--- Comment #1 from hp at gcc dot gnu dot org 2009-10-26 21:46 ---
To whomever will fix this: beware not to introduce warnings for targets where
the packed layout is the default. (As has happened in the past for other
packed warnings.)
--
hp at gcc dot gnu dot org changed:
--- Comment #3 from jamborm at gcc dot gnu dot org 2009-10-26 21:53 ---
Created an attachment (id=18903)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=18903action=view)
Proposed fix
Indeed, the users of build_ref_for_offset in ipa-prop.c and ipa-cp.c
do not unshare the base
--- Comment #6 from b3timmons at speedymail dot org 2009-10-26 21:59
---
Right, I just tried it with a later gcc build and now cannot reproduce it.
Looks like we should close it.
--
b3timmons at speedymail dot org changed:
What|Removed |Added
/gcc/configure --with-mpfr=/usr/local
--with-gmp=/usr/local --with-ppl=/usr/local --with-cloog=/usr/local
--with-mpc=/usr/local --with-libelf=/usr/local --enable-languages=c,c++
--enable-__cxa_atexit --enable-targets=all
Thread model: posix
gcc version 4.5.0 20091026 (experimental) (GCC
--- Comment #1 from b3timmons at speedymail dot org 2009-10-26 22:27
---
Created an attachment (id=18904)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=18904action=view)
preprocessed source triggering failure
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41837
--- Comment #2 from b3timmons at speedymail dot org 2009-10-26 22:28
---
Created an attachment (id=18905)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=18905action=view)
preprocessed source triggering failure
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41837
--- Comment #3 from b3timmons at speedymail dot org 2009-10-26 22:30
---
Created an attachment (id=18906)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=18906action=view)
preprocessed source triggering failure
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41837
--- Comment #4 from b3timmons at speedymail dot org 2009-10-26 22:33
---
Created an attachment (id=18907)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=18907action=view)
gzipped preprocessed source triggering failure
Had to gzip file due to submission timing out
--
--- Comment #2 from kargl at gcc dot gnu dot org 2009-10-26 22:39 ---
My conclusion is that there is a problem when the
NAN is passed from the C to the fortran.
That's not my conclusion. :)
double C2F(returnanan)(void)
{
(snip)
PROGRAM test
DATA ONE/1.0D0/
--- Comment #1 from jyasskin at gmail dot com 2009-10-26 23:04 ---
Created an attachment (id=18908)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=18908action=view)
File with incorrect strict-aliasing warning
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41838
The attached Triple.i, when compiled with `g++ -c -O2 -Wstrict-aliasing
Triple.i`, returns:
Triple.i: In function 'void setOSName(const StringRef)':
Triple.i:9: warning: dereferencing pointer 'anonymous' does break
strict-aliasing rules
Triple.i:9: note: initialized from here
Line 9 is:
if
--- Comment #3 from jakub at gcc dot gnu dot org 2009-10-26 23:11 ---
Fixed.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #8 from jakub at gcc dot gnu dot org 2009-10-26 23:12 ---
Fixed. x86_64-linux and i686-linux bootstrapped/regtested with
--disable-checking just fine.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #2 from pinskia at gcc dot gnu dot org 2009-10-26 23:42 ---
There's no aliasing through the void*.
This is not true. In fact aliasing is not about what the type of the pointers
are but what the dynamic type of what the pointer points to and the access
type.
But that is
--- Comment #2 from dave at hiauly1 dot hia dot nrc dot ca 2009-10-26
23:43 ---
Subject: Re: cc1: error: Cannot load plugin ./selfassign.so
There is configure fluff in gcc/configure.ac at the --enable-plugin handling
that should deal with this (it tries -rdynamic at the moment).
--- Comment #4 from pinskia at gcc dot gnu dot org 2009-10-27 00:26 ---
I have a C testcase which shows the same issue do you want it? It shows up
while building libxml2.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41808
--- Comment #5 from mckelvey at maskull dot com 2009-10-27 01:04 ---
(In reply to comment #4)
Subject: Re: ICE in LTO when linking
On Sun, 25 Oct 2009, mckelvey at maskull dot com wrote:
--- Comment #3 from mckelvey at maskull dot com 2009-10-25 16:32
---
--- Comment #5 from pinskia at gcc dot gnu dot org 2009-10-27 01:14 ---
(In reply to comment #4)
I have a C testcase which shows the same issue do you want it? It shows up
while building libxml2.
Actually looks related but different, there is no inlining involved there.
--
--- Comment #6 from pinskia at gcc dot gnu dot org 2009-10-27 01:17 ---
stest0.o and libPatternDriver.a
Actually since you are not using the gold linker plugin we only really need the
preprocessed source for stest0.o . Since collect2 does not understand archives
right now.
--
Take the following two files:
typedef struct _xmlDict xmlDict;
struct _xmlDict {
int ref_counter;
};
void xmlDictCreate(void) {
xmlDict * dict;
}
--- CUT ---
typedef struct _xmlDict xmlDict;
struct _xmlDoc {
struct _xmlDict *dict;
};
void xmlAddEntity(struct _xmlDoc *a) {
xmlDict * dict =
--- Comment #1 from pinskia at gcc dot gnu dot org 2009-10-27 02:31 ---
Note this originally comes from libxml2.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41839
--- Comment #3 from pinskia at gcc dot gnu dot org 2009-10-27 02:38 ---
Confirmed.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
CC|
--- Comment #1 from pinskia at gcc dot gnu dot org 2009-10-27 02:42 ---
Confirmed.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #5 from pinskia at gcc dot gnu dot org 2009-10-27 02:44 ---
These might not be bogus as there is extra inlining at -O3 which causes us to
strip out the address taking and make search_table_entries, etc. look like real
variables.
--
Following code is not compiling with g++ 4.2.3, but compiles fine with g++
3.3.3:
==
#includeiostream
class mybase {
public:
virtual int afunc()=0;
};
main()
{
mybase (*ptrs)[10];
}
==
Error given by g++4.2.3 is as below for the above code.
with: /home/b3po/build/gcc/gcc/configure --with-mpfr=/usr/local
--with-gmp=/usr/local --with-ppl=/usr/local --with-cloog=/usr/local
--with-mpc=/usr/local --with-libelf=/usr/local --enable-languages=c,c++
--enable-__cxa_atexit --enable-targets=all
Thread model: posix
gcc version 4.5.0 20091026 (experimental
--- Comment #1 from b3timmons at speedymail dot org 2009-10-27 05:07
---
Created an attachment (id=18910)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=18910action=view)
gzipped preprocessed source triggering failure
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41841
This happened in real code, where a required header file was missing.
$ cat bug.i
void f() {
char x[g(h)];
}
$ gcc-4.4.2 -c bug.i
bug.i: In function f:
bug.i:2: error: h undeclared (first use in this function)
bug.i:2: error: (Each undeclared identifier is reported only once
bug.i:2:
101 - 128 of 128 matches
Mail list logo
