seen with the 4.3 branch and trunk 20080711 on i486-linux-gnu, the following
preprocessed source doesn't finish to build after 120 cpu minutes. works with
4.2.4, or with -O2.
--
Summary: [4.3/4.4 regression] endless (?) loop building with -O3
Product: gcc
--- Comment #1 from doko at ubuntu dot com 2008-07-12 07:07 ---
Created an attachment (id=15903)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15903action=view)
preprocessed source
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36811
--- Comment #2 from doko at ubuntu dot com 2008-07-12 07:09 ---
command to build:
$ gcc -g -O3 -c utils.i
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36811
--- Comment #10 from dominiq at lps dot ens dot fr 2008-07-12 09:39 ---
Formatted OPEN also hangs:
open(unit=11,form='formatted')
end
It seems that all I/Os involving external files hang, but 'print *,' or
'write(tmp,*)' (where 'tmp' is an internal buffer) works.
--
compiling the attached preprocessed source, triggers an ICE with g++-4.4:
[EMAIL PROTECTED]:~/workspace/nova$ g++-4.4 -v
Using built-in specs.
Target: i686-pc-linux-gnu
Configured with: ./configure --program-suffix=-4.4
Thread model: posix
gcc version 4.4.0 20080703 (experimental) (GCC)
--- Comment #1 from tim at klingt dot org 2008-07-12 10:47 ---
Created an attachment (id=15904)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15904action=view)
compressed pre-processed source
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36812
--- Comment #6 from domob at gcc dot gnu dot org 2008-07-12 11:27 ---
Subject: Bug 33221
Author: domob
Date: Sat Jul 12 11:26:50 2008
New Revision: 137737
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=137737
Log:
2008-07-12 Daniel Kraft [EMAIL PROTECTED]
* resolve.c
--- Comment #4 from janus at gcc dot gnu dot org 2008-07-12 15:04 ---
Rev. 137717 works again, and compiles the test case without any error. Has
anyone fixed this?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36788
--- Comment #5 from ebotcazou at gcc dot gnu dot org 2008-07-12 15:23
---
Has anyone fixed this?
2008-07-10 Daniel Berlin [EMAIL PROTECTED]
* tree-ssa-pre.c (create_expression_by_pieces): Add fold_convert
calls.
(eliminate): Ditto.
(execute_pre):
--- Comment #15 from gcc at abeckmann dot de 2008-07-12 16:01 ---
Hi,
this bug (not searching for libgomp.spec in the correct place) is still present
in the gcc-4_3-branch as of today (r137745).
Configured on an amd64 Debian testing/unstable machine:
../gcc-4_3-branch/configure
--- Comment #7 from tromey at gcc dot gnu dot org 2008-07-12 17:06 ---
Fixed on trunk.
--
tromey at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #3 from tromey at gcc dot gnu dot org 2008-07-12 17:10 ---
I tried this and it is fixed on trunk.
Closing.
--
tromey at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #6 from tromey at gcc dot gnu dot org 2008-07-12 17:18 ---
This was fixed for 4.3, and I think it unlikely that anyone will
backport the doc fix. So, I am closing this.
--
tromey at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #7 from tromey at gcc dot gnu dot org 2008-07-12 17:19 ---
This was fixed for 4.3 and I think it unlikely that anyone will backport
the doc fix. So, I am closing this.
--
tromey at gcc dot gnu dot org changed:
What|Removed |Added
The second error message issued by gcc 4.3.0 for the ill-formed program below
looks wrong. I would expect it to read the same as the first one.
$ cat t.cpp g++ t.cpp
template class T struct A { };
template class T, class U struct B { typedef AT Y; };
template class T void foo (T*) { }
template
--- Comment #1 from pinskia at gcc dot gnu dot org 2008-07-12 19:44 ---
the error message is slightly wrongly written but this is true for any local
typedef, I can find the other bug where templates are not involved.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36813
--- Comment #15 from pinskia at gcc dot gnu dot org 2008-07-12 19:46
---
*** Bug 36813 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 2008-07-12 19:46 ---
*** This bug has been marked as a duplicate of 15672 ***
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Hello,
I've compiled the code below using `g++ -Wuninitialized -O test.cc -o test':
bool cMyClass::Init()
{
bool retval;
if (some_function())
{
// If the `goto' is executed, 'return retval;' will use the
// uninitialized value of 'retval'.
goto error;
}
With --enable-languages=c,fortran, system trys to build gnattools and fails
with this error:
gnu-make[2]: Entering directory `/home/delisle/gcc/obj44/gcc'
gnu-make[2]: Leaving directory `/home/delisle/gcc/obj44/gcc'
/bin/sh: line 1: gnatls: not found
gnu-make[2]: Entering directory
--- Comment #1 from pinskia at gcc dot gnu dot org 2008-07-12 20:14 ---
Did you remove the Ada subdirectory?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36815
Gcc 4.3.0 diagnoses the line marked error? in the program below.
I'd like to believe the program is well-formed and that the compiler should
be able to correctly deduce the template arguments from the initialed
expression without explicitly providing the template argument list.
template class T
--- Comment #2 from kkojima at gcc dot gnu dot org 2008-07-12 23:46 ---
Created an attachment (id=15905)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15905action=view)
A reduced test case for -O1 -fpic
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36782
--- Comment #3 from kkojima at gcc dot gnu dot org 2008-07-12 23:53 ---
I've added a reduced testcase and confirmed that 4.2.4 works too.
The RTL dumps show what is going on in the problematic case.
The backend creates a REG_EQUAL note for the GOT access and CSE
pass locates a memory
--- Comment #2 from kkojima at gcc dot gnu dot org 2008-07-13 00:05 ---
I've confirmed that 4.2.4 works and the patch suggested at #1
fixes this error on 4.3. I'll test it when 4.3 can be built
for SH again.
--
kkojima at gcc dot gnu dot org changed:
What|Removed
25 matches
Mail list logo
