-
Notifications
You must be signed in to change notification settings - Fork 42
Open
Description
Dear Developers,
recently since the system update of our cluster's system. I have to recompile the extrae in our system.
And I noticed that there the latest version is 4.2.5.
Although the compilation of 4.2.5 has no problem(looks like), if I want to run the example case provided in the source code, it just shows Segmentation fault. If I use the same script to compile 4.0.6 and run the same test code, everything works as before.
is there any change log somewhere for external users?
What has been changed from 4.0.6??
Attached is the error message
Welcome to Extrae 4.2.5
Extrae: Parsing the configuration file (./extrae.xml) begins
Extrae: Tracing package is located on .......
Extrae: Generating intermediate files for Paraver traces.
Extrae: MPI routines will collect HW counters information.
Extrae: All MPI_Comm_* calls will be traced.
Extrae: Warning! <openmp> tag will be ignored. This library does not support OpenMP.
Extrae: Tracing 3 level(s) of MPI callers: [ 1 2 3 ]
Extrae: Warning! change-at-time time units not specified. Using seconds
Extrae: PAPI domain set to ALL for HWC set 1
Extrae: HWC set 1 contains following counters < PAPI_TOT_INS (0x80000032) PAPI_TOT_CYC (0x8000003b) PAPI_L1_DCM (0x80000000) PAPI_L2_DCM (0x80000002) PAPI_L3_TCM (0x80000008) PAPI_BR_INS (0x80000037) PAPI_BR_MSP (0x8000002e) > - never changes
Extrae: Warning! change-at-time time units not specified. Using seconds
Extrae: PAPI domain set to ALL for HWC set 2
Extrae: HWC set 2 contains following counters < PAPI_TOT_INS (0x80000032) PAPI_TOT_CYC (0x8000003b) PAPI_VEC_SP (0x80000069) PAPI_SR_INS (0x80000036) PAPI_LD_INS (0x80000035) PAPI_FP_INS (0x80000034) > - never changes
Extrae: Cannot parse HWC RESOURCE_STALLS in set 1, skipping
Extrae: Error! Cannot query information for hardware counter RESOURCE_STALLS (0x00000000). Check set 1.
Extrae: Error! Hardware counter PAPI_FP_INS (0x80000034) cannot be added in set 2 (task 1, thread 0)
PAPI error -8: Event exists, but cannot be counted due to hardware resource limits
Extrae: Error! Hardware counter PAPI_FP_INS (0x80000034) cannot be added in set 2 (task 0, thread 0)
PAPI error -8: Event exists, but cannot be counted due to hardware resource limits
MPI startup(): Warning: I_MPI_PMI_LIBRARY will be ignored since the hydra process manager was found
Extrae: Resource usage is disabled at flush buffer.
Extrae: Memory usage is disabled at flush buffer.
Extrae: Tracing buffer can hold 5000000 events
Extrae: Circular buffer disabled.
Extrae: Sampling buffers will be written at instrumentation points
Extrae: Warning! <merge> tag will be ignored. This library does not have merging process embedded.
Extrae: Dynamic memory instrumentation is disabled.
Extrae: Basic I/O memory instrumentation is disabled.
Extrae: System calls instrumentation is disabled.
Extrae: Parsing the configuration file (./extrae.xml) has ended
Extrae: Intermediate traces will be stored in ........./ld-preload
Extrae: Tracing mode is set to: Detail.
Extrae: Successfully initiated with 2 tasks and 1 threads
MPI startup(): Warning: I_MPI_PMI_LIBRARY will be ignored since the hydra process manager was found
./trace.sh: line 10: 3935534 Segmentation fault (core dumped) $*
===================================================================================
= BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
= RANK 0 PID 3935521 RUNNING AT login01
= KILLED BY SIGNAL: 9 (Killed)
===================================================================================
attached is the installation script I created. it works fine for 4.0.6
ml load intel-oneapi-advisor/2024.1.0
ml load intel-oneapi-compilers/2023.2.4
ml load intel-oneapi-dal/2024.2.0-tbb
ml load intel-oneapi-inspector/2024.1.0
ml load intel-oneapi-itac/2022.1.0
ml load intel-oneapi-mkl/2024.1.0-impi-tbb
ml load intel-oneapi-mpi/2021.12.1
ml load intel-oneapi-tbb/2021.12.0
ml load intel-oneapi-vtune/2024.1.0
FILE_NAME="extrae-4.2.5-src.tar.bz2"
EXTRACTED_DIR="extrae-4.2.5"
if [ ! -f "$FILE_NAME" ]; then
echo "错误:文件 $FILE_NAME 不存在!"
exit 1
fi
echo "正在解压缩文件 $FILE_NAME..."
tar -xjf "$FILE_NAME"
if [ $? -ne 0 ]; then
echo "错误:解压缩文件失败!"
exit 1
fi
if [ ! -d "$EXTRACTED_DIR" ]; then
echo "错误:解压后的文件夹 $EXTRACTED_DIR 不存在!"
exit 1
fi
cd $EXTRACTED_DIR
./bootstrap
CC=gcc
CFLAGS='-g -O2 -fno-optimize-sibling-calls'
CXX=g++
CXXFLAGS='-g -O2 -fno-optimize-sibling-calls'
FC='gfortran'
FCFLAGS='-g -O2'
echo "Config done"
installpath=/......../zzz_extrae_compile_test/spackInteltest_01
./configure \
--prefix=$installpath \
--with-mpi=$I_MPI_ROOT \
--with-mpi-libs=${I_MPI_ROOT}/lib/release \
--with-binutils=......../binutils_install \
--with-papi=....../PAPI_install \
--with-unwind=....../libunwind_install \
--with-xml-prefix=...../libxml2 \
--without-dyninst \
--disable-parallel-merge > config.log
make -j 8
make install
echo "正在删除文件夹 $EXTRACTED_DIR..."
rm -rf "$EXTRACTED_DIR"
if [ $? -eq 0 ]; then
echo "成功:文件夹 $EXTRACTED_DIR 已被删除!"
else
echo "错误:删除文件夹 $EXTRACTED_DIR 失败!"
exit 1
fi
followings are the packages
binutils-2.44.tar
libunwind-1.8.1.tar.gz
papi-7.2.0b1.tar.gz
libxml2-2.13.6.tar.xz
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
No labels