--- Comment #19 from irar at il dot ibm dot com 2008-09-22 12:32 ---
(In reply to comment #18)
Created an attachment (id=16377)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16377action=view) [edit]
patch
Actually this one vectorizes slp-perm-1.c for me on a ppc64 cross. It
--- Comment #52 from rguenth at gcc dot gnu dot org 2008-09-22 12:40
---
Can we please have
1) -fno-exceptions
This flag turns off C++ exception handling support, which is indicated at
compile time by __GXX_EXCEPTIONS being undefined. Use of the keywords try,
catch, or throw produces
--- Comment #20 from irar at il dot ibm dot com 2008-09-22 12:41 ---
(In reply to comment #18)
Created an attachment (id=16377)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16377action=view) [edit]
patch
Actually this one vectorizes slp-perm-1.c for me on a ppc64 cross. It
--- Comment #53 from rguenth at gcc dot gnu dot org 2008-09-22 12:41
---
CCing C++ FE maintainers.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
Ran gcc testsuite and got a lot of c++ testsuite failures like for
g++.dg/eh/cond1.C:
...
terminate called after throwing an instance of 'int'
terminate called recursively
...
Compiler details:
Target: arm-linux-gnueabi
Configured with:
Follow up to PR 37486.
see also http://gcc.gnu.org/ml/fortran/2008-09/msg00366.html
The following program should print 5 but it prints 0 with default options
in gfortran. The problem is that in the program, the common object is padded as
4bytes padding + i + r8
and then in the subroutine, only
--- Comment #4 from manu at gcc dot gnu dot org 2008-09-22 13:28 ---
I don't understand why this is not closed as wontfix. These warnings are coming
from the front-end.
Unless we do one of the following:
a) some CCP and VRP in the FE, or
b) move the warnings to the middle-end,
this
--- Comment #1 from sposelenov at emcraft dot com 2008-09-22 13:32 ---
Created an attachment (id=16378)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16378action=view)
Reduced testcase
Build as:
arm-linux-gnueabi-g++ -o cond1 cond1.C
--
--- Comment #2 from daniel dot diaz at univ-paris1 dot fr 2008-09-22 14:08
---
Richard Guenther says it is impossible in C... In C maybe but not in GNU C :-)
This worked perfectly under older versions (may gcc 3.x.x) and it is a pitty if
this feature is lost now.
--
--- Comment #54 from paolo dot carlini at oracle dot com 2008-09-22 14:08
---
(In reply to comment #52)
It seems the library maintainers are not willing to see this problem.
Asking for a new C++ frontend feature as an excuse to not fix the
library problem is very lame IMHO.
For the
--- Comment #6 from dave at hiauly1 dot hia dot nrc dot ca 2008-09-22
14:11 ---
Subject: Re: [4.4 Regression] libgcc2.c:1981: ICE:
vector VEC(m em_ref_p,base) index domain error, in
create_vop_ref_mapping_loop at tree-ssa-lo op-im.c:1519
Attached .i.
Dave
---
--- Comment #39 from b0ntrict0r at yandex dot ru 2008-09-22 15:12 ---
Created an attachment (id=16380)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16380action=view)
Updated fixed
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14912
--- Comment #30 from dje at gcc dot gnu dot org 2008-09-22 15:17 ---
Subject: Bug 33642
Author: dje
Date: Mon Sep 22 15:15:56 2008
New Revision: 140551
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=140551
Log:
PR rtl-optimization/33642
*
--- Comment #5 from tromey at gcc dot gnu dot org 2008-09-22 15:19 ---
No, I think we have to record what the user actually wrote.
For instance, consider:
#include string
using namespace std;
std::string x1;
string x2;
If we record the fully qualified name, we can't distinguish these
--- Comment #6 from jan dot kratochvil at redhat dot com 2008-09-22 15:32
---
Tom, could you elaborate why x1 and x2 should be printed differently?
I do not say they should not but I do not see a clear reason for either way.
Should we also try to record the source name of this
--- Comment #5 from zackw at panix dot com 2008-09-22 15:46 ---
I'm not monitoring consensus of developers anymore, but I think we *should*
either move these warnings to the middle end or do some CCP/VRP in the front
ends. The -Wuninitialized warnings are a lot less trouble than the
Dear Nrao Email Account User,
We wrote to you on 29 july 2008 advising that you change the
password on your account in order to prevent any unauthorised
account access following the network instruction we previously
communicated.
All Mailhub systems will undergo regularly scheduled maintenance.
--- Comment #8 from sje at cup dot hp dot com 2008-09-22 16:35 ---
I'll try to review the IA64 sel-sched patch over the next couple of weeks. I
can't approve the workaround since it is not IA64 specific (it is in
haifa-sched.c).
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37381
--- Comment #6 from manu at gcc dot gnu dot org 2008-09-22 16:43 ---
Then, try to raise the issue in GCC and outline a clear plan. Otherwise, I
assure you this is going to stay as-is for the next 5-10 years because nobody
has a clear idea on how to tackle this in a way that pleases a
--- Comment #1 from janus at gcc dot gnu dot org 2008-09-22 16:50 ---
Created an attachment (id=16381)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16381action=view)
patch
The attached patch is as far as I got with this up to now. It regtests fine and
makes the following modified
Lines with 'break', 'goto', and 'continue' are not available for
debugging.
On GCC compilers, one can not set a breakpoint on lines with 'goto', 'break',
or
'continue' statements. This is incorrect, since these lines contain
user statements. Also, for example IBM XLC compilers on AIX 6.1 have
--- Comment #3 from paolo dot carlini at oracle dot com 2008-09-22 17:25
---
I'm fixing this, together with DR 23.
--
paolo dot carlini at oracle dot com changed:
What|Removed |Added
--- Comment #2 from bkoz at gcc dot gnu dot org 2008-09-22 19:06 ---
All the test and example links in pb_ds that point to source files are broken.
These include links linked to from these pages:
libstdc++-v3/doc/html/ext/pb_ds/assoc_examples.html
--- Comment #5 from rguenth at gcc dot gnu dot org 2008-09-22 19:17 ---
Fixed.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
not
recursively generate expressions if running FRE.
* gcc.c-torture/compile/20080922-1.c: New testcase.
* gcc.dg/tree-ssa/ssa-fre-7.c: Remove XFAIL.
* gcc.dg/tree-ssa/ssa-fre-8.c: Likewise.
* gcc.dg/tree-ssa/ssa-fre-9.c: Likewise.
Added:
trunk/gcc
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Severity|normal |enhancement
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37537
--- Comment #1 from pinskia at gcc dot gnu dot org 2008-09-22 19:39 ---
Related to PR 29609.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37616
--- Comment #2 from pinskia at gcc dot gnu dot org 2008-09-22 19:44 ---
Also related to PR 36690.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37616
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNCONFIRMED |WAITING
Component|c |bootstrap
--- Comment #3 from daney at gcc dot gnu dot org 2008-09-22 19:49 ---
Subject: Bug 37593
Author: daney
Date: Mon Sep 22 19:48:09 2008
New Revision: 140563
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=140563
Log:
2008-09-22 David Daney [EMAIL PROTECTED]
PR
--- Comment #4 from daney at gcc dot gnu dot org 2008-09-22 19:52 ---
Fix committed.
--
daney at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #4 from pinskia at gcc dot gnu dot org 2008-09-22 19:54 ---
This is invalid code as the difference between ((EDIDDetailed *)
pExt-checksum) - pDTD is 88 which is not a multiple of 18. This happens the
same way on powerpc.
--
pinskia at gcc dot gnu dot org changed:
--- Comment #4 from pinskia at gcc dot gnu dot org 2008-09-22 19:55 ---
Is part of GNU C++98 now?
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #2 from pinskia at gcc dot gnu dot org 2008-09-22 19:57 ---
Can you try this again?
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #16 from daney at gcc dot gnu dot org 2008-09-22 20:08 ---
This is biting me now. Perhaps I may look at fixing it (or maybe not...).
--
daney at gcc dot gnu dot org changed:
What|Removed |Added
-bootstrap --disable-static --prefix=/home/edwin/gcc_inst/
Thread model: posix
gcc version 4.4.0 20080922 (experimental) (GCC)
$ x86_64-unknown-linux-gnu-gcc-4.4.0 -v
Using built-in specs.
Target: x86_64-unknown-linux-gnu
Configured with: ../gcc/configure --enable-languages=c --disable-multilib
--- Comment #1 from edwintorok at gmail dot com 2008-09-22 20:18 ---
/* testcase compile this with -O1 */
typedef struct TCase TCase;
typedef void (*TFun) (int);
typedef struct Suite Suite;
void _tcase_add_test (TCase * tc, TFun tf, const char *fname, int signal,
--- Comment #2 from burnus at gcc dot gnu dot org 2008-09-22 20:22 ---
(In reply to comment #1)
Created an attachment (id=16381)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16381action=view) [edit]
Although this test case compiles without error, it gives the wrong output:
If
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
Keywords||ice-on-valid-code
Summary|gcc 4.4 regression: ICE on
--- Comment #5 from john dot spelis at 3dlabs dot com 2008-09-22 20:40
---
Subject: Re: Pointer arithmetic yields strange result
Apologies; never realised that pointer arithmetic was defined
in that fashion (i.e. must always be a multiple of the object size).
Sorry to have bothered
--- Comment #3 from bkoz at gcc dot gnu dot org 2008-09-22 20:56 ---
Should be fixed now.
--
bkoz at gcc dot gnu dot org changed:
What|Removed |Added
Target
--- Comment #4 from bkoz at gcc dot gnu dot org 2008-09-22 20:57 ---
Subject: Bug 37391
Author: bkoz
Date: Mon Sep 22 20:56:08 2008
New Revision: 140564
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=140564
Log:
2008-09-22 Benjamin Kosnik [EMAIL PROTECTED]
PR
--- Comment #2 from pinskia at gcc dot gnu dot org 2008-09-22 21:00 ---
This worked with:
gcc version 4.4.0 20080917 (experimental) [trunk revision 140434] (GCC)
But fails with:
GNU C (GCC) version 4.4.0 20080922 (experimental) [trunk revision 140563]
(i386-apple-darwin8.11.1
--- Comment #3 from pinskia at gcc dot gnu dot org 2008-09-22 21:09 ---
Reduced even further and will reproduce with both LP32 and LP64 targets (as
long as float is 32bits and double is 64bits):
typedef union
{
char *string;
double dval;
float fval;
} yystype;
char *f(void)
{
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|--- |4.4.0
Version|lto |4.4.0
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
CC||pinskia at gcc dot gnu dot
|
--- Comment #40 from pluto at agmk dot net 2008-09-22 21:54 ---
(In reply to comment #39)
Created an attachment (id=16380)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16380action=view) [edit]
Updated fixed
1). still there is a warning:
../../gcc/cp/error.c: In function
--- Comment #8 from janis at gcc dot gnu dot org 2008-09-22 22:12 ---
Created an attachment (id=16382)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16382action=view)
small C testcase that fails with current trunk
This version of the small C testcase fails with current mainline
--- Comment #9 from pinskia at gcc dot gnu dot org 2008-09-22 22:45 ---
The one thing I noticed is that fsel is used in the -ffast-math case and it
does a subtraction.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37449
--- Comment #13 from hjl at gcc dot gnu dot org 2008-09-22 22:46 ---
Subject: Bug 37535
Author: hjl
Date: Mon Sep 22 22:44:53 2008
New Revision: 140568
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=140568
Log:
2008-09-22 Vladimir Makarov [EMAIL PROTECTED]
PR
--- Comment #10 from janis at gcc dot gnu dot org 2008-09-22 23:08 ---
The new testcase passes with -O1 -funsafe-math-optimizations
-fno-tree-dominator-opts. The dom1 dump for -O1 -funsafe-math-optimizations
twice reports Invalid sum of incoming frequencies.
--
--- Comment #17 from daney at gcc dot gnu dot org 2008-09-22 23:56 ---
For me the testcase always gets a ConcurrentModificationException in w.clear()
very soon after starting. This is on GCC trunk 140563 on x86_64-pc-linux-gnu.
If I synchronize(w) for accesses to w there is no
--- Comment #4 from paolo at gcc dot gnu dot org 2008-09-23 00:47 ---
Subject: Bug 32422
Author: paolo
Date: Tue Sep 23 00:47:02 2008
New Revision: 140574
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=140574
Log:
2008-09-23 Paolo Carlini [EMAIL PROTECTED]
PR
--- Comment #5 from paolo dot carlini at oracle dot com 2008-09-23 00:49
---
Fixed for 4.4.0.
--
paolo dot carlini at oracle dot com changed:
What|Removed |Added
--- Comment #16 from jvdelisle at gcc dot gnu dot org 2008-09-23 03:53
---
Subject: Bug 37498
Author: jvdelisle
Date: Tue Sep 23 03:52:19 2008
New Revision: 140576
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=140576
Log:
2008-09-22 Jerry DeLisle [EMAIL PROTECTED]
PR
--- Comment #1 from linuxl4 at sohu dot com 2008-09-23 05:06 ---
can anybody comfirm it?
I built gcc ppl polylib cloog mpfr gmp with CFLAGS=-pipe -O3 -ftree-vectorize
-march=pentium4 -mfpmath=sse
glibc and binutils are of my gentoo 2008.0 hosts'
--
--- Comment #7 from abel at gcc dot gnu dot org 2008-09-22 06:20 ---
The patch implementing ia64 changes needed for the selective scheduler can be
found at http://gcc.gnu.org/ml/gcc-patches/2008-08/msg01669.html. [The first
version of the patch with some summary is at
When I try to use protected function pointers in a shared library, I experience
strange behaviour: directly passing a function name triggers an invalid
relocation during linking stage while it works when assigning the function name
to a global variable first.
This report might be a duplicate to
--- Comment #1 from gernot dot hillier at siemens dot com 2008-09-22 07:19
---
Created an attachment (id=16375)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16375action=view)
C file showing the problem
See the body of f1 for the two cases. Using f3( f2) triggers the problem
--- Comment #2 from gernot dot hillier at siemens dot com 2008-09-22 07:21
---
Created an attachment (id=16376)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16376action=view)
preprocessed version of the code triggering the bug
--
--- Comment #4 from irar at gcc dot gnu dot org 2008-09-22 07:57 ---
Subject: Bug 37482
Author: irar
Date: Mon Sep 22 07:55:39 2008
New Revision: 140544
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=140544
Log:
PR tree-optimization/37482
* tree-vectorizer.h
--- Comment #38 from pluto at agmk dot net 2008-09-22 09:21 ---
(In reply to comment #37)
Created an attachment (id=16361)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16361action=view) [edit]
Updated patch
Could someone test updated patch?
it doesn't build.
--- Comment #14 from irar at il dot ibm dot com 2008-09-22 09:24 ---
This patch causes the following failures on ppc:
FAIL: gcc.dg/vect/pr35821-altivec.c scan-tree-dump-times vect vectorized 1
loops 1
FAIL: gcc.dg/vect/slp-perm-1.c scan-tree-dump-times vect vectorized 1 loops 1
FAIL:
--- Comment #4 from sherpya at netfarm dot it 2008-09-22 09:51 ---
I've also seen crashes in alloca(), at least according to gdb backtrace
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37584
--
janus at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |janus at gcc dot gnu dot org
|dot org
--- Comment #15 from rguenth at gcc dot gnu dot org 2008-09-22 10:07
---
This is because the original access is through a restricted pointer, so the
check is conservatively correct at this point. We can move it to the
point where the vector pointer is created (but somehow I cannot
--- Comment #16 from irar at il dot ibm dot com 2008-09-22 10:33 ---
(In reply to comment #15)
This is because the original access is through a restricted pointer, so the
check is conservatively correct at this point. We can move it to the
point where the vector pointer is created
I get configure error in libgcc/configure.
the new gcc compiler seem build and here, but the script as(attached) does exec
a empty name in line 77 and report exec : : not found.
this happen too when i want compile a program by hand and type
xgcc test.c
where test.c contain this
main()
{
}
--- Comment #17 from rguenther at suse dot de 2008-09-22 11:14 ---
Subject: Re: [4.4 Regression] Revision 140257 causes
vectorizer tests failures
On Mon, 22 Sep 2008, irar at il dot ibm dot com wrote:
--- Comment #16 from irar at il dot ibm dot com 2008-09-22 10:33 ---
--- Comment #1 from brian at dessent dot net 2008-09-22 11:15 ---
Subject: Re: New: libgcc/configure (as script)fail with line 77
exec : : not found.
The error means you don't have a cross-assembler installed correctly
prior to configuring gcc. gcc is just a compiler, you need
in this simple function
bug.c:
void restore() {
register long reg_ebx asm (ebx);
reg_ebx = saved_ebx;
}
$ gcc -v
Utilisation des specs internes.
Target: i386-redhat-linux
Configuré avec: ../configure --prefix=/usr --mandir=/usr/share/man
--infodir=/usr/share/info
--- Comment #1 from rguenth at gcc dot gnu dot org 2008-09-22 11:27 ---
reg_ebx is obviously unused in your function so the assignment is DCEd. I
guess
what you try to do is not really possible with C.
--
rguenth at gcc dot gnu dot org changed:
What|Removed
--- Comment #9 from janus at gcc dot gnu dot org 2008-09-22 11:46 ---
Subject: Bug 37486
Author: janus
Date: Mon Sep 22 11:45:02 2008
New Revision: 140546
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=140546
Log:
2008-09-22 Janus Weil [EMAIL PROTECTED]
PR
--- Comment #2 from rguenth at gcc dot gnu dot org 2008-09-22 11:54 ---
Technically gcc.dg/tree-ssa/ssa-fre-10.c also fails (or rather it doesn't
properly test for what it is supposed to).
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37145
--- Comment #3 from rguenth at gcc dot gnu dot org 2008-09-22 11:54 ---
Subject: Bug 37145
Author: rguenth
Date: Mon Sep 22 11:53:29 2008
New Revision: 140547
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=140547
Log:
2008-09-22 Richard Guenther [EMAIL PROTECTED]
PR
--- Comment #10 from janus at gcc dot gnu dot org 2008-09-22 11:50 ---
Fixed with r140546. Closing.
--
janus at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #18 from rguenth at gcc dot gnu dot org 2008-09-22 11:48
---
Created an attachment (id=16377)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16377action=view)
patch
Actually this one vectorizes slp-perm-1.c for me on a ppc64 cross. It also
passes the x86_64 vect
--- Comment #5 from tim at klingt dot org 2008-09-22 12:11 ---
i have the same issue with both gcc-4.3 and gcc-4.4 ...
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34624
--- Comment #3 from rguenth at gcc dot gnu dot org 2008-09-22 12:20 ---
Current PRE simplifies only a subset of previous PRE - see
tree-ssa-pre.c:fully_constant_expression.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37542
--- Comment #4 from rguenth at gcc dot gnu dot org 2008-09-22 12:22 ---
All XFAILs left are due to the SCCVN union value-numbering optimization being
disabled:
#if FIXME
/* If this is a reference to a union member, record the union
member size as operand. Do so
80 matches
Mail list logo
