--- Comment #23 from fxcoudert at gcc dot gnu dot org 2005-10-29 12:44
---
Patch for that PR submitted
(http://gcc.gnu.org/ml/fortran/2005-10/msg00682.html).
--
fxcoudert at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #7 from fxcoudert at gcc dot gnu dot org 2005-10-29 12:55
---
Confirmed and marked as an enhancement.
The standard clearly forbids the behaviour the reporter would like (and that
some commercial compilers exhibit), so it's not a bug. However, it could be an
interesting feat
--- Comment #33 from fxcoudert at gcc dot gnu dot org 2005-10-29 12:57
---
(In reply to comment #30)
> I do agree with you. Can you re-submit the patch on the fortran ml for
> approval?
ping*2
--
fxcoudert at gcc dot gnu dot org changed:
What|Removed
--- Comment #8 from pinskia at gcc dot gnu dot org 2005-10-29 13:21 ---
And I really think this is just an manufacting of PR 22298 in a different way.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
CC||pinskia at gcc dot gnu dot
|
--- Comment #1 from fxcoudert at gcc dot gnu dot org 2005-10-29 13:42
---
Confirmed. I'm working on this and related problems in io.c.
--
fxcoudert at gcc dot gnu dot org changed:
What|Removed |Added
---
--- Comment #9 from fxcoudert at gcc dot gnu dot org 2005-10-29 13:47
---
(In reply to comment #8)
> And I really think this is just an manufacting of PR 22298 in a different way.
No it's not. The same problem happens with dynamic linking, static linking has
nothing to do with this bug
--- Comment #10 from pinskia at gcc dot gnu dot org 2005-10-29 14:09
---
Can you try the following "work around":
gfortran -fPIC -shared -Wl,-Bsymbolic -o libtest.so test.f \
-Wl,--whole-archive /usr/lib64/libgfortran.a -Wl,--no-whole-archive
This should work and if it does, then the
--- Comment #11 from fxcoudert at gcc dot gnu dot org 2005-10-29 14:24
---
(In reply to comment #10)
> Can you try the following "work around":
> gfortran -fPIC -shared -Wl,-Bsymbolic -o libtest.so test.f \
> -Wl,--whole-archive /usr/lib64/libgfortran.a -Wl,--no-whole-archive
"That wor
--- Comment #12 from iwan at irs dot phy dot nrc dot ca 2005-10-29 15:01
---
Subject: Re: Fortran I/O to same unit number in main program and in a shared
library
Using -Wl,--whole-archive /usr/lib64/libgfortran.a -Wl,--no-whole-archive
works for me too.
Thanks.
--
http://gcc.g
--- Comment #9 from steven at gcc dot gnu dot org 2005-10-29 15:11 ---
I'm testing the patch from comment #8 on a few targets.
--
steven at gcc dot gnu dot org changed:
What|Removed |Added
---
--- Comment #2 from steven at gcc dot gnu dot org 2005-10-29 15:15 ---
How can this possibly be a GCC 4.0/4.1 regression?!
Before GCC 4.0 we didn't even have a Fortran with support for derived types.
--
steven at gcc dot gnu dot org changed:
What|Removed
--- Comment #3 from pinskia at gcc dot gnu dot org 2005-10-29 15:20 ---
(In reply to comment #2)
> How can this possibly be a GCC 4.0/4.1 regression?!
Because in 4.0.0, this worked.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
---
--- Comment #13 from fxcoudert at gcc dot gnu dot org 2005-10-29 15:39
---
Then we can close this PR as duplicate of 22298.
*** This bug has been marked as a duplicate of 22298 ***
--
fxcoudert at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #2 from fxcoudert at gcc dot gnu dot org 2005-10-29 15:39
---
*** Bug 24576 has been marked as a duplicate of this bug. ***
--
fxcoudert at gcc dot gnu dot org changed:
What|Removed |Added
--
--- Comment #3 from pinskia at gcc dot gnu dot org 2005-10-29 15:52 ---
This really does not need to be targeted towards 4.1 (this is a bug I filed
looking at code gen problems with some weird code which I made up).
--
pinskia at gcc dot gnu dot org changed:
What|Remo
--- Comment #5 from steven at gcc dot gnu dot org 2005-10-29 16:28 ---
This does not fail for me... Neither the original test case nor the reduced
one. The compiler I used was "GNU C++ version 4.1.0 20051029 (experimental)
(x86_64-unknown-linux-gnu)"
--
http://gcc.gnu.or
The following code should not abort:
struct vbase {};
struct foo : virtual vbase {
foo() {
throw "exception in foo ctor";
}
};
main()
{
struct bar : public foo {};
try {
bar a;
}
catch ( ... ) {
}
return 0;
};
This is forwarded from:
http:
--- Comment #6 from steven at gcc dot gnu dot org 2005-10-29 16:41 ---
This does ICE for i686 or for AMD64 -m32. For 64 bits it passes for me.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24365
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|--- |4.0.3
Version|unknown |4.1.0
http://
--- Comment #1 from pinskia at gcc dot gnu dot org 2005-10-29 16:42 ---
Confirmed as a bug in:
http://gcc.gnu.org/ml/gcc/2005-09/msg00329.html
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--
--- Comment #2 from clum at users dot sourceforge dot net 2005-10-29 17:57
---
This is really weird. I have used gcc many, many times, (but never on Knoppix
before) and have never had such a problem. I'm having the exact same problem.
It happens right away when I compile, even if I only
--- Comment #7 from pinskia at gcc dot gnu dot org 2005-10-29 18:01 ---
I have a fix but it could cause missed optimizations but from the looks of it,
that is because we don't use SFT on complex variables (or it might be on
RESULT_DECL which is the real issue).
--
pinskia at gcc dot
--- Comment #6 from pinskia at gcc dot gnu dot org 2005-10-29 18:09 ---
Another one of these I filed this bug after looking at someone else's bug for
code gen regressions.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
-
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Component|regression |middle-end
Keywords||missed-opt
/*
* The C99 standard intends x+I*y to be used for this, but x+I*y is
* currently unusable because gcc introduces many overflow,
* underflow, sign and efficiency bugs by rewriting I*y as
* (0.0+I)*(y+0.0*I) and laboriously computing the full complex product.
* In particular, I*Inf is corrupted
--- Comment #1 from joseph at codesourcery dot com 2005-10-29 20:09 ---
Subject: Re: New: Complex arithmetic on special cases is
incorrect.
On Sat, 29 Oct 2005, kargl at gcc dot gnu dot org wrote:
> * underflow, sign and efficiency bugs by rewriting I*y as
> * (0.0+I)*(y+0.0*I) an
--- Comment #2 from pinskia at gcc dot gnu dot org 2005-10-29 20:10 ---
Hmm, I want to say this is a defect in the standard but what do I know.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
-
--- Comment #3 from pinskia at gcc dot gnu dot org 2005-10-29 20:14 ---
(In reply to comment #1)
> I suspect there are lots of presumptions in the compiler that arithmetic
> operations such as PLUS_EXPR and MULT_EXPR have both operands of the same
> type, which would need to be fixed t
--- Comment #4 from sgk at troutmask dot apl dot washington dot edu
2005-10-29 20:27 ---
Subject: Re: Complex arithmetic on special cases is incorrect.
On Sat, Oct 29, 2005 at 08:09:45PM -, joseph at codesourcery dot com wrote:
>
> > * underflow, sign and efficiency bugs by rewr
--- Comment #2 from pinskia at gcc dot gnu dot org 2005-10-29 20:45 ---
Should this block PR 16620 and PR 16989?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21343
--- Comment #3 from joseph at codesourcery dot com 2005-10-29 20:48 ---
Subject: Re: [4.0/4.1 Regression] incompatible internal linkage
declarations in different scopes not diagnosed
On Sat, 29 Oct 2005, pinskia at gcc dot gnu dot org wrote:
> Should this block PR 16620 and PR 16989?
--- Comment #3 from pinskia at gcc dot gnu dot org 2005-10-29 20:50 ---
Confirmed.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNCON
--- Comment #5 from pinskia at gcc dot gnu dot org 2005-10-29 20:51 ---
Confirmed.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Status|UNCON
--- Comment #2 from steven at gcc dot gnu dot org 2005-10-29 21:52 ---
You'll have to investigate the drop a bit more. That patch you identified was
supposed to fix the performance problems of a previous fix for some bug. It
should not have any adverse effects. You'll have to identify
Compile the to-be-attached test case, and you get an ICE in the front end after
errors:
jigdo-file.cc: At global scope:
jigdo-file.cc:491: error: 'class JigdoFileCmd::Command' is not a class or
namespace
jigdo-file.cc:491: error: ISO C++ forbids declaration of 'cmdOptions' with no
type
int::cmdOp
--- Comment #1 from steven at gcc dot gnu dot org 2005-10-29 22:10 ---
Created an attachment (id=10080)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=10080&action=view)
testcase
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24582
--- Comment #2 from pinskia at gcc dot gnu dot org 2005-10-29 22:23 ---
This is my fault, will fix but tomorrow.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #5 from amodra at gcc dot gnu dot org 2005-10-29 22:35 ---
Subject: Bug 23070
Author: amodra
Date: Sat Oct 29 22:35:37 2005
New Revision: 105998
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=105998
Log:
PR target/23070
* config/rs6000/rs6000.c (funct
--- Comment #26 from steven at gcc dot gnu dot org 2005-10-29 22:36 ---
Waiting for someone to look into this...
--
steven at gcc dot gnu dot org changed:
What|Removed |Added
-
--- Comment #6 from amodra at bigpond dot net dot au 2005-10-29 22:36
---
Fixed.
--
amodra at bigpond dot net dot au changed:
What|Removed |Added
Status|ASSI
--- Comment #20 from steven at gcc dot gnu dot org 2005-10-29 22:38 ---
amacleod, are you going to post your patch and/or commit it??
--
steven at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #8 from steven at gcc dot gnu dot org 2005-10-29 22:43 ---
Is the only problem here one single extra SET produced by expand, or do we have
a bug here somewhere?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23335
--- Comment #6 from pcarlini at suse dot de 2005-10-29 22:45 ---
Fixed.
--
pcarlini at suse dot de changed:
What|Removed |Added
Status|ASSIGNED
When a Java program is run in a chroot environment on Linux,
InetAddress.getLocalHost() returns a broken InetAddress object (broken in the
same way as it was in bug 17069).
Test case:
import java.net.InetAddress;
import java.net.UnknownHostException;
public class TestAddr {
public stati
--- Comment #1 from greenrd at gcc dot gnu dot org 2005-10-30 00:21 ---
Also, the hostname command does work inside the chroot - it returns the
hostname of the real machine. So I don't think this is a reoccurrence of bug
17069.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24583
--- Comment #3 from mckelvey at maskull dot com 2005-10-30 01:23 ---
OK, I updated my cygwin and it built. You can close the bug.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24394
Hey Jennie,
Remember I told you about that e-mail I got the other day saying I could get my
B/A, B/S/C, M/A, M/S/C, M/B/A, P/H/D, in 2 weeks without any study required and
100% verifiable ?
Well bro I gave it a shot and now 2 weeks later I have my BA and
working on getting my Masters which I w
--- Comment #4 from jvdelisle at gcc dot gnu dot org 2005-10-30 02:59
---
OK I have a patch for the segfault I was seeing and now I see the error the OP
discusses. I will continue on that as soon as I get the other bug fix
submitted for review. (I will have to create a new PR for the
The following program segfaults when the input file has an empty string
specified.
PROGRAM foo
IMPLICIT NONE
CHARACTER(len=10) var
CHARACTER(len=10) sname
NAMELIST /input/ var
sname = 'howdy'
WRITE(*,*) sname, "<<<"
OPEN(unit=1,file="
--- Comment #1 from jvdelisle at gcc dot gnu dot org 2005-10-30 03:56
---
With this patch there is a side effect with read_null_string.f90 in
gfortran.fortran-torture. Checking on it.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24584
--- Comment #2 from jvdelisle at gcc dot gnu dot org 2005-10-30 04:51
---
OK Now, Regrssion tested OK Will submit to list for approval
Index: io/list_read.c
===
--- io/list_read.c (revision 105987)
+++ io/list_read.
--- Comment #4 from pinskia at gcc dot gnu dot org 2005-10-30 05:48 ---
Closing as works for me then.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
53 matches
Mail list logo
