--- Comment #5 from jakub at gcc dot gnu dot org 2010-08-26 07:14 ---
It isn't in terrible shape on many targets, it is only on targets with very
small amount of registers where -fschedule-insns didn't in 4.4 and earlier
(since forever) work reliably and wasn't enabled by default either
--- Comment #2 from jakub at gcc dot gnu dot org 2010-08-26 07:33 ---
That's not the main problem here. The problem is that this is an aggregate and
we SRA only automatic vars, not global vars. And it isn't constant either
(neither declared, nor actually, so can't be detected as consta
Hello,
"-Wno-unused-variable" and friends such as "-Wno-unused-parameter" would be
welcome additions for folks who like to use "-Wall" and "-Wextra".
Reasons it would be a good idea to include the "-no" forms:
(1) its easy to remember
(2) its easy to use
(3) when using "-Wall" and "-Wextra", the
Command line:
$ g++ -O2 -fipa-cp-clone -ftracer testcase.C
Valgrind output:
$ valgrind -q --trace-children=yes
/mnt/svn/gcc-trunk/binary-163468-lto-fortran-checking-yes-rtl-df/bin/g++ -O2
-fipa-cp-clone -ftracer testcase.C
==31083== Invalid read of size 8
==31083==at 0x112BDF0: search_line_sse
--- Comment #1 from zsojka at seznam dot cz 2010-08-26 09:04 ---
Created an attachment (id=21569)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21569&action=view)
reduced testcase
$ g++ -O2 -fipa-cp-clone -ftracer pr45412.C
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45412
--- Comment #1 from noloader at gmail dot com 2010-08-26 09:09 ---
Though (5) and (8) are specific to me, there's nothing really special about me
- there's probably lots of lazy programmers who would also like the warning
options.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45411
--- Comment #2 from jakub at gcc dot gnu dot org 2010-08-26 09:23 ---
There of course are
-Wno-unused-{function,label,parameter,value,variable,but-set-{parameter,variable}}
options.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45411
--- Comment #2 from rguenth at gcc dot gnu dot org 2010-08-26 09:42 ---
Confirmed.
The tracer seriously messes up virtual operands. I'm not sure the copy
tables work like the author thought they would.
After duplicating we have
virtual int S::vm() (struct S * const this)
{
int stat
--- Comment #3 from rguenth at gcc dot gnu dot org 2010-08-26 09:46 ---
Well, you need flow-sensitive analysis to figure out that when visiting
ss.j += 88; you can assume that ss.j is still 2.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45410
--- Comment #4 from jakub at gcc dot gnu dot org 2010-08-26 09:59 ---
Sure, one that would also be able to assume that main is called just once.
Anyway, even to optimize ss.i into 77 you'd need a way to mark fields in global
aggregates as const if they aren't ever modified...
--
htt
--- Comment #3 from redi at gcc dot gnu dot org 2010-08-26 10:14 ---
You didn't say which version of GCC you're using, but it doesn't really matter
because these options have been in place for many years.
(In reply to comment #0)
> (5) I'm lazy and don't want to locate the applicable ma
--- Comment #17 from howarth at nitro dot med dot uc dot edu 2010-08-26
10:18 ---
Testresults with proposed patch at
http://gcc.gnu.org/ml/gcc-testresults/2010-08/msg02586.html.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44812
The scan-assembler-not "12345" test in gcc.target/i386/combine-mul.c fails on
x86_64-apple-darwin10 due to that target defaulting to -fPIC. Adding -fno-PIC
eliminates the failure. This testcase needs...
Index: gcc.target/i386/combine-mul.c
==
Hi,
I agree that "const void *" is the better parameter for _mm_prefetch, but it
leads to incompatible code. ICC defines _mm_prefetch as:
void _mm_prefetch(char const *, int)
and MSVC documents it as:
void _mm_prefetch(char * p , int i )
(which is even worse). I would expect the GCC header to us
--- Comment #1 from bernds at gcc dot gnu dot org 2010-08-26 10:57 ---
Ok, I could apply that, but why is it failing? What assembly output is being
produced for it?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45413
--- Comment #2 from howarth at nitro dot med dot uc dot edu 2010-08-26
11:13 ---
Created an attachment (id=21570)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21570&action=view)
assembly file for gcc.target/i386/combine-mul.c on x86_64-apple-darwin10
--
http://gcc.gnu.org/bu
gcc.c-torture/execute/960416-1.c fails this way with -O1
-fno-tree-dominator-opts, 4.5 is fine.
> /abuild/rguenther/trunk-g/gcc/xgcc -B/abuild/rguenther/trunk-g/gcc/
> /space/rguenther/src/svn/trunk/gcc/testsuite/gcc.c-torture/execute/960416-1.c
> -w -O1 -lm -fno-tree-dominator-opts -o
> /abuil
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
CC||matz at gcc dot gnu dot org
Known to work|
This is a performance/size regression between 4.6 (4.5) and 4.4.
The C code:
int foo(long long a)
{
if (a & (long long) 0x400)
return 1;
return 0;
}
Assemble code generated by 4.6 trunk:
foo:
.LFB0:
.cfi_startproc
pushq %rbp
.cfi_def_cfa_offset 16
mo
--- Comment #1 from rguenth at gcc dot gnu dot org 2010-08-26 12:43 ---
It works for me.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45416
--- Comment #2 from bmei at broadcom dot com 2010-08-26 12:47 ---
Sorry, I first observed this on our target. Then I tried to reproduce on x86,
but I forgot to turn on optimization flags. It does work for x86. Please delete
this report. I will figure out what happen with my target.
-
--- Comment #3 from bmei at broadcom dot com 2010-08-26 12:55 ---
I found I can reproduce the bug with ARM
ARM trunk -Os:
foo:
@ args = 0, pretend = 0, frame = 0
@ frame_needed = 0, uses_anonymous_args = 0
@ link register save eliminated.
mov r2, #102
--
fxcoudert at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|fxcoudert at gcc dot gnu dot|unassigned at gcc dot gnu
|org
See thread starting at
http://j3-fortran.org/pipermail/j3/2010-August/003747.html
The first example is valid (state == 0), the second valid (state /= 0), and the
third one is invalid.
gfortran produces, http://j3-fortran.org/pipermail/j3/2010-August/003753.html,
0 .
Expected: The second example
Hello,
this may just not be implemented yet, but in case it was missed:
struct A1 {
};
struct A2 {
A2(){}
};
template struct B {
T tab[1];
B(T const& t):tab({t}) {} // here
};
int main(){
int i=2;
B b{i};
A1 a1;
B e1{a1};
A2 a2;
--- Comment #4 from mikpe at it dot uu dot se 2010-08-26 14:01 ---
(In reply to comment #3)
> I found I can reproduce the bug with ARM
I see this too on armv5tel-linux-gnueabi. gcc-4.4.4 -Os generates 3
instructions for the body of foo(), 4.5.1 and 4.6.0 generate 8 instructions.
--
--- Comment #3 from hjl dot tools at gmail dot com 2010-08-26 14:16 ---
It is caused by revision 162842:
http://gcc.gnu.org/ml/gcc-cvs/2010-08/msg00053.html
--
hjl dot tools at gmail dot com changed:
What|Removed |Added
---
--- Comment #1 from hjl dot tools at gmail dot com 2010-08-26 14:19 ---
It is caused by revision 162413:
http://gcc.gnu.org/ml/gcc-cvs/2010-07/msg00767.html
--
hjl dot tools at gmail dot com changed:
What|Removed |Added
---
--- Comment #8 from rguenth at gcc dot gnu dot org 2010-08-26 14:21 ---
Fixed.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #9 from rguenth at gcc dot gnu dot org 2010-08-26 14:21 ---
Subject: Bug 45255
Author: rguenth
Date: Thu Aug 26 14:20:45 2010
New Revision: 163565
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=163565
Log:
2010-08-26 Richard Guenther
PR tree-optimization/
--- Comment #2 from rguenth at gcc dot gnu dot org 2010-08-26 14:21 ---
I'd say it merely exposed it.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45415
--- Comment #4 from noloader at gmail dot com 2010-08-26 14:54 ---
> You didn't say which version of GCC you're using, but it doesn't really matter
> because these options have been in place for many years.
Moot point indeed! But for completeness
$ gcc --version
gcc (Ubuntu 4.4.3-4u
--- Comment #5 from redi at gcc dot gnu dot org 2010-08-26 15:09 ---
(In reply to comment #4)
>
> > (In reply to comment #0)
> >> (5) I'm lazy and don't want to locate the applicable man page
>
> > Here it is:
> > http://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html
> Interesting. I
--- Comment #18 from howarth at nitro dot med dot uc dot edu 2010-08-26
16:01 ---
Jan's alternative patch...
Index: lto-cgraph.c
===
--- lto-cgraph.c(revision 163494)
+++ lto-cgraph.c(working copy)
@@ -1109
--- Comment #2 from rth at gcc dot gnu dot org 2010-08-26 16:27 ---
Created an attachment (id=21571)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21571&action=view)
Use
Rather than use gcc builtins directly, use the standard header file.
Hopefully this will work with Apple's GCC
--- Comment #3 from rth at gcc dot gnu dot org 2010-08-26 16:27 ---
The patch bootstraps on ppc64-linux; please test on Darwin.
--
rth at gcc dot gnu dot org changed:
What|Removed |Added
-
--- Comment #8 from jakub at gcc dot gnu dot org 2010-08-26 16:39 ---
Subject: Bug 44485
Author: jakub
Date: Thu Aug 26 16:39:26 2010
New Revision: 163568
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=163568
Log:
PR tree-optimization/44485
* calls.c (flags_from_
--- Comment #1 from rwild at gcc dot gnu dot org 2010-08-26 16:46 ---
Please show
grep '^SHELL =' alphaev67-dec-osf5.1/libjava/libltdl/Makefile
and if you want Java enabled, please set and export CONFIG_SHELL to some bash
instance before running $CONFIG_SHELL path/to/toplevel/configur
--- Comment #9 from jakub at gcc dot gnu dot org 2010-08-26 16:48 ---
Should be fixed now.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
Status
--- Comment #4 from dominiq at lps dot ens dot fr 2010-08-26 16:57 ---
Following a discussion on IRC I have tested the following patch that does not
work with "gcc version 4.0.1 (Apple Inc. build 5493)" and without 'GCC_VERSION
>= 4005'.
--- /opt/gcc/gcc-4.6-work/libcpp/lex.c 2010-08-
Command line:
$ gcc -fcompare-debug -O3 testcase-min6.i
Compiler output:
$ gcc -fcompare-debug -O3 testcase-min6.i -w
gcc: error: testcase-min6.i: -fcompare-debug failure (length)
Tested revisions:
r163468 - fail
r162940 - OK
r161659 - OK
r161170 - OK
The failure keeps appearing and disappearing
--- Comment #1 from zsojka at seznam dot cz 2010-08-26 17:12 ---
Created an attachment (id=21572)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21572&action=view)
testcase
Delta (with line granularity) fails to reduce the file further. The difference
in debug and non-debug runs is
Reported by Salvatore at http://gcc.gnu.org/ml/fortran/2010-08/msg00351.html.
Here is a slightly simplified version of the test case:
module base_mat_mod
type :: base_sparse_mat
contains
procedure, pass(a) :: get_fmt => base_get_fmt
end type base_sparse_mat
contains
function base_get_
--- Comment #2 from hjl dot tools at gmail dot com 2010-08-26 17:46 ---
I can't reproduce it on Fedora 13. But valgrind reports:
Compiler executable checksum: d32748fe5bfee4f2f9ecf4e95f2e1498
==17242== Conditional jump or move depends on uninitialised value(s)
==17242==at 0x7ED1AF:
--- Comment #1 from janus at gcc dot gnu dot org 2010-08-26 17:54 ---
The dump shows that the first call to 'get_fmt' is executed dynamically as
'a.$vptr->get_fmt(...)', while the ones inside the SELECT TYPE block are
resolved statically to 'd_base_get_fmt'. For the TYPE IS clause this i
--- Comment #3 from jakub at gcc dot gnu dot org 2010-08-26 17:58 ---
Unfortunately can't reproduce it myself, neither with bootstrapped, nor with
just stage1 built gcc.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45419
Between 20100709 and 20100823, IRIX6.5 bootstrap with Ada included broke:
$ /vol/gcc/obj/regression/trunk/6.5-gcc/build/./prev-gcc/xgcc
-B/vol/gcc/obj/regression/trunk/6.5-gcc/build/./prev-gcc/
-B/vol/gcc/mips-sgi-irix6.5/bin/ -B/vol/gcc/mips-sgi-irix6.5/bin/
-B/vol/gcc/mips-sgi-irix6.5/lib/ -isys
compiling the to-be-attached testcase as:
gfortran -c -ftime-report -cpp -fbounds-check -g -O3 -ffast-math -funroll-loops
-ftree-vectorize -march=native -ffree-form test.f90
massively slowed down in 4.6
4.4: 51s
4.5: 50s
4.6: 434s
the timing report look like:
Execution times (seconds)
garba
--- Comment #1 from jv244 at cam dot ac dot uk 2010-08-26 18:34 ---
Created an attachment (id=21573)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21573&action=view)
testcase
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45422
--- Comment #4 from zsojka at seznam dot cz 2010-08-26 18:37 ---
Created an attachment (id=21574)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21574&action=view)
different testcase
Thank you for answers! This not yet reduced testcase fails in a similiar way
(r163468, x86_64-linux
bool b;
int i;
int
foo (void)
{
#pragma omp atomic
b ^= 1;
#pragma omp barrier
#pragma omp atomic
b |= 1;
#pragma omp barrier
#pragma omp atomic
b &= 1;
#pragma omp barrier
#pragma omp atomic
b -= 1;
#pragma omp barrier
#pragma omp atomic
i ^= 1;
#pragma omp
--
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 #5 from hjl dot tools at gmail dot com 2010-08-26 19:48 ---
valgrind issue is caused by revision 162156:
http://gcc.gnu.org/ml/gcc-cvs/2010-07/msg00510.html
--
hjl dot tools at gmail dot com changed:
What|Removed |Added
---
--- Comment #13 from domob at gcc dot gnu dot org 2010-08-26 19:49 ---
Subject: Bug 38936
Author: domob
Date: Thu Aug 26 19:48:43 2010
New Revision: 163572
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=163572
Log:
2010-08-26 Daniel Kraft
PR fortran/38936
PR
--- Comment #9 from domob at gcc dot gnu dot org 2010-08-26 19:49 ---
Subject: Bug 45384
Author: domob
Date: Thu Aug 26 19:48:43 2010
New Revision: 163572
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=163572
Log:
2010-08-26 Daniel Kraft
PR fortran/38936
PR f
--- Comment #2 from domob at gcc dot gnu dot org 2010-08-26 19:49 ---
Subject: Bug 44047
Author: domob
Date: Thu Aug 26 19:48:43 2010
New Revision: 163572
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=163572
Log:
2010-08-26 Daniel Kraft
PR fortran/38936
PR f
--- Comment #3 from domob at gcc dot gnu dot org 2010-08-26 19:51 ---
Fixed.
--
domob at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #10 from domob at gcc dot gnu dot org 2010-08-26 19:52 ---
Fixed.
--
domob at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #14 from domob at gcc dot gnu dot org 2010-08-26 19:54 ---
This fixed association to CLASS entities and also reworked SELECT TYPE to use
the same workings internally. Still open: Association to strings and
derived-types, and the extended checks for illegal usage of names as
--- Comment #2 from janus at gcc dot gnu dot org 2010-08-26 19:55 ---
It turns out this bug is rather easy to fix. The problem was the we used the
temporary needed for the TYPE IS clause also in the CLASS DEFAULT clause (where
we need none). The following patch fixes it (haven't checked
--- Comment #6 from hjl dot tools at gmail dot com 2010-08-26 20:02 ---
(In reply to comment #4)
> Created an attachment (id=21574)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21574&action=view) [edit]
> different testcase
>
> Thank you for answers! This not yet reduced testcase
--- Comment #19 from dominiq at lps dot ens dot fr 2010-08-26 20:03 ---
With the patch in comment #18, with -m64 I see the following failure
FAIL: gcc.dg/lto/ipareference2 c_lto_ipareference2_0.o-c_lto_ipareference2_1.o
execute -O1 -fwhopr -fwhole-program
which was only present with -
--
janus at gcc dot gnu dot org changed:
What|Removed |Added
Summary|[OOP] poylmorphic TBP calls |[OOP] polymorphic TBP call
|in a CLASS DEFAULT clause
--- Comment #5 from mikpe at it dot uu dot se 2010-08-26 21:13 ---
The code size regression on ARM is caused by r146817, Matz' expand from SSA
patch: http://gcc.gnu.org/ml/gcc-cvs/2009-04/msg01459.html
Here's the diff in the assembly code generated by a cross to
armv5tel-linux-gnueabi,
--- Comment #20 from dominiq at lps dot ens dot fr 2010-08-27 00:09 ---
The patch in comment #18 works also on powerpc-apple-darwin9:
Running target unix/-m32
Using /sw/share/dejagnu/baseboards/unix.exp as board description file for
target.
Using /sw/share/dejagnu/config/unix.exp as gen
--- Comment #21 from howarth at nitro dot med dot uc dot edu 2010-08-27
00:42 ---
Created an attachment (id=21575)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21575&action=view)
patch which completely eliminates -m32 darwin lto failures
--
http://gcc.gnu.org/bugzilla/show_b
--- Comment #22 from howarth at nitro dot med dot uc dot edu 2010-08-27
00:44 ---
The m32_lto_fix2.diff patch expands the usage of DECL_WEAK into
gcc/lto-cgraph.c and is sufficient to eliminate all of the darwin-specific lto
failures at -m32. The only remaining ones are...
FAIL: g++.d
--- Comment #10 from rwild at gcc dot gnu dot org 2010-08-27 04:27 ---
Subject: Bug 40868
Author: rwild
Date: Fri Aug 27 04:27:40 2010
New Revision: 163580
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=163580
Log:
Fix building ecjx in cross compile mode.
libjava/:
PR l
--- Comment #11 from rwild at gcc dot gnu dot org 2010-08-27 04:30 ---
Fixed.
--
rwild at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #8 from jvdelisle at gcc dot gnu dot org 2010-08-27 06:50
---
Subject: Bug 43217
Author: jvdelisle
Date: Fri Aug 27 06:50:03 2010
New Revision: 163581
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=163581
Log:
2010-08-27 Jerry DeLisle
PR fortran/43217
70 matches
Mail list logo
