--- Comment #4 from uros at gcc dot gnu dot org 2007-06-21 07:06 ---
Subject: Bug 32389
Author: uros
Date: Thu Jun 21 07:06:29 2007
New Revision: 125911
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=125911
Log:
PR target/32389
* config/i386/i386.h (enum ix86_sta
--- Comment #5 from uros at gcc dot gnu dot org 2007-06-21 07:13 ---
Subject: Bug 32389
Author: uros
Date: Thu Jun 21 07:13:16 2007
New Revision: 125912
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=125912
Log:
PR target/32389
* config/i386/i386.h (enum ix86_sta
--- Comment #6 from ubizjak at gmail dot com 2007-06-21 07:14 ---
Fixed on branches.
--
ubizjak at gmail dot com changed:
What|Removed |Added
Status|ASSIGNED
Hi,
I have successfully built SH4-Linux toolchain based on
(binutils-2.17, gcc-4.2-20061205, glibc-2.5) for Renesas SH target.
I am facing problems while compiling the following testcase-
--
TESTCASE
#include
#include
extern int s
--- Comment #20 from rguenther at suse dot de 2007-06-21 07:59 ---
Subject: Re: ICE: fold check: original tree changed
by fold with --enable-checking=fold
On Thu, 20 Jun 2007, spop at gcc dot gnu dot org wrote:
>
>
> --- Comment #17 from spop at gcc dot gnu dot org 2007-06-20
--- Comment #21 from rguenther at suse dot de 2007-06-21 08:01 ---
Subject: Re: ICE: fold check: original tree changed
by fold with --enable-checking=fold
On Thu, 20 Jun 2007, spop at gcc dot gnu dot org wrote:
>
>
> --- Comment #18 from spop at gcc dot gnu dot org 2007-06-20
--- Comment #1 from malitzke at metronets dot com 2007-06-21 08:08 ---
Disclosure:
I am not an IBM hater and never was. My first significant program ran on an IBM
1410 with a Tape Operating System (TOS) on the day President Kennedy was
assassinated. I then used an IBM 1620 II extensivel
--- Comment #22 from richard dot guenther at gmail dot com 2007-06-21
08:16 ---
Subject: Re: ICE: fold check: original tree changed by fold with
--enable-checking=fold
On 6/21/07, Sebastian Pop <[EMAIL PROTECTED]> wrote:
> Hi, here is the final fix for the remaining cases, closing thi
--- Comment #23 from spop at gcc dot gnu dot org 2007-06-21 08:21 ---
Subject: Re: ICE: fold check: original tree changed by fold with
--enable-checking=fold
On 6/21/07, Richard Guenther <[EMAIL PROTECTED]> wrote:
>
> The fold_binary change looks unnecessary.
ok.
> The rest is ok.
>
--- Comment #1 from schwab at suse dot de 2007-06-21 08:45 ---
may define strcmp as a macro. This has nothing to do with the
compiler.
--
schwab at suse dot de changed:
What|Removed |Added
-
--- Comment #24 from rguenther at suse dot de 2007-06-21 08:48 ---
Subject: Re: ICE: fold check: original tree changed
by fold with --enable-checking=fold
On Thu, 21 Jun 2007, spop at gcc dot gnu dot org wrote:
>
>
> --- Comment #23 from spop at gcc dot gnu dot org 2007-06-21
--- Comment #3 from rob1weld at aol dot com 2007-06-21 08:49 ---
GCC printed no warning about disliking a conversion.
I hacked the program in accordance with your suggestion and now it prints:
n= 4 QUESTION1=+0.123 QUESTION2=+0.123 Q(n)=+0.000
n= 5 QUESTION1=+0.099 QUESTION3=+0.0
--- Comment #20 from tbm at cyrius dot com 2007-06-21 08:53 ---
It's interesting that you mention problems with va_list and prologues
because I have such problems on IA64 too since the df merge. I've no
idea if they're related but maybe you want to check them out:
Error: .prologue with
--- Comment #3 from pinskia at gcc dot gnu dot org 2007-06-21 08:58 ---
Patch which fixes this bug:
Index: builtins.c
===
--- builtins.c (revision 125776)
+++ builtins.c (working copy)
@@ -4649,7 +4649,18 @@
std_expand_bu
--- Comment #4 from pinskia at gcc dot gnu dot org 2007-06-21 09:01 ---
First temp is SRA'd so the debugging info should be correct unless tracking is
wrong. Second iv-opts is messing up the debuging info for i by creating a new
variable.
I think this is a feature we would like to supp
CP2K (see PR 29975) compiles fine with current trunk using
'-O3 -ffast-math -ftree-vectorize -march=native -funroll-loops'
but the code stops with a bogus error message using
'-O3 -ffast-math -ftree-vectorize -march=native -funroll-loops -pg'
to reproduce see comment 112 in PR 29975 to obtain
--- Comment #5 from pinskia at gcc dot gnu dot org 2007-06-21 09:03 ---
Actually SRA should be ok, because we do keep tracking them on the tree level.
Daniel Berlin added this code.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32444
--- Comment #115 from jv244 at cam dot ac dot uk 2007-06-21 09:05 ---
trying to investigate the culprit of the slowdown mentioned in comment 112 I
found out that adding -pg to the compile flags leads to a miscompiled code.
I've filed PR 32450 to track the issue
--
http://gcc.gnu.org
--- Comment #4 from pinskia at gcc dot gnu dot org 2007-06-21 09:06 ---
abs converts the float/double to an integer type so this is not a bug. If you
use 4.3, you can use -Wconversion.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32448
--- Comment #1 from pinskia at gcc dot gnu dot org 2007-06-21 09:07 ---
Most likely -pg is changing the alignment of the stack which is incorrect. Oh
-pg code is emitted by the target specific code so this is a target issue.
--
pinskia at gcc dot gnu dot org changed:
Wha
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-06-21 09:11 ---
You are misrepresenting the problem again. So the issue is that the size of
cc1/cc1plus is increased with the decimal float back-end support. Now
Disabling this support in general is a hard problem. I would dare y
--- Comment #22 from sunjoong at gmail dot com 2007-06-21 09:14 ---
I checked VOLATILE statement passing in gfortran 4.3 .
Unfortunately I had failed to buid gfortran (and gcc) 4.3 in my i686,
I only checked that passing with ia64 binary on another ia64 machine.
(There is no gfortran 4
--- Comment #8 from pinskia at gcc dot gnu dot org 2007-06-21 09:15 ---
But again, even if sqrtl is not part of the library, this is not a GCC issue.
We provide builtins for faster usage and not to get around libc not including
the function.
--
pinskia at gcc dot gnu dot org changed
--- Comment #1 from daney at gcc dot gnu dot org 2007-06-21 09:15 ---
Created an attachment (id=13754)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13754&action=view)
Proposed patch.
This is a problem with __builtin_eh_return().
At -O0 the new return address is being stored in t
--- Comment #2 from rsandifo at gcc dot gnu dot org 2007-06-21 09:30
---
Does the pattern get deleted, or moved somewhere inappropriate?
If the former, does that mean that (unspec)s aren't allowed
in top-level patterns any more? Other backends besides MIPS
seem to use this construct.
./cc1plus -quiet -O3 1.2.ii -fdump-tree-all-details
1.2.ii: In function 'void test1()':
1.2.ii:104: error: true/false edge after a non-COND_EXPR in bb 15
1.2.ii:104: internal compiler error: verify_flow_info failed
Please submit a full bug report,
with preprocessed source if appropriate.
See http:/
--- Comment #1 from rguenth at gcc dot gnu dot org 2007-06-21 09:48 ---
Created an attachment (id=13755)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13755&action=view)
testcase
testcase reduced from libstdc++-v3/testsuite/25_algorithms/equal/1.cc
--
http://gcc.gnu.org/bugzi
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|--- |4.3.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32451
--- Comment #25 from richard dot guenther at gmail dot com 2007-06-21
10:07 ---
Subject: Re: ICE: fold check: original tree changed by fold with
--enable-checking=fold
On 6/21/07, Sebastian Pop <[EMAIL PROTECTED]> wrote
> On 6/21/07, Richard Guenther <[EMAIL PROTECTED]> wrote:
> >
> >
--- Comment #43 from ebotcazou at gcc dot gnu dot org 2007-06-21 10:11
---
Subject: Bug 25737
Author: ebotcazou
Date: Thu Jun 21 10:11:19 2007
New Revision: 125915
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=125915
Log:
PR tree-optimization/25737
* misc.c (gn
--- Comment #44 from ebotcazou at gcc dot gnu dot org 2007-06-21 10:13
---
Works on mainline.
--
ebotcazou at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #2 from jv244 at cam dot ac dot uk 2007-06-21 10:16 ---
(In reply to comment #1)
> Most likely -pg is changing the alignment of the stack which is incorrect. Oh
> -pg code is emitted by the target specific code so this is a target issue.
>
Is there an easy way for me to ch
--- Comment #2 from rguenth at gcc dot gnu dot org 2007-06-21 10:19 ---
Can be triggered with -O2 as well, with --param max-aliased-vops=1000
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32451
--- Comment #3 from rguenth at gcc dot gnu dot org 2007-06-21 10:26 ---
Zdenek, I suppose this is caused by your loop-preserving for DOM.
Note this may be jump threading related:
Threaded jump 9 --> 15 to 27
(gdb) call debug_bb_n (15)
;; basic block 15, loop depth 1, count 0
;; prev
--- Comment #4 from rguenth at gcc dot gnu dot org 2007-06-21 10:38 ---
The problem is introduced by tree-ssa-threadupdate.c:thread_single_edge where
we have the single_pred_p case but fail to clear the edge flags as we do
for example in DOMs propagate_rhs_into_lhs. So the following see
--- Comment #5 from rguenth at gcc dot gnu dot org 2007-06-21 10:42 ---
Looks like so, the other caller in tree-ssa-threadupdate.c does so.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32451
--- Comment #6 from rguenth at gcc dot gnu dot org 2007-06-21 10:44 ---
Mine.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|unassigned
--- Comment #1 from rguenth at gcc dot gnu dot org 2007-06-21 10:57 ---
I bet this is the same as PR32421.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #1 from rguenth at gcc dot gnu dot org 2007-06-21 10:58 ---
Can you run the compile inside gdb and check periodically where it wastes its
time?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32439
--- Comment #4 from rguenth at gcc dot gnu dot org 2007-06-21 11:00 ---
You shouldn't introduce calls to langhooks. Why not use mode_for_size?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32441
(Debian bug #429934, forwarding upstream)
Package: gnat-4.1
Version: 4.1.1-22
Severity: normal
Tags: confirmed, upstream
In the test case below, GCC emits correct type information for
Debugging.A but the type information for External.B is incorrect.
package External is
type External_Type is a
--- Comment #8 from rguenth at gcc dot gnu dot org 2007-06-21 11:19 ---
Ping?
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
CC|
--- Comment #1 from ludovic at ludovic-brenta dot org 2007-06-21 11:24
---
Forgot to add the output of gcc -v:
$ gcc-4.1 -v
Using built-in specs.
Target: x86_64-linux-gnu
Configured with: ../src/configure -v
--enable-languages=c,c++,fortran,objc,obj-c++,treelang --prefix=/usr
--enable-
#0 fancy_abort (file=0xe96e48 "/space/rguenther/src/svn/trunk/gcc/tree.c",
line=3074, function=0xe9825f "build2_stat")
at /space/rguenther/src/svn/trunk/gcc/diagnostic.c:656
#1 0x00a3bb1a in build2_stat (code=PLUS_EXPR, tt=0x2ad09f6def00,
arg0=0x2ad09fb625c0, arg1=0x2ad09fb1
--- Comment #1 from rguenth at gcc dot gnu dot org 2007-06-21 11:26 ---
Created an attachment (id=13756)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13756&action=view)
testcase
testcase from glibc, build with -O2
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32453
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
CC||pinskia at gcc dot gnu dot
|
--- Comment #5 from ubizjak at gmail dot com 2007-06-21 11:27 ---
(In reply to comment #3)
> GCC printed no warning about disliking a conversion.
It just happens that gcc is not like microwave oven that has to include
warnings about not drying animals in it.
> Sometimes the answer is _
--- Comment #6 from rob1weld at aol dot com 2007-06-21 11:30 ---
Thanks for everyones input.
The only issues related to this 'bug' are:
1): printf _sometimes_ prints "-0.000" and sometimes prints "+0.000" - the
reason it is even showing the "+" or "-" is because I enabled them using "
--- Comment #2 from rguenth at gcc dot gnu dot org 2007-06-21 11:38 ---
I have a patch.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|
L PROTECTED]>
PR tree-optimization/32451
* tree-ssa-threadupdate.c (thread_single_edge): Fixup edge flags.
* g++.dg/torture/20070621-1.C: New testcase.
Added:
trunk/gcc/testsuite/g++.dg/torture/20070621-1.C
Modified:
trunk/gcc/ChangeLog
trunk/gcc/testsuite/ChangeLog
--- Comment #2 from jakub at gcc dot gnu dot org 2007-06-21 12:11 ---
Subject: Bug 32362
Author: jakub
Date: Thu Jun 21 12:11:00 2007
New Revision: 125917
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=125917
Log:
PR middle-end/32362
* omp-low.c (lookup_decl_in_o
--- Comment #3 from jakub at gcc dot gnu dot org 2007-06-21 12:16 ---
Subject: Bug 32362
Author: jakub
Date: Thu Jun 21 12:15:53 2007
New Revision: 125918
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=125918
Log:
PR middle-end/32362
* omp-low.c (lookup_decl_in_o
--- Comment #3 from jakub at gcc dot gnu dot org 2007-06-21 12:21 ---
Subject: Bug 31866
Author: jakub
Date: Thu Jun 21 12:20:42 2007
New Revision: 125919
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=125919
Log:
PR tree-optimization/31866
* tree-ssa-coalesce.c
--- Comment #4 from jakub at gcc dot gnu dot org 2007-06-21 12:23 ---
Fixed.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #4 from jakub at gcc dot gnu dot org 2007-06-21 12:24 ---
Fixed.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #2 from zadeck at naturalbridge dot com 2007-06-21 12:29
---
Subject: Re: ICE in df_refs_verify, at df-scan.c:4066
rask at sygehus dot dk wrote:
> --- Comment #1 from rask at sygehus dot dk 2007-06-20 16:58 ---
> Created an attachment (id=13747)
--> (http://gcc.g
Should give with -fbounds-check something like the following (NAG f95 -C=all):
Rank 1 of array operand has extent 8 instead of 4
Program terminated by fatal error
In BOUNDSERROR, line 7 of test.f90
I think it might a duplicate of some PR, though I couldn't find it.
program boundsError
implicit
--- Comment #1 from burnus at gcc dot gnu dot org 2007-06-21 12:33 ---
I forgot to mention: I think this file is valid Fortran 2003 and only invalid
Fortran 95. Maybe using:
integer, dimension(4) :: y
is a better test case.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32454
--- Comment #5 from zadeck at naturalbridge dot com 2007-06-21 12:49
---
this was fixed with the commit.
--
zadeck at naturalbridge dot com changed:
What|Removed |Added
--- Comment #14 from dir at lanl dot gov 2007-06-21 12:57 ---
>What is actually the expected result? Depending on the compiler and compiler
>setting, I get completely different results for the second triangular matrix.
>(The first matrix remains always the same.)
What the program does i
--- Comment #7 from bardeau at iram dot fr 2007-06-21 13:08 ---
(In reply to comment #6)
> Fixed on mainline and 4.2. Unless you really want to backport it to 4.1.3, I'm
> closing this bug.
>
The bug still appears under cygwin with gcc 4.3:
program test
integer*4 a
a=1
print
--- Comment #15 from dir at lanl dot gov 2007-06-21 13:23 ---
>BTW I do not see (beside obfuscation) the interest of the constructs:
It is the construct:
jt=t(j2)
tt=tt+tt
t(j2)=jt
that is being optmized away or done incorrectly when the second matrix stays
the same
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-06-21 13:32 ---
*** Bug 32435 has been marked as a duplicate of this bug. ***
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
---
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-06-21 13:32 ---
Yes this is the same issue, we have POINTER_PLUS_EXPR and we are trying to
create it with a vector type.
*** This bug has been marked as a duplicate of 32421 ***
--
pinskia at gcc dot gnu dot org changed:
--- Comment #1 from zadeck at naturalbridge dot com 2007-06-21 13:37
---
What is the configure string that i use to recreate this?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32423
--- Comment #5 from pinskia at gmail dot com 2007-06-21 13:47 ---
Subject: Re: ICE in expand_expr_real_1, at expr.c:7109
> You shouldn't introduce calls to langhooks. Why not use mode_for_size?
I was just copying code from fold-const.c. I have the mode already, I
need an integer tree
--- Comment #2 from jv244 at cam dot ac dot uk 2007-06-21 13:50 ---
(In reply to comment #1)
> Can you run the compile inside gdb and check periodically where it wastes its
> time?
>
I have a few gdb backtraces, but it looks like it is just writing the .s file.
At the point where f951
HARDWARE:
-
System: HP-UX B.11.00
Machine: 9000/785
COMPILATION:
/usr/local/bin/gcc -v -save-temps -c TP.c -o TP_c.o
Using built-in specs.
Target: hppa2.0w-hp-hpux11.11
Configured with: ../gcc/configure
Thread model: posix
gcc version 4.1.2
/usr/local/libexec/gcc/hppa2.0w-hp
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Severity|blocker |normal
Component|c |target
GCC ta
! echo "z" > foo.dat
program test
implicit none
integer :: i
open(99,file="foo.dat")
read(99,*) i
print *, i
end program
gfortran:
At line 5 of file x.f90
Fortran runtime error: Bad integer for item 1 in list input
Expected: gfortran prints out the filename and/or unit as other compiler
--- Comment #3 from jv244 at cam dot ac dot uk 2007-06-21 14:28 ---
this is the list of options I have tested, the comment indicates if this yields
a failure or not, basically, you need -O2 and -march=native to trigger the bug
using '-O1 -march=native -pg' or '-O2 -pg' are not sufficient
--- Comment #3 from malitzke at metronets dot com 2007-06-21 14:33 ---
Thanks for helping out again. Enjoy Japan. I was there quite often, dealing
with NEC and Mitsubishi, but as a buyer representative for for multi-million $
projects. At that level it was pleasure to do business, even e
--- Comment #9 from hubicka at ucw dot cz 2007-06-21 14:39 ---
Subject: Re: [4.3 Regression] cannot take address of bit field
>
>
> --- Comment #8 from rguenth at gcc dot gnu dot org 2007-06-21 11:19
> ---
> Ping?
I tought the bug is long fixed by moving the folding from f
--- Comment #1 from frederic dot schuh at neuf dot fr 2007-06-21 14:49
---
Created an attachment (id=13757)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13757&action=view)
preprocessed file in attachment
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32455
This is with the polyhedron test gas_dyn.f90
http://www.polyhedron.co.uk/pb05/polyhedron_benchmark_suite.html
It only occurs for gas_dyn.f90 and only with -m32 (-m64 is ok).
This is with gcc-Version 4.3.0 20070621 on x86_64-unknown-linux-gnu.
gfortran -m32 -march=opteron -ftree-vectorize -O1
--- Comment #10 from rguenther at suse dot de 2007-06-21 14:52 ---
Subject: Re: [4.3 Regression] cannot take address of
bit field
On Thu, 21 Jun 2007, hubicka at ucw dot cz wrote:
>
>
> --- Comment #9 from hubicka at ucw dot cz 2007-06-21 14:39 ---
> Subject: Re: [4.3 Reg
--- Comment #3 from rguenth at gcc dot gnu dot org 2007-06-21 14:54 ---
Subject: Bug 32453
Author: rguenth
Date: Thu Jun 21 14:54:47 2007
New Revision: 125922
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=125922
Log:
2007-06-21 Richard Guenther <[EMAIL PROTECTED]>
PR
--- Comment #4 from rguenth at gcc dot gnu dot org 2007-06-21 14:56 ---
Fixed.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #8 from rguenth at gcc dot gnu dot org 2007-06-21 14:56 ---
Fixed.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
Status|ASSIGNED
--- Comment #1 from burnus at gcc dot gnu dot org 2007-06-21 15:58 ---
Missed to say that yesterday's version (r125874) was ok.
Occurs here with 125909 (bootstrapped) and 125922 (only build). svn status
shows no changes in my tree.
However, it could not be reproduced by Uros with 125920
--- Comment #9 from rask at sygehus dot dk 2007-06-21 15:59 ---
I tried this on top of the patch in comment 3 of bug 32441:
Index: gcc/config/m32c/m32c.c
===
--- gcc/config/m32c/m32c.c (revision 125892)
+++ gcc/config/
--- Comment #10 from rask at sygehus dot dk 2007-06-21 16:15 ---
Created an attachment (id=13758)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13758&action=view)
Preprocessed source code for the problem in comment 9.
$ ./xgcc -B./ -S -dp -o /dev/null ~/complex_io.cc
/home/rask/bu
--- Comment #16 from dir at lanl dot gov 2007-06-21 16:16 ---
Created an attachment (id=13759)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13759&action=view)
Warning free version
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32393
I used the simple program provided in the gcc help pages.
i am using gcc-4.2.0 cross compiler for ARM (I think the problem
should persist in native compilers as well)
a.c
#include
int main(void) {
static int a __attribute__ ((section (".offsets"))) = 0;
static int myname __at
--- Comment #17 from dir at lanl dot gov 2007-06-21 16:29 ---
I have attached version that generates no warnings with gfortran or g95. As I
reduced, it the bug changed - that is the problem with optmization bugs - they
are hard to trap. Anyway there is still a bug for some compilers. gfo
--- Comment #2 from sje at cup dot hp dot com 2007-06-21 16:30 ---
Here is a preprocessed test case"
typedef __builtin_va_list __gnuc_va_list;
typedef __gnuc_va_list va_list;
unsigned char inb_local(unsigned long port) { unsigned char value; __asm__
__vol
atile__("in" "b" " %w1, %" "b"
--- Comment #5 from aldot at gcc dot gnu dot org 2007-06-21 16:44 ---
Without combine, the attribute is ignored:
$ gcc-4.3.orig-HEAD -c pr.c -o /dev/null
pr.c: In function 'f1':
pr.c:3: warning: '__weakref__' attribute ignored
pr.c: In function 'f2':
pr.c:7: warning: '__weakref__' att
hich finds directly one
problem:
Subscript 1 of IA (value 2) is out of range (1:1)
In PRMX, line 134 of test.f
I changed ia(1) to ia(20) - then it works. Actually, it works on my system
(x86_64-unknown-linux-gnu, 4.3.0 20070621) up to the option
-O3 -ffast-math -ftree-vectorize -funroll-all-loops
--- Comment #3 from burnus at gcc dot gnu dot org 2007-06-21 16:58 ---
Dump a valid program which contains equivalence to show a harder case for the
checks (NAG f95 chokes on it).
program main
implicit none
integer :: i, it, jt
real:: tt
equivalence
--- Comment #19 from dominiq at lps dot ens dot fr 2007-06-21 17:04 ---
> Subscript 1 of IA (value 2) is out of range (1:1)
I don't think it really matter as
dimension ia(1),a(20)
is the old style for passing arrays. However there are many uninitialized
variables:
ca
--- Comment #16 from spop at gcc dot gnu dot org 2007-06-21 17:06 ---
Subject: Bug 19590
Author: spop
Date: Thu Jun 21 17:06:05 2007
New Revision: 125925
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=125925
Log:
PR tree-optimization/19590
* tree-vrp.c (adjust_ra
--- Comment #3 from sje at cup dot hp dot com 2007-06-21 17:10 ---
Slightly shorter test case:
unsigned char inb_local(unsigned long port)
{
unsigned char value;
__asm__ __volatile__("in" "b" " %w1, %" "b" "0" : "=a"(value) :
"Nd"(port));
return value;
}
void
x_
--- Comment #4 from sje at cup dot hp dot com 2007-06-21 17:13 ---
With optimization, it looks like we die because df_insn_change_bb can handle
insn_info being NULL but it can't handle insn_info->defs (or uses or eq_uses)
being NULL and that is what is happening with -O2. When no optimi
--- Comment #20 from burnus at gcc dot gnu dot org 2007-06-21 17:14 ---
> as far as I can tell none of them are initialized.
This is why I'm eagerly waiting for Asher fixing PR20441 and why g95 has
-fzero.
> Tracing what is wrong with this code explains why so many people don't like
> f
--- Comment #2 from burnus at gcc dot gnu dot org 2007-06-21 17:22 ---
Further note:
-march=pentium2 works
-march=pentium3 fails
-march=pentium2 -msse fails
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32457
* the exact version of GCC;
gcc -v
Using built-in specs.
Target: powerpc-linux-gnu
Configured with: ../src/configure -v
--enable-languages=c,c++,fortran,objc,obj-c++,treelang --prefix=/usr
--enable-shared --with-system-zlib --libexecdir=/usr/lib
--without-included-gettext --enable-threads=posix
--- Comment #1 from dfranke at gcc dot gnu dot org 2007-06-21 17:49 ---
http://gcc.gnu.org/onlinedocs/gfortran/Error-and-Warning-Options.html
-Wsurprising
[...] This currently produces a warning under the following circumstances:
* An INTEGER SELECT construct has a CASE that can neve
--- Comment #19 from pault at gcc dot gnu dot org 2007-06-21 18:12 ---
(In reply to comment #18)
> Due to huge time constraints, I won't be able to
> do anything with this for the next few weeks. Unassigning
> myself for the time.
> If anybody wants to look over my partial patch and fly
--- Comment #7 from pault at gcc dot gnu dot org 2007-06-21 18:14 ---
This fixes Tobias' testcase of comment #4. Note that it is a diff relative to
the patch for 32298. I'll resubmit the latter in this new form.
Paul
Index: gcc/fortran/trans-intrinsic.c
==
--- Comment #11 from rask at sygehus dot dk 2007-06-21 18:19 ---
Disregard the comment about the code label use count. They have
LABEL_PRESERVE_P (/s) set.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32418
1 - 100 of 144 matches
Mail list logo
