[Bug libgomp/35364] New: ICE on ia64 with vector declaration inside #pragma omp parallel
Overview
The following code fails to compile on ia64 platform with standard options. The
problem is gone when:
- removing the #pragma or
- declaring a double instead of a vector or
- not explicitly assigning a length to the vector
- inlining foo()
Problematic code:
=
#include vector
#include omp.h
class A {
void foo();
};
void A::foo() {
#pragma omp parallel
{
std::vectordouble x(0);
}
}
Steps to reproduce
Compile using
g++ -Wall -fopenmp -save-temps -o ice.o -c ice.h
Build Date Platform
=
Linux 2.4.21-27.0.4.EL #1 SMP ia64 GNU/Linux
GCC version: SVN revision 131821 (2008-01-25) (I don't know about earlier or
later versions)
GCC was configured with
Configured with: ../gcc/configure --prefix=(...) --enable-languages=c++,fortran
--with-gmp=(...) --with-mpfr=(...) CC=gcc CXX=g++
The bug does not show up on amd64 using the same revision.
--
Summary: ICE on ia64 with vector declaration inside #pragma omp
parallel
Product: gcc
Version: 4.3.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: libgomp
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: steigers at phys dot ethz dot ch
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35364
[Bug libgomp/35364] ICE on ia64 with vector declaration inside #pragma omp parallel
--- Comment #1 from steigers at phys dot ethz dot ch 2008-02-25 10:35 --- Created an attachment (id=15221) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15221action=view) Preprocessed file (.ii) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35364
[Bug target/35364] ICE on ia64 with vector declaration inside #pragma omp parallel
--- Comment #3 from steigers at phys dot ethz dot ch 2008-02-25 18:20 --- (In reply to comment #2) What is the ICE? Segmentation fault. GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096 Compiler executable checksum: 84cf8bc685ded1e544606791a260313c ice.h: In member function 'void A::foo()': ice.h:11: internal compiler error: Segmentation fault -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35364
[Bug libgomp/33333] New: libgomp omp-low.c:4251 generates internal compiler error
Overview = When compiling the simple file which is attached to this report, I get the following message suggesting that I report it here: compiler_error.cc:18: internal compiler error: in lower_regimplify, at omp-low.c:4251 Steps to reproduce Compile the attached file compiler_error.cc: g++-4.2.1 -c -Wall -fopenmp -save-temps -o compiler_error.o compiler_error.cc Build Date Platform = Red Hat Enterprise Linux AS release 4 (Nahant Update 5) Linux 2.6.19.2 #8 SMP Mon Jun 25 2007 x86_64 GNU/Linux gcc was configured with: .../gcc-4.2.1/configure --prefix=.../gcc-4.2.1 --program-suffix=-4.2.1 --enable-threads --enable-languages=c++,fortran The bug shows up both with the RedHat gcc version AND the official version. The bug does not occur when I compile the source using the portland compiler, for example. Additional Information == 1. When I remove the try...catch statement, the problem is gone. 2. When I do not loop up to the class variable A::x but a constant (say, 1000), the problem is gone. -- Summary: libgomp omp-low.c:4251 generates internal compiler error Product: gcc Version: 4.2.1 Status: UNCONFIRMED Severity: normal Priority: P3 Component: libgomp AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: steigers at phys dot ethz dot ch http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3
[Bug libgomp/33333] libgomp omp-low.c:4251 generates internal compiler error
--- Comment #2 from steigers at phys dot ethz dot ch 2007-09-07 14:42 --- Created an attachment (id=14170) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14170action=view) Preprocessed file -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3
[Bug libgomp/33333] libgomp omp-low.c:4251 generates internal compiler error
--- Comment #1 from steigers at phys dot ethz dot ch 2007-09-07 14:41 --- Created an attachment (id=14169) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14169action=view) Source code which fails to compile -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3
[Bug c++/30711] New: Ifstream bug using Red Hat g++-4.1.0 compared to g++-4.0.2
When an ifstream object is used in several open()-close() cycles, it seems to remember past cycles in the Red Hat gcc version. STEPS TO REPRODUCE: 1) Compile the attached code using gcc version 4.1.0 20060515 (Red Hat 4.1.0-18) 2) Compile the attached code using gcc version 4.0.2 3) Create a file named filetest.mat in the directory where you exectude the binaries 4) Compare the results: They are different ACTUAL RESULTS: (Output from program compiled with gcc 4.1.0 20060515 (Red Hat 4.1.0-18)) -- Try 1: ifstream-declaration outside loop (not working) could not open file ./not_existing_directory/filetest.mat could not open file ./filetest.mat -- Try 2: ifstream-declaration inside loop (working) could not open file ./not_existing_directory/filetest.mat could open file ./filetest.mat -- EXPECTED RESULTS: (Output from program compiled with gcc 4.0.2) -- Try 1: ifstream-declaration outside loop (not working) could not open file ./not_existing_directory/filetest.mat could open file ./filetest.mat -- Try 2: ifstream-declaration inside loop (working) could not open file ./not_existing_directory/filetest.mat could open file ./filetest.mat -- BUILD INFOS: Faulty version: --- gcc version 4.1.0 20060515 (Red Hat 4.1.0-18) Using built-in specs. Target: x86_64-redhat-linux Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --with-gxx-include-dir=/usr/include/c++/3.4.3 --enable-libgcj-multifile --enable-languages=c,c++,java,f95 --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=x86_64-redhat-linux Thread model: posix Working version: gcc version 4.0.2 Reading specs from /usr/pack/gcc-4.0.2-mo/amd64-debian-linux3.1/lib/gcc/x86_64-unknown-linux-gnu/4.0.2/specs Target: x86_64-unknown-linux-gnu Configured with: ../configure --prefix=/usr/pack/gcc-4.0.2-mo --exec-prefix=/usr/pack/gcc-4.0.2-mo/amd64-debian-linux3.1 --with-gxx-include-dir=/usr/pack/gcc-4.0.2-mo/include/c++ --disable-nls --enable-languages=c,c++,f95 Thread model: posix -- Summary: Ifstream bug using Red Hat g++-4.1.0 compared to g++- 4.0.2 Product: gcc Version: 4.1.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: c++ AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: steigers at phys dot ethz dot ch http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30711
[Bug c++/30711] Ifstream bug using Red Hat g++-4.1.0 compared to g++-4.0.2
--- Comment #1 from steigers at phys dot ethz dot ch 2007-02-06 10:05 --- Created an attachment (id=13012) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13012action=view) Code for testing the bug -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30711
[Bug c++/30711] Ifstream bug using Red Hat g++-4.1.0 compared to g++-4.0.2
--- Comment #2 from steigers at phys dot ethz dot ch 2007-02-06 10:33 --- Apparently if the close() statement is followed by a clear() statement it works. I t also works if close() is omitted and only clear() is called. I therefore suspect that someone forgot to change the status of the stream in the open() method of the fulty compiler. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30711
