/4.0/4.1 Regression] wrong error message (int instead of
bool)
"pinskia at gcc dot gnu dot org" <[EMAIL PROTECTED]> writes:
| Hmm, it is looking at the wrong type, it is looking at the RESUL
--- Comment #5 from bonzini at gnu dot org 2005-11-29 17:55 ---
Frederic, "-Wstrict-aliasing" is only working for C up until 4.1 (included).
Your code, rewritten to C, would issue a warning with 4.1 and earlier compilers
as well. Also note that -Wstrict-aliasing does not prevent the co
--- Comment #3 from gdr at integrable-solutions dot net 2005-11-29 17:48
---
Subject: Re: [3.4/4.0/4.1 Regression] wrong error message (int instead of
bool)
"pinskia at gcc dot gnu dot org" <[EMAIL PROTECTED]> writes:
| Hmm, it is looking at the wrong type, it is looking at the RESUL
--- Comment #4 from pinskia at gcc dot gnu dot org 2005-11-29 17:47 ---
(In reply to comment #3)
> I know it breaks aliasing rules, but this problem was not present in gcc-3.3.
So ...
GCC 4.1 includes more optimization than 3.3 did which takes advantage of the
aliasing rules.
Note thi
--- Comment #3 from frederic dot devernay at m4x dot org 2005-11-29 17:42
---
I know it breaks aliasing rules, but this problem was not present in gcc-3.3.
Most gcc users consider "-O2" as being a safe optimizing option, (e.g. redhat
RPMs are compiled with gcc -O2), and generating wron
--- Comment #3 from pinskia at gcc dot gnu dot org 2005-11-29 17:36 ---
I should also note that compare_values is does not respect -fwrapv either as
shown by the testcase in PR 25145 in comment #1 and comment #6.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25148
--- Comment #6 from pinskia at gcc dot gnu dot org 2005-11-29 17:34 ---
(In reply to comment #5)
> Note this was the simple fix which exposes those latent bugs as far as I can
> see that should work, we get the correct range but the rest of VRP goes
> bonkers:
I should also note it does
--- Comment #7 from tromey at gcc dot gnu dot org 2005-11-29 17:13 ---
Fix checked in everywhere.
--
tromey at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #4 from tromey at gcc dot gnu dot org 2005-11-29 17:09 ---
*** Bug 25040 has been marked as a duplicate of this bug. ***
--
tromey at gcc dot gnu dot org changed:
What|Removed |Added
-
--- Comment #2 from tromey at gcc dot gnu dot org 2005-11-29 17:09 ---
*** This bug has been marked as a duplicate of 15411 ***
--
tromey at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #2 from pinskia at gcc dot gnu dot org 2005-11-29 16:46 ---
Hmm, it is looking at the wrong type, it is looking at the RESULT_DECL's type
instead of the FUNCTION_TYPE's type which has the correct type, I might fix
this later tonight.
--
http://gcc.gnu.org/bugzilla/show_b
--- Comment #1 from pinskia at gcc dot gnu dot org 2005-11-29 16:40 ---
Confirmed, all an integer types give the wrong error (well except for int).
3.3 gave:
t.c: In function `bool func()':
t.c:3: error: return-statement with no value, in function declared with a
non-void return type
--- Comment #1 from schwab at suse dot de 2005-11-29 16:13 ---
This is the effect of the integer promotion rules.
--
schwab at suse dot de changed:
What|Removed |Added
bool func()
{
return;
}
a.cc:3: error: return-statement with no value, in function returning 'int'
--
Summary: wrong error message (int instead of bool)
Product: gcc
Version: 4.0.2
Status: UNCONFIRMED
Severity: minor
Priority: P
--- Comment #38 from carlo at gcc dot gnu dot org 2005-11-29 16:05 ---
That never works because it always defines _REENTRANT.
You probably mean:
#ifdef _REENTRANT
#define GCC_BUGFIX 1
#endif
#include
#undef _REENTRANT
#ifdef GCC_BUGFIX
#define _REENTRANT 1
#endif
The drawback of this
I don't know whether it is a bug or not.
Consider the following example:
short int sh1 = 1;
short int sh2 = 2;
printf("Size of short: %d\n", sizeof(short));
printf("Size of int: %d\n", sizeof(int));
printf("Size of expr: %d\n", sizeof(sh1+sh2));
Output:
Size of short: 2
Size of int: 4
Size of e
--- Comment #2 from pinskia at gcc dot gnu dot org 2005-11-29 16:01 ---
A start is done on the branch libobjc_noheaders. I almost forgot about
thrd-objc.c.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24775
1 - 100 of 130 matches
Mail list logo
