--- Additional Comments From mmitchel at gcc dot gnu dot org 2005-02-10
08:00 ---
John Spicer and I have had a long talk about this issue, and we're of differring
opinions. There's going to be a core issue about this, but until that's
resolved I don't think we know whether to call this
--
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |mark at codesourcery dot com
|dot org |
Status|NEW
--- Additional Comments From giovannibajo at libero dot it 2005-02-10
08:15 ---
Let's suspend it then. It's still questionable whether we should let GCC change
behaviour between 3.4 and 4.0 on this unresolved issue, but I guess that Mark
had fixed it back to the previous behaviour,
--- Additional Comments From Thomas dot Koenig at online dot de 2005-02-10
08:52 ---
(In reply to comment #2)
There are a gazillion places where we just check if (optimize) without
any specific flag. It would be quite a lot of work to introduce flags for all
of them, and I'm not sure
--- Additional Comments From steven at gcc dot gnu dot org 2005-02-10
09:08 ---
The slowdown is probably some unfortunate icache effect - ccould be anything
from alignment, the slightly larger instructions due to using r8 instead of
rcx. I guess we should not care too much about such
--- Additional Comments From aj at gcc dot gnu dot org 2005-02-10 09:22
---
With my patch, the results look good again (this is on x86-64 with multilibs):
=== libjava Summary for unix ===
# of expected passes3726
# of expected failures 14
#
--- Additional Comments From rguenth at gcc dot gnu dot org 2005-02-10
09:39 ---
It's CCP that for foo_void is able to propagate i[0] into the comparison here:
struct Foo * const this;
void * D.1798;
size_t D.1795;
void * __p;
int i[2];
struct Foo * i.6;
int D.1786;
--- Additional Comments From rguenth at gcc dot gnu dot org 2005-02-10
09:49 ---
The difference between foo_void and foo_void_offset is that for foo_void PRE
cannot see that (struct Foo *) i[0] is equivalent to (struct Foo *) i. As
such, for foo_void we end up with
bb 0:
__p_2 =
program wtest
implicit none
open(1,file='wtest.out')
write(1,'(1 2 3 4 5 6 7 8 9)')
close(1)
open(1,file='wtest.out')
write(1,'(9 8 7 6)')
close(1)
end
[EMAIL PROTECTED] gfortran]$ gfc a.f
[EMAIL PROTECTED] gfortran]$ ./a.out
[EMAIL PROTECTED]
--- Additional Comments From steven at gcc dot gnu dot org 2005-02-10
10:06 ---
'size' for susan_edged_mod_1 .o files
33 = pre 3.3.3-suse (hammer branch
40 = CVS head 20050209
patched = CVS head 20050209 with the 'TER hack' patch applied.
i686:
textdata bss dec
--- Additional Comments From krischik at users dot sourceforge dot net
2005-02-10 10:13 ---
Created an attachment (id=8160)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=8160action=view)
Source needed to recreate the bug.
After seperating the source with gnatchop use the following
--- Additional Comments From krischik at users dot sourceforge dot net
2005-02-10 10:17 ---
Hello Arno,
I have added the sources, as you requested. I hope I got the gnatchop stuff
right. I have tested the sources with the provided command and the error
persists. Version:
GNAT 4.0.0
--- Additional Comments From Thomas dot Koenig at online dot de 2005-02-10
10:17 ---
It appears the problem is caused by one of the
optimization options that cannot be controlled with
flags.
One suspect is this code snippet from gcc/config/ia64.c :
static bool
ia64_rtx_costs (rtx x,
--- Additional Comments From krischik at users dot sourceforge dot net
2005-02-10 10:20 ---
(From update of attachment 8160)
Mime type was not correctly determined.
--
What|Removed |Added
--- Additional Comments From rguenth at gcc dot gnu dot org 2005-02-10
10:20 ---
The C++ frontend doesn't give us the opportunity to canonicalize i to i[0] as
it doesn't call fold in typeck:build_address or decay_conversion.
I'm lost here.
--
--- Additional Comments From schwab at suse dot de 2005-02-10 10:33 ---
$ find . -name '*.c' | xargs grep '[(|!] *optimize[) =!|]' | wc -l
204
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19848
--- Additional Comments From steven at gcc dot gnu dot org 2005-02-10
10:37 ---
http://gcc.gnu.org/ml/gcc-patches/2005-02/msg00410.html is a start.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19309
--- Additional Comments From steven at gcc dot gnu dot org 2005-02-10
10:50 ---
The real problem here is that the tree alias analyses do not take full
advantage of 'restrict'. There are more PRs about this, and it is also *the*
major source of regressions in a well-known commercial
--- Additional Comments From steven at gcc dot gnu dot org 2005-02-10
10:55 ---
Add some dependencies to other restrict-related problem reports.
--
What|Removed |Added
--- Additional Comments From aph at gcc dot gnu dot org 2005-02-10 10:58
---
It looks like the patch was applied to the wrong place in the file: it certainly
was my intention to apply it to all Linux. And indeed, my testing was not on
m68k, but on x86-64.
The obvious question is
--- Additional Comments From steven at gcc dot gnu dot org 2005-02-10
10:58 ---
wrong-code, the worst kind we have...
--
What|Removed |Added
Severity|normal
--- Additional Comments From steven at gcc dot gnu dot org 2005-02-10
11:02 ---
In comment #3 Zdenek said Possibly even better would be to add generation of
autoincrements to loop optimizer, but this would require fixing cse so that it
handles them correctly. Zdenek, can you
--- Additional Comments From steven at gcc dot gnu dot org 2005-02-10
11:06 ---
Is this really a regression if, really, 3.3 was buggy?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19210
--- Additional Comments From rakdver at atrey dot karlin dot mff dot cuni
dot cz 2005-02-10 11:12 ---
Subject: Re: [4.0 Regression] Poor quality code after loop unrolling.
In comment #3 Zdenek said Possibly even better would be to add generation of
autoincrements to loop optimizer,
The current source (CVS or 3.1.1, same) of XaoS (http://xaos.sf.net), if you
compile it and run bin/xaos, gives a segmentation fault if you press the button
B twice (setting perturbation on/off). Now if I add the following line to
src/ui-hlp/menu.c in the end of uih_persw():
printf();
there will
--- Additional Comments From cvs-commit at gcc dot gnu dot org 2005-02-10
12:19 ---
Subject: Bug 19342
CVSROOT:/cvs/gcc
Module name:gcc
Changes by: [EMAIL PROTECTED] 2005-02-10 12:18:52
Modified files:
gcc: ChangeLog c-typeck.c
--- Additional Comments From jakub at gcc dot gnu dot org 2005-02-10 12:20
---
Fixed.
--
What|Removed |Added
Status|NEW |RESOLVED
Enhanced testcase for PR c/19342 still ICEs e.g. on x86_64 at -O2,
particularly in emit_move_insn.
expand_value_return is called with val = (reg:QI 64), but
return_reg has different mode, (reg:SI 58 [ result ]).
--
Summary: ICE in emit_move_insn with __attribute__((mode (QI)))
--- Additional Comments From jakub at gcc dot gnu dot org 2005-02-10 12:27
---
Created an attachment (id=8161)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=8161action=view)
pr19874-test.patch
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19874
--
What|Removed |Added
Status|UNCONFIRMED |NEW
Ever Confirmed||1
Last reconfirmed|-00-00 00:00:00
--- Additional Comments From rguenth at gcc dot gnu dot org 2005-02-10
12:46 ---
Patch at
http://gcc.gnu.org/ml/gcc-patches/2005-01/msg00656.html
pinged. Or WONTFIX - it's up to Mark.
--
What|Removed |Added
--- Additional Comments From steven at gcc dot gnu dot org 2005-02-10
12:48 ---
In reply to comment #13 - I have tested the patch on i686, amd64, ppc, and
ia64.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11707
--- Additional Comments From bonzini at gcc dot gnu dot org 2005-02-10
12:50 ---
No, it is not, still I would not close it as WONTFIX. I'd rather see it
suspended, and even better I'd like to see an approval for Zdenek's unsafe loop
optimization patch.
--
--- Additional Comments From bonzini at gcc dot gnu dot org 2005-02-10
12:51 ---
Oh, and VTOP notes were killed on mainline.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17860
--- Additional Comments From steven at gcc dot gnu dot org 2005-02-10
13:13 ---
.
--
What|Removed |Added
Status|NEW |RESOLVED
--- Additional Comments From cvs-commit at gcc dot gnu dot org 2005-02-10
13:14 ---
Subject: Bug 19309
CVSROOT:/cvs/gcc
Module name:gcc
Changes by: [EMAIL PROTECTED] 2005-02-10 13:14:03
Modified files:
gcc: ChangeLog
gcc/doc:
--- Additional Comments From falk at debian dot org 2005-02-10 13:16
---
(In reply to comment #0)
The current source (CVS or 3.1.1, same) of XaoS (http://xaos.sf.net), if you
compile it and run bin/xaos, gives a segmentation fault if you press the
button
B twice (setting
--- Additional Comments From rguenth at gcc dot gnu dot org 2005-02-10
13:18 ---
Actually, exchanging loc_test for
int loc_test(void)
{
const Loc2 k(0, 1);
return k[0].first();
}
shows the same problem:
bb 0:
D.2541 = (struct Loc1 *) k.D.2190.D.2155.domain_m.buffer;
--- Additional Comments From kovzol at math dot u-szeged dot hu 2005-02-10
13:21 ---
OK, I see that this is not really a good bug report, I know.
Now I tried the same compilation under gcc-2.95.3 (SuSE 8.0, x86) and the same
source works perfectly.
--
--- Additional Comments From cvs-commit at gcc dot gnu dot org 2005-02-10
13:53 ---
Subject: Bug 19386
CVSROOT:/cvs/gcc
Module name:gcc
Changes by: [EMAIL PROTECTED] 2005-02-10 13:53:24
Modified files:
gcc/ada: decl.c utils2.c utils.c
Log
--- Additional Comments From Ulrich dot Beingesser at t-systems dot com
2005-02-10 13:54 ---
(In reply to comment #1)
Works in 3.4.3:
http://gcc.gnu.org/ml/gcc-testresults/2004-11/msg00294.html
So something is wrong, but I don't know because there is not enough
information in this
--- Additional Comments From cvs-commit at gcc dot gnu dot org 2005-02-10
13:55 ---
Subject: Bug 16592
CVSROOT:/cvs/gcc
Module name:gcc
Changes by: [EMAIL PROTECTED] 2005-02-10 13:55:33
Modified files:
gcc/ada: Makefile.in
Added files:
--- Additional Comments From charlet at gcc dot gnu dot org 2005-02-10
14:01 ---
Should now be fixed.
Arno
--
What|Removed |Added
Status|UNCONFIRMED
--- Additional Comments From charlet at gcc dot gnu dot org 2005-02-10
14:02 ---
Fixed by forcing -static-libgcc when building the tools, as intended.
--
What|Removed |Added
--
Bug 17464 depends on bug 16592, which changed state.
Bug 16592 Summary: Ada tools don't use the newly built shared libraries
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16592
What|Old Value |New Value
--- Additional Comments From uweigand at gcc dot gnu dot org 2005-02-10
14:03 ---
Some more information about the problem.
The __builtin_memset call gets as V_MAY_DEF operands all global variables
(using the call_clobbered_vars mechanism). Initially, this does *not*
include
--- Additional Comments From peter at pogma dot com 2005-02-10 14:15
---
http://gcc.gnu.org/ml/gcc-patches/2005-02/msg00439.html
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18810
On OSF1, when compiling R-2.0.1 in directory:
R-2.0.1/src/modules/lapack
when running:
g77 -mieee -g -O2 -c dlapack0.f -o dlapack0.lo
Got the following error:
dlapack0.f: In subroutine `dlasda':
dlapack0.f:18513: Internal compiler error in reload, at reload1.c:1100
--
Summary:
--- Additional Comments From falk at debian dot org 2005-02-10 15:05
---
3.0 is no longer supported. Can you retry with a newer version, such as 3.3.4 or
3.4.3? If you can still reproduce it, please attach the preprocessed source as
generated when adding -save-temps.
--
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-10
15:25 ---
(In reply to comment #2)
OK, I see that this is not really a good bug report, I know.
Now I tried the same compilation under gcc-2.95.3 (SuSE 8.0, x86) and the
same
source works perfectly.
Could
--
What|Removed |Added
Severity|critical|normal
Component|fortran |middle-end
GCC target triplet|
--
What|Removed |Added
Target Milestone|--- |4.0.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19830
--
What|Removed |Added
Target Milestone|3.4.4 |4.0.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19309
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-10
15:40 ---
Ok, lets remove the target milestone and suspend this bug then.
--
What|Removed |Added
--- Additional Comments From belyshev at depni dot sinp dot msu dot ru
2005-02-10 15:55 ---
3.4.4 4.0.0 delta
hashes100.c:
-O0 3.663.81 4%
-O1 6.43 11.4778%
-O2
--- Additional Comments From reichelt at gcc dot gnu dot org 2005-02-10
16:14 ---
Here's a reduced testcase without sizeof:
=
templateint N struct A
{
static const int i = 8/N;
char c[i];
};
A0 a;
=
The error
--- Additional Comments From cvs-commit at gcc dot gnu dot org 2005-02-10
17:11 ---
Subject: Bug 19579
CVSROOT:/cvs/gcc
Module name:gcc
Branch: gcc-3_4-branch
Changes by: [EMAIL PROTECTED] 2005-02-10 17:11:13
Modified files:
gcc:
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-10
17:19 ---
Fixed also in 3.4.4.
--
What|Removed |Added
Known to work|3.3.3 4.0.0
--- Additional Comments From jakub at gcc dot gnu dot org 2005-02-10 17:21
---
This looks like fold_truthop bug, will look at it.
--
What|Removed |Added
--- Additional Comments From ebotcazou at gcc dot gnu dot org 2005-02-10
17:22 ---
This was confirmed by Devang at some point.
--
What|Removed |Added
--- Additional Comments From ayqazi at yahoo dot co dot uk 2005-02-10
17:23 ---
Once GCC 4.0 is out, I'll experiment with it and submit test cases etc. for it.
No use trying to fix an older release's optimisations IMHO.
Thanks anyway.
--
What|Removed
--- Additional Comments From ebotcazou at gcc dot gnu dot org 2005-02-10
17:34 ---
Can we close this PR, Brad?
--
What|Removed |Added
CC|
--- Additional Comments From ebotcazou at gcc dot gnu dot org 2005-02-10
17:42 ---
The C++ pragma init_priority is enabled for several sparc targets, but not for
the generic sparc-elf (--target=sparc-elf).
Confirmed, a fallout of the illegitimate dependency on the Solaris config
--
What|Removed |Added
CC||ebotcazou at gcc dot gnu dot
||org
--
What|Removed |Added
CC||ebotcazou at gcc dot gnu dot
||org
--- Additional Comments From law at redhat dot com 2005-02-10 18:01 ---
Subject: Re: [4.0 Regression] Poor quality
code after loop unrolling.
On Thu, 2005-02-10 at 12:12 +0100, Zdenek Dvorak wrote:
In comment #3 Zdenek said Possibly even better would be to add generation
Hi
I'm developing a linux distribution with friends, and we decided to use gcc 4.
A few days ago, KDE 3.3.92 was released.
But it doesn't compile :( Especially, with the kdepim package, g++ starts
eating all the memory and all the CPU on a file.
I can't submit this file
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-10
18:06 ---
Did you read http://gcc.gnu.org/bugs.html. We need the preprocessed source.
--
What|Removed |Added
--- Additional Comments From jakub at gcc dot gnu dot org 2005-02-10 18:07
---
Actually, I see there multiple problems elsewhere.
First is on
int i;
int foo (void)
{
return i ~(unsigned int)3;
}
First is that
if (change)
return fold (build2 (BIT_AND_EXPR, type,
--- Additional Comments From kenner at vlsi1 dot ultra dot nyu dot edu
2005-02-10 18:12 ---
Subject: Re: [4.0 Regression] Poor quality code after loop unrolling.
It's been about a decade since I looked at cse vs autoincrements, so
the details have faded from memory. [The
--- Additional Comments From pied at fnux dot org 2005-02-10 18:13 ---
Created an attachment (id=8162)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=8162action=view)
The prepocessed source
I'm sorry, really sorry !
I generated the file, but forget to send it :/
--
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-10
18:30 ---
Ok, I can confirm this with the following options on powerpc-darwin (which
means it is most likely a
tree optimizator problem but I could be wrong).
-fno-exceptions -fno-check-new -fno-common -O2
It
--- Additional Comments From lucier at math dot purdue dot edu 2005-02-10
18:34 ---
Subject: Re: Parallel bootstrap failure: No rule to make target `intl.h',
needed by `c-parse.o'.
Yes, close it; I think it is a generic parallel build problem when the
build file system is mounted
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-10
18:35 ---
PRE is eating up a huge amount of memory.
--
What|Removed |Added
CC|
--
What|Removed |Added
CC||ericw at evcohs dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19684
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-10
18:36 ---
Closing as requested by the submitter.
--
What|Removed |Added
Status|WAITING
--
What|Removed |Added
CC||ericw at evcohs dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19815
--- Additional Comments From jiri at gaisler dot com 2005-02-10 18:41
---
Subject: Re: C++ init_priority is not enabled for generic
sparc-elf target
It not fixed on 3.4 or the 4.0 as far as I can see.
Jiri.
ebotcazou at gcc dot gnu dot org wrote:
--- Additional Comments From
Sometimes when I reconfigure libgcj, configure decides that
mmap doesn't work on my machine. This breaks the .db feature.
I don't know exactly how to reproduce, but I have seen it more than
once.
--
Summary: sometimes reconfiguring leads to incorrect config.h
Product: gcc
--- Additional Comments From ebotcazou at gcc dot gnu dot org 2005-02-10
18:51 ---
It not fixed on 3.4 or the 4.0 as far as I can see.
What do you mean by as far as I can see? Did you really try with 4.0.0pre?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19715
--- Additional Comments From ericw at evcohs dot com 2005-02-10 18:59
---
The testcase compiles successfully with avr-gcc on 3.3.2, and 3.4.3, using
-mmcu=atmega128.
Could someone with sufficient permissions please set the Known To Work field.
Dieter, could you confirm which device
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-10
19:04 ---
Reverting the following patch fixes the bug:
2005-01-14 Steven Bosscher [EMAIL PROTECTED]
* tree-ssa-dce.c (visited_control_parents): New sbitmap to
replace BB_VISITED uses.
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-10
19:05 ---
Confirmed.
--
What|Removed |Added
Status|UNCONFIRMED |NEW
--- Additional Comments From dberlin at gcc dot gnu dot org 2005-02-10
19:06 ---
Pinski is correct (i know because i told him that was the cause :P)
It no longer converges (i suspect some strange bug in your logic).
Before it converged in 2 iterations.
I stopped it at 958 iterations
--- Additional Comments From rsandifo at gcc dot gnu dot org 2005-02-10
19:08 ---
Hope to look at this over the weekend.
--
What|Removed |Added
--- Additional Comments From steven at gcc dot gnu dot org 2005-02-10
19:15 ---
I'll look into it.
--
What|Removed |Added
CC|steven at gcc dot gnu dot
--- Additional Comments From jiri at gaisler dot com 2005-02-10 19:24
---
Subject: Re: C++ init_priority is not enabled for generic
sparc-elf target
It is not fixed. In gcc-4.0.0, the file that needs to be fixed is
gcc/config/sparc/sp-elf.h . It should contain the following:
#undef
--
What|Removed |Added
CC||pinskia at gcc dot gnu dot
||org
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-10
19:40 ---
Fixed, thanks zdenek.
--
What|Removed |Added
Status|NEW
--
Bug 18048 depends on bug 19701, which changed state.
Bug 19701 Summary: [4.0 regression] Way too many IVs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19701
What|Old Value |New Value
--- Additional Comments From ericw at evcohs dot com 2005-02-10 19:43
---
Dieter, could you please try this out with a more recent snapshot?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19636
ICE on mainline (20050210) when compiling the the attached file f1.c
Compiles fine on 3.4.
g++ -o f1.o f1.c
f1.c: In function 'void __static_initialization_and_destruction_0(int, int)':
f1.c:7: internal compiler error: in import_export_decl, at cp/decl2.c:1718
Please submit a full bug report
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-10
19:56 ---
Confirmed, here is the most reduced testcase:
struct S {
char k;
};
char const volatile S::* const p01 = S::k;
--
What|Removed |Added
--- Additional Comments From ebotcazou at gcc dot gnu dot org 2005-02-10
20:05 ---
It is not fixed.
That's wrong. It is fixed in 4.0.0pre:
gcc -E -DIN_GCC -DCROSS_COMPILE -W -Wall -Wwrite-strings -Wstrict-prototypes
-Wmissing-prototypes -fno-common -DHAVE_CONFIG_H-I. -Icp
--- Additional Comments From pluto at pld-linux dot org 2005-02-10 20:17
---
(In reply to comment #2)
As Andrew pointed out, the merge of the eor and the rotate is now done on
mainline in 4.0.
Hmm, it doesn't work on my gcc.
# arm-pld-linux-gcc reversing_the_bytes_in_word.c -s
--- Additional Comments From ebotcazou at gcc dot gnu dot org 2005-02-10
20:24 ---
And it is not present in all 3.x versions either, only in 3.4.x.
--
What|Removed |Added
--- Additional Comments From Thomas dot Koenig at online dot de 2005-02-10
20:31 ---
*** This bug has been marked as a duplicate of 19848 ***
--
What|Removed |Added
--- Additional Comments From Thomas dot Koenig at online dot de 2005-02-10
20:31 ---
*** Bug 19825 has been marked as a duplicate of this bug. ***
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19848
--
Bug 5900 depends on bug 19825, which changed state.
Bug 19825 Summary: -fno-loop-optimize2 does not work
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19825
What|Old Value |New Value
--- Additional Comments From Thomas dot Koenig at online dot de 2005-02-10
20:35 ---
(In reply to comment #4)
$ find . -name '*.c' | xargs grep '[(|!] *optimize[) =!|]' | wc -l
204
Any idea how I should go about further debugging PR 5900? There is a
wrong-code for ia-64 there,
1 - 100 of 190 matches
Mail list logo
