- added CP_SIZE_AT_OPEN metric (records size of file when opened)

- moved sync cost to cumulative write time rather than metadata time
- fixed more bugs in access size table in summary
- added sync to counter graph in summary


git-svn-id: https://svn.mcs.anl.gov/repos/darshan/trunk@144 3b7491f3-a168-0410-bf4b-c445ed680a29

darshan-log-format.h

@@ -178,6 +178,7 @@ enum darshan_indices
     CP_ACCESS3_COUNT,
     CP_ACCESS4_COUNT,
     CP_DEVICE,                     /* device id reported by stat */
+    CP_SIZE_AT_OPEN,
 
     CP_NUM_INDICES,
 };

darshan-logutils.c

@@ -150,6 +150,7 @@ char *darshan_names[] = {
     "CP_ACCESS3_COUNT",
     "CP_ACCESS4_COUNT",
     "CP_DEVICE",
+    "CP_SIZE_AT_OPEN",
 
     "CP_NUM_INDICES"
 };
@@ -178,9 +179,10 @@ char *darshan_f_names[] = {
  *
  * - added:
  *   - CP_DEVICE
+ *   - CP_SIZE_AT_OPEN
  * - changed params:
- *   - CP_FILE_RECORD_SIZE: 1240 to 1244
- *   - CP_NUM_INDICES: 138 to 139
+ *   - CP_FILE_RECORD_SIZE: 1240 to 1248
+ *   - CP_NUM_INDICES: 138 to 140
  */
 #define CP_NUM_INDICES_1_22 138
 struct darshan_file_1_22
@@ -204,11 +206,12 @@ static void shift_missing_1_22(struct darshan_file* file);
  *   - CP_MAX_READ_TIME_SIZE
  *   - CP_MAX_WRITE_TIME_SIZE
  *   - CP_DEVICE
+ *   - CP_SIZE_AT_OPEN
  *   - CP_F_MAX_READ_TIME
  *   - CP_F_MAX_WRITE_TIME
  * - changed params:
- *   - CP_FILE_RECORD_SIZE: 1184 to 1244
- *   - CP_NUM_INDICES: 133 to 139
+ *   - CP_FILE_RECORD_SIZE: 1184 to 1248
+ *   - CP_NUM_INDICES: 133 to 140
  *   - CP_F_NUM_INDICES: 12 to 14
  * - so 60 bytes worth of new indices are the only difference
  */
@@ -483,7 +486,9 @@ void darshan_log_print_version_warnings(struct darshan_job *job)
     {
         printf("# WARNING: version 1.22 log format does not support the following parameters:\n");
         printf("#   CP_DEVICE\n");
-        printf("# It also does not record mounted file systems, mount points, or fs types.\n");
+        printf("#   CP_SIZE_AT_OPEN\n");
+        printf("# It does not record mounted file systems, mount points, or fs types.\n");
+        printf("# It also attributes syncs to cumulative metadata time, rather than cumulative write time.\n");
         return;
     }
 
@@ -496,9 +501,11 @@ void darshan_log_print_version_warnings(struct darshan_job *job)
         printf("#   CP_MAX_READ_TIME_SIZE\n");
         printf("#   CP_MAX_WRITE_TIME_SIZE\n");
         printf("#   CP_DEVICE\n");
+        printf("#   CP_SIZE_AT_OPEN\n");
         printf("#   CP_F_MAX_READ_TIME\n");
         printf("#   CP_F_MAX_WRITE_TIME\n");
         printf("# It also does not record mounted file systems, mount points, or fs types.\n");
+        printf("# It also attributes syncs to cumulative metadata time, rather than cumulative write time.\n");
         printf("#\n");
         return;
     }
@@ -523,6 +530,7 @@ static void shift_missing_1_21(struct darshan_file* file)
         CP_MAX_READ_TIME_SIZE,
         CP_MAX_WRITE_TIME_SIZE,
         CP_DEVICE,
+        CP_SIZE_AT_OPEN,
         -1};
     int missing_f_counters[] = {
         CP_F_MAX_READ_TIME,
@@ -574,6 +582,7 @@ static void shift_missing_1_22(struct darshan_file* file)
     int c_index = 0;
     int missing_counters[] = {
         CP_DEVICE,
+        CP_SIZE_AT_OPEN,
         -1};
 
     c_index = 0;

darshan-parser.c

@@ -81,7 +81,7 @@ int main(int argc, char **argv)
     /* print table of mounted file systems */
     ret = darshan_log_getmounts(file, &devs, &mnt_pts, &fs_types, &mount_count,
         &no_files_flag);
-    printf("\n# mounted file systems (device, fs type, and mount point)\n");
+    printf("\n# mounted file systems (device, mount point, and fs type)\n");
     printf("# -------------------------------------------------------\n");
     for(i=0; i<mount_count; i++)
     {
@@ -135,6 +135,7 @@ int main(int argc, char **argv)
     printf("#   CP_ACCESS*_ACCESS: the four most common access sizes.\n");
     printf("#   CP_ACCESS*_COUNT: count of the four most common access sizes.\n");
     printf("#   CP_DEVICE: device id reported by stat().\n");
+    printf("#   CP_SIZE_AT_OPEN: size of file when first opened.\n");
     printf("#   CP_F_OPEN_TIMESTAMP: timestamp of first open (mpi or posix).\n");
     printf("#   CP_F_*_START_TIMESTAMP: timestamp of first read/write (mpi or posix).\n");
     printf("#   CP_F_*_END_TIMESTAMP: timestamp of last read/write (mpi or posix).\n");

lib/darshan-mpi-io.c

@@ -551,7 +551,7 @@ int MPI_File_sync(MPI_File fh)
         file = darshan_file_by_fh(fh);
         if(file)
         {
-            CP_F_INC(file, CP_F_MPI_META_TIME, (tm2-tm1));
+            CP_F_INC(file, CP_F_MPI_WRITE_TIME, (tm2-tm1));
             CP_INC(file, CP_SYNCS, 1);
         }
         CP_UNLOCK();
@@ -1317,8 +1317,9 @@ static void darshan_file_reduce(void* infile_v,
                 inoutfile->counters[CP_MAX_READ_TIME_SIZE];
         }
 
-        /* pick one device id */
+        /* pick one device id and file size */
         tmp_file.counters[CP_DEVICE] = infile->counters[CP_DEVICE];
+        tmp_file.counters[CP_SIZE_AT_OPEN] = infile->counters[CP_SIZE_AT_OPEN];
 
         /* pick one name suffix */
         strcpy(tmp_file.name_suffix, infile->name_suffix);

lib/darshan-posix.c

@@ -195,6 +195,7 @@ static void cp_access_counter(struct darshan_file_runtime* file, ssize_t size,
         if(fstat(file->fd, &cp_stat_buf) == 0) {\
             CP_SET(file, CP_DEVICE, cp_stat_buf.st_dev); \
             CP_SET(file, CP_FILE_ALIGNMENT, cp_stat_buf.st_blksize); \
+            CP_SET(file, CP_SIZE_AT_OPEN, cp_stat_buf.st_size); \
         }\
     }\
     file->log_file->rank = my_rank; \
@@ -326,7 +327,7 @@ int __wrap_fsync(int fd)
     file = darshan_file_by_fd(fd);
     if(file)
     {
-        CP_F_INC(file, CP_F_POSIX_META_TIME, (tm2-tm1));
+        CP_F_INC(file, CP_F_POSIX_WRITE_TIME, (tm2-tm1));
         CP_INC(file, CP_POSIX_FSYNCS, 1);
     }
     CP_UNLOCK();
@@ -350,7 +351,7 @@ int __wrap_fdatasync(int fd)
     file = darshan_file_by_fd(fd);
     if(file)
     {
-        CP_F_INC(file, CP_F_POSIX_META_TIME, (tm2-tm1));
+        CP_F_INC(file, CP_F_POSIX_WRITE_TIME, (tm2-tm1));
         CP_INC(file, CP_POSIX_FDSYNCS, 1);
     }
     CP_UNLOCK();
@@ -1078,6 +1079,7 @@ void darshan_condense(void)
 
                 /* pick one */
                 case CP_DEVICE:
+                case CP_SIZE_AT_OPEN:
                     CP_SET(base_file, i, CP_VALUE(iter_file, i));
                     break;
 

util/bin/darshan-job-summary.pl

@@ -97,6 +97,7 @@ while ($line = <TRACE>) {
 open(COUNTS, ">$tmp_dir/counts.dat") || die("error opening output file: $!\n");
 print COUNTS "# P=POSIX, MI=MPI-IO indep., MC=MPI-IO coll., R=read, W=write\n";
 print COUNTS "# PR, MIR, MCR, PW, MIW, MCW, Popen, Pseek, Pstat\n";
+my $total_syncs = $summary{CP_POSIX_FSYNCS} + $summary{CP_POSIX_FDSYNCS};
 print COUNTS "Read, ", $summary{CP_POSIX_READS}, ", ",
     $summary{CP_INDEP_READS}, ", ", $summary{CP_COLL_READS}, "\n",
     "Write, ", $summary{CP_POSIX_WRITES}, ", ", 
@@ -105,7 +106,8 @@ print COUNTS "Read, ", $summary{CP_POSIX_READS}, ", ",
     $summary{CP_COLL_OPENS}, "\n",
     "Stat, ", $summary{CP_POSIX_STATS}, ", 0, 0\n",
     "Seek, ", $summary{CP_POSIX_SEEKS}, ", 0, 0\n",
-    "Mmap, ", $summary{CP_POSIX_MMAPS}, ", 0, 0\n";
+    "Mmap, ", $summary{CP_POSIX_MMAPS}, ", 0, 0\n",
+    "Fsync, ", $total_syncs, ", 0, 0\n";
 close COUNTS;
 
 # histograms of reads and writes
@@ -259,8 +261,8 @@ access size \& count \\\\
 
 # sort access sizes (descending)
 my $i = 0;
-foreach $value (sort {$access_hash{$b} cmp $access_hash{$a} } keys %access_hash) {
-    if($i == 3) {
+foreach $value (sort {$access_hash{$b} <=> $access_hash{$a} } keys %access_hash) {
+    if($i == 4) {
         last;
     }
     if($access_hash{$value} == 0) {