Merge branch 'master' of https://xgitlab.cels.anl.gov/darshan/darshan

ChangeLog

@@ -2,13 +2,27 @@
 Darshan Release Change Log
 --------------------------
 
+Darshan-3.1.2
+=============
+* bug fix in resolving underlying call to fopen64 when using the LD_PRELOAD
+  instrumentation mechanism
+* bug fix to make sure Lustre module headers are installed when installing
+  darshan-util
+  - reported by Matthieu Dorrier
+
+Darshan-3.1.1
+=============
+* bug fix to MPIIO_F_WRITE_START_TIMESTAMP, which may have produced incorrect
+  timestamps in some cases
+  - reported by Wucherl (William) Yoo
+
 Darshan-3.1.0
 =============
 * add stdio I/O library instrumentation module (Philip Carns)
     - this handles instrumentation of file stream I/O functions
       like fopen(), fprintf(), fscanf(), etc.
     - this module also captures stats on the standard streams (stdin,
-      stdout, & stdin)
+      stdout, & stderr)
 * add Lustre instrumentation module (Glenn Lockwood)
     - this module provides Lustre striping details (e.g., stripe
       width, stripe size, list of OSTs a file is striped over)

darshan-mpiio-log-format.h

@@ -8,7 +8,7 @@
 #define __DARSHAN_MPIIO_LOG_FORMAT_H
 
 /* current MPI-IO log format version */
-#define DARSHAN_MPIIO_VER 1
+#define DARSHAN_MPIIO_VER 2
 
 /* TODO: maybe use a counter to track cases in which a derived datatype is used? */
 

darshan-runtime/configure

 #! /bin/sh
 # Guess values for system-dependent variables and create Makefiles.
-# Generated by GNU Autoconf 2.69 for darshan-runtime 3.1.0.
+# Generated by GNU Autoconf 2.69 for darshan-runtime 3.1.2.
 #
 #
 # Copyright (C) 1992-1996, 1998-2012 Free Software Foundation, Inc.
@@ -577,8 +577,8 @@ MAKEFLAGS=
 # Identity of this package.
 PACKAGE_NAME='darshan-runtime'
 PACKAGE_TARNAME='darshan-runtime'
-PACKAGE_VERSION='3.1.0'
-PACKAGE_STRING='darshan-runtime 3.1.0'
+PACKAGE_VERSION='3.1.2'
+PACKAGE_STRING='darshan-runtime 3.1.2'
 PACKAGE_BUGREPORT=''
 PACKAGE_URL=''
 
@@ -1259,7 +1259,7 @@ if test "$ac_init_help" = "long"; then
   # Omit some internal or obsolete options to make the list less imposing.
   # This message is too long to be a string in the A/UX 3.1 sh.
   cat <<_ACEOF
-\`configure' configures darshan-runtime 3.1.0 to adapt to many kinds of systems.
+\`configure' configures darshan-runtime 3.1.2 to adapt to many kinds of systems.
 
 Usage: $0 [OPTION]... [VAR=VALUE]...
 
@@ -1321,7 +1321,7 @@ fi
 
 if test -n "$ac_init_help"; then
   case $ac_init_help in
-     short | recursive ) echo "Configuration of darshan-runtime 3.1.0:";;
+     short | recursive ) echo "Configuration of darshan-runtime 3.1.2:";;
    esac
   cat <<\_ACEOF
 
@@ -1335,7 +1335,7 @@ Optional Features:
                           Set log files to be group readable
   --disable-bgq-mod       Disables compilation and use of BG/Q module (for BG/Q systems)
   --disable-lustre-mod    Disables compilation and use of the Lustre module
-  --enable-mmmap-logs     Enables ability to mmap I/O data to log file
+  --enable-mmap-logs      Enables ability to mmap I/O data to log file
 
 Optional Packages:
   --with-PACKAGE[=ARG]    use PACKAGE [ARG=yes]
@@ -1431,7 +1431,7 @@ fi
 test -n "$ac_init_help" && exit $ac_status
 if $ac_init_version; then
   cat <<\_ACEOF
-darshan-runtime configure 3.1.0
+darshan-runtime configure 3.1.2
 generated by GNU Autoconf 2.69
 
 Copyright (C) 2012 Free Software Foundation, Inc.
@@ -1783,7 +1783,7 @@ cat >config.log <<_ACEOF
 This file contains any messages produced by compilers while
 running configure, to aid debugging if configure makes a mistake.
 
-It was created by darshan-runtime $as_me 3.1.0, which was
+It was created by darshan-runtime $as_me 3.1.2, which was
 generated by GNU Autoconf 2.69.  Invocation command line was
 
   $ $0 $@
@@ -4360,7 +4360,7 @@ else
   MPICH_LIB_OLD=0
 fi
 
-DARSHAN_VERSION="3.1.0"
+DARSHAN_VERSION="3.1.2"
 
 
 
@@ -4880,7 +4880,7 @@ cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1
 # report actual input values of CONFIG_FILES etc. instead of their
 # values after options handling.
 ac_log="
-This file was extended by darshan-runtime $as_me 3.1.0, which was
+This file was extended by darshan-runtime $as_me 3.1.2, which was
 generated by GNU Autoconf 2.69.  Invocation command line was
 
   CONFIG_FILES    = $CONFIG_FILES
@@ -4942,7 +4942,7 @@ _ACEOF
 cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
 ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //; s/[\\""\`\$]/\\\\&/g'`"
 ac_cs_version="\\
-darshan-runtime config.status 3.1.0
+darshan-runtime config.status 3.1.2
 configured by $0, generated by GNU Autoconf 2.69,
   with options \\"\$ac_cs_config\\"
 

darshan-runtime/configure.in

@@ -5,7 +5,7 @@ dnl Process this file with autoconf to produce a configure script.
 dnl You may need to use autoheader as well if changing any DEFINEs
 
 dnl sanity checks, output header, location of scripts used here
-AC_INIT([darshan-runtime], [3.1.0])
+AC_INIT([darshan-runtime], [3.1.2])
 AC_CONFIG_SRCDIR([darshan.h])
 AC_CONFIG_AUX_DIR(../maint/config)
 AC_CONFIG_HEADER(darshan-runtime-config.h)
@@ -61,7 +61,7 @@ AC_ARG_ENABLE(
 )
 
 AC_ARG_ENABLE(mmap-logs,
-[  --enable-mmmap-logs     Enables ability to mmap I/O data to log file],
+[  --enable-mmap-logs      Enables ability to mmap I/O data to log file],
 [if test "x$enableval" = "xyes" ; then
     AC_DEFINE(__DARSHAN_ENABLE_MMAP_LOGS, 1, Define if Darshan should mmap data structures to log file)
     ENABLE_MMAP_LOGS=1

darshan-runtime/doc/darshan-runtime.txt

@@ -219,7 +219,7 @@ should be compiled using the normal, unmodified MPI compiler.
 
 To use this mechanism, set the `LD_PRELOAD` environment variable to the full
 path to the Darshan shared library. The preferred method of inserting Darshan
-instrumentation in this case is to set the LD_PRELOAD variable specifically
+instrumentation in this case is to set the `LD_PRELOAD` variable specifically
 for the application of interest. Typically this is possible using
 command line arguments offered by the `mpirun` or `mpiexec` scripts or by
 the job scheduler:
@@ -249,6 +249,11 @@ unrelated binaries:
 export LD_PRELOAD=/home/carns/darshan-install/lib/libdarshan.so
 ----
 
+[NOTE]
+For SGI systems running the MPT environment, it may be necessary to set the `MPI_SHEPHERD`
+environment variable equal to `true` to avoid deadlock when preloading the Darshan shared
+library.
+
 === Instrumenting dynamically-linked Fortran applications
 
 Please follow the general steps outlined in the previous section.  For
@@ -444,8 +449,9 @@ method has been briefly tested using both GNU and Intel compilers.
 [NOTE]
 ====
 Darshan is only known to work with C and C++ executables generated by the
-Intel MPI suite.  Darshan will not produce instrumentation for Fortran
-executables.  For more details please check this Intel forum discussion:
+Intel MPI suite in versions prior to the 2017 version -- Darshan will not
+produce instrumentation for Fortran executables in these earlier versions (pre-2017).
+For more details on this issue please check this Intel forum discussion:
 
 http://software.intel.com/en-us/forums/showthread.php?t=103447&o=a&s=lr
 ====

darshan-runtime/lib/darshan-mpiio.c

@@ -210,7 +210,7 @@ static int enable_dxt_io_trace = 0;
         rec_ref->file_rec->counters[MPIIO_RW_SWITCHES] += 1; \
     rec_ref->last_io_type = DARSHAN_IO_WRITE; \
     if(rec_ref->file_rec->fcounters[MPIIO_F_WRITE_START_TIMESTAMP] == 0 || \
-     rec_ref->file_rec->fcounters[MPIIO_F_READ_START_TIMESTAMP] > __tm1) \
+     rec_ref->file_rec->fcounters[MPIIO_F_WRITE_START_TIMESTAMP] > __tm1) \
         rec_ref->file_rec->fcounters[MPIIO_F_WRITE_START_TIMESTAMP] = __tm1; \
     rec_ref->file_rec->fcounters[MPIIO_F_WRITE_END_TIMESTAMP] = __tm2; \
     if(rec_ref->file_rec->fcounters[MPIIO_F_MAX_WRITE_TIME] < __elapsed) { \

darshan-test/regression/run-all.sh

@@ -45,17 +45,6 @@ fi
 # set up environment for tests according to platform
 source $DARSHAN_TESTDIR/$DARSHAN_PLATFORM/env.sh
 
-# in case of using LD_PRELOAD, it is possible errors with Darshan symbols
-# (e..g, forgetting to resolve a symbol Darshan wraps using dlsym) can
-# cause arbitrary binaries to crash. We check the output of the true
-# command to successfully identify failures like this and exit
-true_out=`/bin/true 2>&1`
-if [ $? -ne 0 -o -n "$true_out" ]; then
-    echo -n $true_out > tmp
-    echo "environment setup failed"
-    exit 1
-fi
-
 failure_count=0
 
 for i in `ls $DARSHAN_TESTDIR/test-cases/*.sh`; do

darshan-test/regression/workstation-dynamic/env.sh

@@ -38,3 +38,20 @@ FULL_MPICC_PATH=`which mpicc`
 export LD_PRELOAD=`dirname $FULL_MPICC_PATH`/../lib/libfmpich.so:$DARSHAN_PATH/lib/libdarshan.so:$LD_PRELOAD
 
 export DARSHAN_RUNJOB="mpiexec -n $DARSHAN_DEFAULT_NPROCS"
+
+# in case of using LD_PRELOAD, it is possible errors with Darshan symbols
+# (e..g, forgetting to resolve a symbol Darshan wraps using dlsym) can
+# cause arbitrary binaries to crash. We check the output of the true
+# command and ls to try to identify failures like this and exit
+true_out=`true`
+if [ $? -ne 0 -o -n "$true_out" ]; then
+    echo -n "$true_out"
+    echo "true command test failed"
+    exit 1
+fi
+ls_out=`ls 1>/dev/null`
+if [ $? -ne 0 -o -n "$ls_out" ]; then
+    echo -n "$ls_out"
+    echo "ls command test failed"
+    exit 1
+fi

darshan-util/Makefile.in

@@ -155,6 +155,7 @@ endif
 	install -m 644 $(srcdir)/darshan-pnetcdf-logutils.h $(includedir)
 	install -m 644 $(srcdir)/darshan-bgq-logutils.h $(includedir)
 	install -m 644 $(srcdir)/darshan-null-logutils.h $(includedir)
+	install -m 644 $(srcdir)/darshan-lustre-logutils.h $(includedir)
 	install -m 644 $(srcdir)/darshan-stdio-logutils.h $(includedir)
 	install -m 644 $(srcdir)/darshan-dxt-logutils.h $(includedir)
 	install -m 644 $(srcdir)/../darshan-null-log-format.h $(includedir)
@@ -163,6 +164,7 @@ endif
 	install -m 644 $(srcdir)/../darshan-hdf5-log-format.h $(includedir)
 	install -m 644 $(srcdir)/../darshan-pnetcdf-log-format.h $(includedir)
 	install -m 644 $(srcdir)/../darshan-bgq-log-format.h $(includedir)
+	install -m 644 $(srcdir)/../darshan-lustre-log-format.h $(includedir)
 	install -m 644 $(srcdir)/../darshan-stdio-log-format.h $(includedir)
 	install -m 644 $(srcdir)/../darshan-dxt-log-format.h $(includedir)
 	install -d $(includedir)/uthash-1.9.2

darshan-util/configure

 #! /bin/sh
 # Guess values for system-dependent variables and create Makefiles.
-# Generated by GNU Autoconf 2.69 for darshan-util 3.1.0.
+# Generated by GNU Autoconf 2.69 for darshan-util 3.1.2.
 #
 #
 # Copyright (C) 1992-1996, 1998-2012 Free Software Foundation, Inc.
@@ -577,8 +577,8 @@ MAKEFLAGS=
 # Identity of this package.
 PACKAGE_NAME='darshan-util'
 PACKAGE_TARNAME='darshan-util'
-PACKAGE_VERSION='3.1.0'
-PACKAGE_STRING='darshan-util 3.1.0'
+PACKAGE_VERSION='3.1.2'
+PACKAGE_STRING='darshan-util 3.1.2'
 PACKAGE_BUGREPORT=''
 PACKAGE_URL=''
 
@@ -1247,7 +1247,7 @@ if test "$ac_init_help" = "long"; then
   # Omit some internal or obsolete options to make the list less imposing.
   # This message is too long to be a string in the A/UX 3.1 sh.
   cat <<_ACEOF
-\`configure' configures darshan-util 3.1.0 to adapt to many kinds of systems.
+\`configure' configures darshan-util 3.1.2 to adapt to many kinds of systems.
 
 Usage: $0 [OPTION]... [VAR=VALUE]...
 
@@ -1309,7 +1309,7 @@ fi
 
 if test -n "$ac_init_help"; then
   case $ac_init_help in
-     short | recursive ) echo "Configuration of darshan-util 3.1.0:";;
+     short | recursive ) echo "Configuration of darshan-util 3.1.2:";;
    esac
   cat <<\_ACEOF
 
@@ -1405,7 +1405,7 @@ fi
 test -n "$ac_init_help" && exit $ac_status
 if $ac_init_version; then
   cat <<\_ACEOF
-darshan-util configure 3.1.0
+darshan-util configure 3.1.2
 generated by GNU Autoconf 2.69
 
 Copyright (C) 2012 Free Software Foundation, Inc.
@@ -1770,7 +1770,7 @@ cat >config.log <<_ACEOF
 This file contains any messages produced by compilers while
 running configure, to aid debugging if configure makes a mistake.
 
-It was created by darshan-util $as_me 3.1.0, which was
+It was created by darshan-util $as_me 3.1.2, which was
 generated by GNU Autoconf 2.69.  Invocation command line was
 
   $ $0 $@
@@ -4117,7 +4117,7 @@ fi
 done
 
 
-DARSHAN_UTIL_VERSION="3.1.0"
+DARSHAN_UTIL_VERSION="3.1.2"
 
 
 
@@ -4633,7 +4633,7 @@ cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1
 # report actual input values of CONFIG_FILES etc. instead of their
 # values after options handling.
 ac_log="
-This file was extended by darshan-util $as_me 3.1.0, which was
+This file was extended by darshan-util $as_me 3.1.2, which was
 generated by GNU Autoconf 2.69.  Invocation command line was
 
   CONFIG_FILES    = $CONFIG_FILES
@@ -4695,7 +4695,7 @@ _ACEOF
 cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
 ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //; s/[\\""\`\$]/\\\\&/g'`"
 ac_cs_version="\\
-darshan-util config.status 3.1.0
+darshan-util config.status 3.1.2
 configured by $0, generated by GNU Autoconf 2.69,
   with options \\"\$ac_cs_config\\"
 

darshan-util/configure.in

@@ -5,7 +5,7 @@ dnl Process this file with autoconf to produce a configure script.
 dnl You may need to use autoheader as well if changing any DEFINEs
 
 dnl sanity checks, output header, location of scripts used here
-AC_INIT([darshan-util], [3.1.0])
+AC_INIT([darshan-util], [3.1.2])
 AC_CONFIG_SRCDIR([darshan-logutils.h])
 AC_CONFIG_AUX_DIR(../maint/config)
 AC_CONFIG_HEADER(darshan-util-config.h)

darshan-util/darshan-convert.c

@@ -360,6 +360,7 @@ int main(int argc, char **argv)
     {
         darshan_log_close(infile);
         darshan_log_close(outfile);
+        unlink(outfile_name);
         return(-1);
     }
 
@@ -368,6 +369,7 @@ int main(int argc, char **argv)
     {
         darshan_log_close(infile);
         darshan_log_close(outfile);
+        unlink(outfile_name);
         return(-1);
     }
 
@@ -389,19 +391,8 @@ int main(int argc, char **argv)
         /* we have module data to convert */
         memset(mod_buf, 0, DEF_MOD_BUF_SIZE);
 
-        ret = mod_logutils[i]->log_get_record(infile, (void **)&mod_buf);
-        if(ret != 1)
-        {
-            fprintf(stderr, "Error: failed to parse the first %s module record.\n",
-                darshan_module_names[i]);
-            darshan_log_close(infile);
-            darshan_log_close(outfile);
-            unlink(outfile_name);
-            return(-1);
-        }
-
         /* loop over each of the module's records and convert */
-        do
+        while((ret = mod_logutils[i]->log_get_record(infile, (void **)&mod_buf)) == 1)
         {
             base_rec = (struct darshan_base_record *)mod_buf;
 
@@ -412,12 +403,22 @@ int main(int argc, char **argv)
                 {
                     darshan_log_close(infile);
                     darshan_log_close(outfile);
+                    unlink(outfile_name);
                     return(-1);
                 }
 
                 memset(mod_buf, 0, DEF_MOD_BUF_SIZE);
             }
-        } while((ret = mod_logutils[i]->log_get_record(infile, (void **)&mod_buf)) == 1);
+        }
+        if(ret < 0)
+        {
+            fprintf(stderr, "Error: failed to parse %s module record.\n",
+                darshan_module_names[i]);
+            darshan_log_close(infile);
+            darshan_log_close(outfile);
+            unlink(outfile_name);
+            return(-1);
+        }
     }
 
     darshan_log_close(infile);

darshan-util/darshan-diff.c

@@ -384,13 +384,11 @@ static int darshan_build_global_record_hash(
             {
                 fprintf(stderr, "Error: unable to read module %s data from log file.\n",
                     darshan_module_names[i]);
-                free(mod_rec->mod_dat);
                 free(mod_rec);
                 return(-1);
             }
             else if(ret == 0)
             {
-                free(mod_rec->mod_dat);
                 free(mod_rec);
                 break;
             }

darshan-util/darshan-logutils.c

@@ -643,6 +643,17 @@ int darshan_log_get_mod(darshan_fd fd, darshan_module_id mod_id,
     if(fd->mod_map[mod_id].len == 0)
         return(0); /* no data corresponding to this mod_id */
 
+    /* assume module will support backwards compatibility, but we obviously
+     * can't provide any sort of "forwards" compatibility
+     */
+    if(fd->mod_ver[mod_id] > darshan_module_versions[mod_id])
+    {
+        fprintf(stderr, "Error: invalid %s module log format version "
+                "(expected %d, got %d)\n", darshan_module_names[mod_id],
+                darshan_module_versions[mod_id], fd->mod_ver[mod_id]);
+        return(-1);
+    }
+
     /* read this module's data from the log file */
     ret = darshan_log_dzread(fd, mod_id, mod_buf, mod_buf_sz);
     if(ret < 0)

darshan-util/darshan-mpiio-logutils.c

@@ -64,7 +64,7 @@ static int darshan_log_get_mpiio_file(darshan_fd fd, void** mpiio_buf_p)
         if(!file)
             return(-1);
     }
-
+    
     ret = darshan_log_get_mod(fd, DARSHAN_MPIIO_MOD, file,
         sizeof(struct darshan_mpiio_file));
 
@@ -165,6 +165,12 @@ static void darshan_log_print_mpiio_description(int ver)
     printf("#   MPIIO_F_*_RANK_TIME: fastest and slowest I/O time for a single rank (for shared files).\n");
     printf("#   MPIIO_F_VARIANCE_RANK_*: variance of total I/O time and bytes moved for all ranks (for shared files).\n");
 
+    if(ver < 2)
+    {
+        printf("\n# WARNING: MPIIO module log format version 1 has the following limitations:\n");
+        printf("# - MPIIO_F_WRITE_START_TIMESTAMP may not be accurate.\n");
+    }
+
     return;
 }
 

darshan-util/doc/darshan-util.txt

@@ -39,7 +39,6 @@ make
 make install
 ----
 
-[NOTE]
 The darshan-util package is intended to be used on a login node or
 workstation.  For most use cases this means that you should
 either leave CC to its default setting or specify a local compiler.  This is
@@ -97,6 +96,12 @@ You can also manually specify the name of the output file using the
 An example of the output produced by darshan-job-summary.pl can be found
 link:http://www.mcs.anl.gov/research/projects/darshan/docs/ssnyder_ior-hdf5_id3655016_9-23-29011-12333993518351519212_1.darshan.pdf[HERE].
 
+*NOTE*: The darshan-job-summary tool depends on a few LaTeX packages that may not
+be availalble by default on all systems, including: lastpage, subfigure, and
+threeparttable. These packages can be found and installed using your system's package
+manager. For instance, the packages can be installed on Debian or Ubuntu systems as
+follows: `apt-get install texlive-latex-extra`
+
 === darshan-summary-per-file.sh
 
 This utility is similar to darshan-job-summary.pl, except that it produces a