update darshan-util to recognize partial log files

darshan-util/darshan-convert.c

@@ -251,7 +251,7 @@ int main(int argc, char **argv)
         return(-1);
  
     comp_type = bzip2 ? comp_type = DARSHAN_BZIP2_COMP : DARSHAN_ZLIB_COMP;
-    outfile = darshan_log_create(outfile_name, comp_type);
+    outfile = darshan_log_create(outfile_name, comp_type, infile->partial_flag);
     if(!outfile)
     {
         darshan_log_close(infile);

darshan-util/darshan-job-summary/bin/darshan-job-summary.pl.in

@@ -36,6 +36,7 @@ my %posix_access_hash = ();
 my %mpiio_access_hash = ();
 my @access_size = ();
 my %hash_files = ();
+my $partial_flag = 0;
 
 # data structures for calculating performance
 my %hash_unique_file_time = ();
@@ -93,10 +94,21 @@ while($line = <PARSE_OUT>)
     {
         if ($line =~ /^# exe: /)
         {
+            $f_save = "";
             ($junk, $cmdline) = split(':', $line, 2);
+
+            print("PRE: $cmdline\n");
             # add escape characters if needed for special characters in
             # command line
-            $cmdline = encode('latex', $cmdline);
+            if ($cmdline =~ /<unknown args>/)
+            {
+                # fortran "<unknown args> seems to throw things off,
+                # so we don't encode that if it's present
+                $f_save = substr($cmdline, -14);
+                $cmdline = substr($cmdline, 0, -14); 
+            }
+            $cmdline = encode('latex', $cmdline) . $f_save;
+            print("POST: $cmdline\n");
         }
         elsif ($line =~ /^# nprocs: /)
         {
@@ -123,6 +135,10 @@ while($line = <PARSE_OUT>)
             ($junk, $version) = split(':', $line, 2);
             $version =~ s/^\s+//;
         }
+        elsif ($line =~ /^# \*WARNING\*: This Darshan log contains incomplete data!/)
+        {
+            $partial_flag = 1;
+        }
     }
     else
     {
@@ -952,12 +968,19 @@ my $latex_cmd_line = "\"\\def\\titlecmd{$cmd} \\
     \\input{summary.tex}\" \\
     @__DARSHAN_PDFLATEX_HALT_ON_ERROR@";
 
+if ($partial_flag == 1)
+{
+    my $partial_log_flags = "\\def\\incompletelog{1} \\";
+    $latex_cmd_line = substr($latex_cmd_line, 0, 1) . $partial_log_flags . substr($latex_cmd_line, 1);
+}
+
 if (defined $summary{MPIIO_INDEP_OPENS})
 {
-    my $mpiio_latex_flags = "\\def\\inclmpiio{1} \\\n";
+    my $mpiio_latex_flags = "\\def\\inclmpiio{1} \\";
     $latex_cmd_line = substr($latex_cmd_line, 0, 1) . $mpiio_latex_flags . substr($latex_cmd_line, 1);
 }
 
+
 $system_rc = system "$pdflatex $latex_cmd_line > latex.output";
 if($system_rc)
 {

darshan-util/darshan-job-summary/share/summary.tex

@@ -6,6 +6,7 @@
 \usepackage{subfigure}
 \usepackage{multirow}
 \usepackage{threeparttable}
+\usepackage{color}
 
 %
 % GET THE MARGINS RIGHT, THE UGLY WAY
@@ -32,6 +33,17 @@
 
 \pagestyle{fancy}
 
+\ifdefined\incompletelog
+\twocolumn[
+\vspace{3.5in}
+\center
+{\bf \textcolor{red}{WARNING}}: This Darshan log contains incomplete data
+which may skew results in this document.
+\endcenter
+]
+\newpage
+\fi
+
 \begin{figure*}[!h]
 \centering
 \subfigure

darshan-util/darshan-logutils.c

@@ -159,7 +159,8 @@ darshan_fd darshan_log_open(const char *name)
  *
  * returns file descriptor on success, NULL on failure
  */
-darshan_fd darshan_log_create(const char *name, enum darshan_comp_type comp_type)
+darshan_fd darshan_log_create(const char *name, enum darshan_comp_type comp_type,
+    int partial_flag)
 {
     darshan_fd tmp_fd;
     int ret;
@@ -188,6 +189,7 @@ darshan_fd darshan_log_create(const char *name, enum darshan_comp_type comp_type
     }
     tmp_fd->state->creat_flag = 1;
     tmp_fd->state->comp_type = comp_type;
+    tmp_fd->partial_flag = partial_flag;
     strncpy(tmp_fd->state->logfile_path, name, PATH_MAX);
 
     /* position file pointer to prealloc space for the log file header
@@ -882,6 +884,7 @@ static int darshan_log_getheader(darshan_fd fd)
     }
 
     state->comp_type = header.comp_type;
+    fd->partial_flag = header.partial_flag;
 
     /* save the mapping of data within log file to this file descriptor */
     fd->job_map.off = sizeof(struct darshan_header);
@@ -913,6 +916,7 @@ static int darshan_log_putheader(darshan_fd fd)
     strcpy(header.version_string, DARSHAN_LOG_VERSION);
     header.magic_nr = DARSHAN_MAGIC_NR;
     header.comp_type = state->comp_type;
+    header.partial_flag = fd->partial_flag;
 
     /* copy the mapping information to the header */
     memcpy(&header.rec_map, &fd->rec_map, sizeof(struct darshan_log_map));

darshan-util/darshan-logutils.h

@@ -27,6 +27,8 @@ struct darshan_fd_s
     /* flag indicating whether byte swapping needs to be
      * performed on log file data */
     int swap_flag;
+    /* flag indicating whether a log file contains partial data */
+    int partial_flag;
     /* log file offset/length maps for each log file region */
     struct darshan_log_map job_map;
     struct darshan_log_map rec_map;
@@ -82,7 +84,8 @@ extern struct darshan_mod_logutil_funcs *mod_logutils[];
 #include "darshan-bgq-logutils.h"
 
 darshan_fd darshan_log_open(const char *name);
-darshan_fd darshan_log_create(const char *name, enum darshan_comp_type comp_type);
+darshan_fd darshan_log_create(const char *name, enum darshan_comp_type comp_type,
+    int partial_flag);
 int darshan_log_getjob(darshan_fd fd, struct darshan_job *job);
 int darshan_log_putjob(darshan_fd fd, struct darshan_job *job);
 int darshan_log_getexe(darshan_fd fd, char *buf);

darshan-util/darshan-parser.c

@@ -321,6 +321,12 @@ int main(int argc, char **argv)
         printf("# mount entry:\t%s\t%s\n", mnt_pts[i], fs_types[i]);
     }
 
+    /* warn user if this log file is incomplete */
+    if(fd->partial_flag)
+        printf("\n# *WARNING*: This Darshan log contains incomplete data!\n"
+               "#            This happens when an application creates\n"
+               "#            more records than Darshan can track.\n");
+
     pdata.rank_cumul_io_time = malloc(sizeof(double)*job.nprocs);
     pdata.rank_cumul_md_time = malloc(sizeof(double)*job.nprocs);
     if (!pdata.rank_cumul_io_time || !pdata.rank_cumul_md_time)