-
Notifications
You must be signed in to change notification settings - Fork 29
/
ChangeLog
860 lines (797 loc) · 39.8 KB
/
ChangeLog
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
--------------------------
Darshan Release Change Log
--------------------------
Darshan-3.4.6
=============
* Added enhancements to Darshan's Lustre instrumentation module to capture
extensive details on the Lustre stripe components constituting a file (e.g.,
in a progressive file layout)
- File records now composed of potentially multiple Lustre component counter
sets (e.g., LUSTRE_COMP1_*, LUSTRE_COMP2_*, etc.)
- STRIPE_OFFSET, OSTS, and MDTS counters are no longer provided
- STRIPE_WIDTH is now referred to as STRIPE_COUNT to match Lustre terminology
* Fixed bugs in Darshan's log compression/decompression routines that are
triggered when using the new zlib-ng software package
- darshan-runtime bug fix to correct problematic log compression strategy
- darshan-util bug fix to correct logs already generated with this issue
* Fixed bug leading to hangs when passing in unproperly formatted Darshan
runtime configuration settings
Darshan-3.4.5
=============
* Added the ability for the Darshan runtime library to properly shutdown for
non-MPI applications that call `_exit()` directly
- This behavior has been commonly observed in the Python `multiprocessing`
package, which has traditionally prevented Darshan from properly
instrumenting applications that use it (e.g., the PyTorch DataLoader)
* Added optional integration with the LDMS data/metrics collection system,
allowing realtime analysis of Darshan instrumented I/O operations.
- For more details, see https://ovis-hpc.readthedocs.io/en/latest/ldms/ldms-streams.html#darshan
- Contributed by Sara Walton (SNL), Ana Luisa Solorzano (Northeastern), and the LDMS team
* Fixed bug to ensure Darshan includes the appropriate instrumentation wrapper
for `fscanf`/`__isoc99_fscanf`
* Fixed bug in HDF5 module causing any call to HDF5's `H5Pset_fapl_mpio()`
routine to fail
Darshan-3.4.4
=============
* Fixed bug leading to inconsistent heatmap record shapes when Darshan
shared file reductions are disabled
- also added a darshan-util fix to resolve this inconsisency on already
impacted logs (any generated with 3.4.0+ versions of Darshan)
* Added workaround for potential undefined symbol errors for 'H5FD_mpio_init'
when LD_PRELOADing an HDF5-enabled runtime library
- bug triggered by 1.13+ versions of HDF5
Darshan-3.4.3
=============
* Added new configure option '--with-username-env' to allow
specification of an env variable to use to find the username
associated with a job (e.g., SLURM_JOB_USER)
* Fixed bug causing crashes for applications that call fork()
and use Darshan app exclusions settings
* Fixed bug related to not closing open HDF5 file ID when
instrumenting H5Fflush() calls (reported/fixed by Zanhua
Huang)
Darshan-3.4.2
=============
* Fixed segfault when defining scalar variables in PnetCDF module
* Fixed memory corruption (and potential segfault) when
reading/writing high-dimensional PnetCDF variables using
vara/vars/varm interfaces
* Fixed crashes related to using PnetCDF vard interfaces with input
MPI_DATATYPE_NULL datatypes
* Fixed bug attributing all PnetCDF variable instrumentation to the
first variable instrumented
Darshan-3.4.1
=============
* Added comprehensive instrumentation of PnetCDF APIs via PNETCDF_FILE and
PNETCDF_VAR modules (contributed by Wei-Keng Liao)
- disabled by default, enabled by passing `--enable-pnetcdf-mod` to
configure
* Modified Darshan log format to support a max of 64 instrumentation
modules, since the current version of Darshan reached the old max (16)
* Modified Darshan to report job start/end times at nanosecond
granularity (previously only second granularity was possible)
* Added support for instrumenting H5Oopen family of calls
* Modified HDF5 module extraction of dataspase selection details
- Extraction of point selections now possible regardless of HDF5 version
- H5S_ALL selections are no longer counted as regular hyperslab accesses
* Fixed bug causing no instrumentation of child processes of fork() calls
(reported by Rui Wang)
* Deprecated --file-list and --file-list-detailed options in darshan-parser
* Added "darshan_accumulator" API to the logutils library
- _create(), _inject(), _emit(), and _destroy()
- generalizes the mechanism for producing summation records and derived
metrics for sets of records from a given module
- refactored darshan-parser to use new API
- implemented support for accumulators in POSIX, STDIO, and MPIIO modules
* Fixed memory leak in darshan-util helper functions used by PyDarshan
- darshan_log_get_name_records
- darshan_log_get_filtered_name_records
* Integrated the µnit Testing Framework in darshan-util
- implemented unit tests for darshan_accumlator API
Darshan-3.4.0
=============
* Fix segfault affecting new DARSHAN_MOD_DISABLE/ENABLE environment variables
* Fix divide-by-zero condition that can potentially be triggered by new
heatmap module
* Fix potential MPI errors related to calling MPI_Type_size() on a
user-supplied MPI_DATATYPE_NULL type (reported by Jim Edwards)
* cuserid() is no longer the default method for determining username, and must
be manually enabled at configure time
* Fix backwards compatibility bug affecting darshan-3.0.0 logs in darshan-util
C library functions used by PyDarshan
* Suppress noisy output warnings when using darshan-job-summary.pl
* Clarify units displayed by darshan-job-summary.pl (reported by Jeff Layton)
Darshan-3.4.0-pre1
==================
* Add a new "heatmap" module that records histograms of I/O traffic over time
on each process
* Performance optimizations to runtime wrappers to limit function call
overhead and locking costs
* Replace mutex locks with atomics in core library
* Add optional, limited support for RDTSCP-based timing, enabled via
--enable-rdtscp at configure time
* Remove unecessary explicit mpich binding libraries from link line when static linking
* Graceful warnings when parsing log files with unknown module types
* Remove deprecated performance estimate methods from darshan-parser
* Add two new derived metrics to darshan-parser --perf: agg_time_by_slowest
(total elapsed time performing I/O by the slowest rank) and
slowest_rank_rw_only_time (total elapsed time performing read/write
operations by the slowest rank)
* Add unified Darshan configuration support via env vars and config files
* Adopt automake and libtool for Darshan build, and allow top-level build
of both darshan-runtime and darshan-util (contributed by Wei-Keng Liao)
* Increase default memory limit for record names (NAMEMEM config setting)
to 1 MiB
Darshan-3.3.1
=============
* fixed github issue 399: undefined llapi_layout_get_by_xattr function in some
scenarios because of incomplete dependency information in Darshan shared
library
* modified Darshan to properly handle for non-MPI applications that call fork()
Darshan-3.3.0
=============
* added AutoPerf modules for tracking MPI communication usage (APMPI) and
for tracking Cray XC performance counters (APXC)
Darshan-3.3.0-pre2
=============
* fixed bug in HDF5 module resulting in warning messages for instrumented
datasets that use H5S_SELECT_NONE (reported by Marta Garcia Martinez)
* disabled DXT's MPI-IO offset tracing for Darshan builds using OpenMPI
to workaround an OpenMPI bug (reported by Tobias Meisel)
* modified Darshan to query Lustre striping information for all
instrumented files to workaround issue of retrieving striping info for
Lustre files that are symlinked to other file systems
Darshan-3.3.0-pre1
=============
* added first version of PyDarshan, which provides interfaces for
accessing Darshan log file data from Python applications
* modified Lustre module to use 'fgetxattr' rather than 'ioctl' for
querying stripe info to avoid known bug in Lustre that could cause
application crashes
* fixed bug leading to potential deadlock when reducing shared file records
for MPI applications (known to affect mvapich2)
* fixed bug preventing usage of Darshan's non-MPI mode in builds that use
MPI compilers
* boosted Darshan's default internal memory allocation from 2 MiB to 4 MiB
* added wrappers for openat, preadv, preadv2, pwritev, and pwritev2 (improves
profiling of ompio)
* various bug fixes to enable usage of HDF5 modules in non-MPI applications
* fixed HDF5 module bug leading to warnings on stdout for applications using
non-MPIIO VFD for HDF5
* fixed HDF5 module bug leading to potentially incorrect values in H5D module
common access counters
* added new H5D record value to indicate corresponding H5F record ID, to
allow easier mapping of HDF5 datasets to files they belong to
* improved error handling in command line parsers to make incomplete logs more
obvious to users
* fixed bug leading to incorrect internal timing information at Darshan
shutdown time
* fixed bug causing negative Darshan counter values from being plotted when
using the darshan-job-summary tool
Darshan-3.2.1
=============
* fixed bug causing corruption of Darshan common access
counters (POSIX, MPI-IO, and HDF5 modules) and potentially
other preceding counters (reported by Gerd Heber)
Darshan-3.2.0
=============
* added detailed instrumentation of HDF5 library file and dataset
interfaces
* added new logutil functions: darshan_log_get_modules,
darshan_log_get_name_records, and darshan_log_get_record (contributed
by Jakob Luettgau)
* extended DXT MPIIO module to trace file offsets
* added `darshan-config --log-path` command option to display the log path(s)
configured for use with Darshan
* silenced stdout from darshan-job-summary.pl and remove --summary option
* removed Darshan stubs library traditionally used in static linking test
cases and other general simplifications of Darshan's static linking options
* fixed bug related to multiple definitions of fscanf symbols when building
Darshan (reported by Francois Trahay)
* updated Darshan's compiler wrapper generators, MPICH profiling config
hooks, and Cray software modules to explicitly link in Darshan in
dynamic linking cases using rpath
* reimplemented Darshan's MPI/PMPI wrapper macros (using functiona aliases)
to avoid potential deadlock with other PMPI tools
* made use of DESTDIR match standard automake practice (contributed by
Christopher Morrone)
* made "--with-mem-align" option optional, with a default of 8 (contributed
by Christopher Morrone)
* corrected pkgconfig path for darshan-util when using DESTDIR (reported by
Matthew Curry)
* added support for non-MPI instrumenation to Darshan, using GCC
construct/destructor attributes for bootstrapping rather than
intercepting MPI_Init()/MPI_Finalize(). This mechanism only
works for Darshan's shared library instrumentation (implemented with the
help of Glenn Lockwood)
Darshan-3.1.8
=============
* added dynamic trace triggering functionality, allowing Darshan to
decide at runtime whether or not to trace specific files with DXT,
with this decision based on triggers in a user-provided config file
Darshan-3.1.7
=============
* bug fix for handling DXT data in darshan-convert utility. Bug reported
by Mahzad Khoshlessan.
* bug fix for darshan-parser backwards compatibility: Darshan logs generated
by Darshan versions prior to 3.1.0 may have included stdio counters that
were not properly up-converted. Reported by Teng Wang.
* bug fix to MiB reported in I/O performance estimate of
darshan-job-summary.pl when both posix and stdio access is present
* added wrapper for __open_2(), bug reported by Cormac Garvey in which open
calls are not intercepted with some versions of glibc/gcc
* added an instrumentation module for the MDHIM key/val storage system
* added support for properly instrumenting dup(), fileno(), fdopen(), and
rename() calls
Darshan-3.1.6
=============
* bug fix for linker failures caused when linking external libraries
that use MPI internally (e.g., PETSc). Reported by Bilel Hadri.
* bug fix in mapping of Darshan's MPI_File_read_all wrapper to the
underlying MPI library call in dynamic linking case
Darshan-3.1.5
=============
* modify Darshan MPI instrumentation method to intercept both MPI and
PMPI symbols to workaround MPI implementations that are calling
PMPI routines directly (e.g., the Fortran bindings of OpenMPI2).
Contributed in part by Chris Zimmer.
* add a new python utility for analzying DXT trace files (dxt_analyzer).
Contributed by Alex Sim.
* bug fix to disable Darshan module instrumentation after the
Darshan shutdown procedure has began, an issue that was
leading to negative timers in some cases
* bug fix for autoconf ignoring specified libbz2 location. Contributed
by Glenn Lockwood.
* add regression tests harnesses for Cray systems at ALCF & NERSC
* add support for DARSHAN_EXCLUDE_DIRS environment variable to
explicitly disable instrumentation for files in given
directories. Contributed by Cristian Simarro.
* correct "undefined reference to `__wrap_H5get_libversion'" linker failure
when compiling some HDF5 programs, reported by Jialin Liu
* bug fix in darshan-merge utility related to logs containing DXT trace
data. Reported by Glenn Lockwood.
Darshan-3.1.4
=============
* bug fix to prevent darshan-parser segfault when parsing logs with
DXT module data present. Reported by James Dickson.
* bug fixes to make the darshan-diff utility functional again.
Darshan-3.1.3
=============
* disable instrumentation for mmap when dynamically linking; this avoids a
potential deadlock condition on Cray systems using dynamically linked
executabes. Reported by Cristian Simarro for dynamic linking case.
Darshan-3.1.3-pre1
=============
* add new DxT instrumentation modules to provide fine-grained read/write
operation tracing at both the POSIX and MPI-IO layers
- this functionality should be enabled at runtime by exporting the
DXT_ENABLE_IO_TRACE environment variable
- trace output is stored within Darshan's traditional log file format
- a corresponding trace parser (darshan-dxt-parser) is offered within
darshan-util to allow the DxT trace modules to be parsed and displayed
- this software was contributed by Cong Xu and Intel's HPDD division.
* add logic to allow Darshan to capture command line arguments from Fortran
applications (contributed by Cristian Simarro)
* skip instrumentation attempts for anonymous mmap() calls; this avoids a
potentential deadlock condition when used with hugepages on Cray systems.
Reported by Glenn Lockwood for static linking case.
* fix segmentation fault in statistics collection for applications that issue
operations with a large number of distince access sizes or strides on the
same file. Reported by Glenn Lockwood.
* disable HDF5 module by default unless enabled using --enable-HDF5-post-1.10
or --enable-HDF5-pre-1.10 configure arguments. These options
vary the wrapper prototypes to match the corresponding HDF5 library ABI.
The initial patch for HDF5 1.10 compatibility was contributed by
Karl-Ulrich Bamberg.
* modified Darshan's path exclusion logic to include a whitelist to prevent
I/O to/from Cray's Datawarp service from being filtered out (it is located
within the /var directory, which historically has been excluded). Reported
by Glenn Lockwood.
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, & 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)
* add new mmap-based logging mechanism that allows Darshan to
generate output logs even in cases where applications don't
call MPI_Finalize()
- these logs are uncompressed and are per-process rather
than per-job
* add the darshan-merge utility to darshan-util to allow per-process
logs generated by the mmap-based logging mechanism to be converted
into Darshan's traditional compressed per-job log files
* augment the POSIX module timestamp counters to also include a
LAST_OPEN & FIRST_CLOSE counters to give more details on application
I/O intervals
* avoid saving duplicate mount point entries in Darshan log files
Darshan-3.0.1
=============
* bug fix in darshan logutil mount parsing code that was
causing file paths to be matched to the first mount point
with a common prefix rather than the one with the longest
common prefix
* bug fix in the darshan-util bzip2 configure check that
was accidentally overwriting Darshan's LDFLAGS
* minor bug fixes to IO start time counters in all modules
to set IO start time to the actual first start time rather
than the first IO op to complete
* update darshan-util perl scripts to get perl bin from
user's path, rather than from /usr/lib (reported by
Kay Thust)
* update Darshan's fortran and cxx compiler wrapper generators
to automatically detect MPICH library names on BG/Q
* fix bug that was calculating Darshan's agg_perf_by_slowest
performance metric incorrectly
* add performance estimate info to darshan-job-summary
graphs
Darshan-3.0.0
=============
* install darshan-null-log-format.h header when installing
darshan-util component, otherwise compiler errors are
generated when building external tools that use
libdarshan-util
* update docs to give debugging tips for cases where
Darshan logs are not generated
* fix shared library regression test script to check for
potential errors with Darshan symbols rather than
failing silently in these cases
* bug fix for determining minimum non-zero counters in
shared file reductions in all modules
* loosen Darshan's PMPI symbol check to prevent inadvertent
disabling of Darshan for some MPICH builds
* update runtime docs to give information on upgrading Darshan
* bug fix for resolving MPI_Gather and MPI_Barrier when LDPRELOADing
Darshan's shared libraries (reported by Richard Hedges and Rob Latham)
* add more helpful error handling when opening 2.x version log files
* port darshan-diff utility over to new log file format
* fix numerous configure bugs on Cray systems
* add synthetic benchmarking hooks for testing Darshan's shutdown
procedure
Darshan-3.0.0-pre3
=============
* add module-specific version fields to header to allow utilities
to handle different versions of a module's I/O data for backwards
compatibility -- NOTE: this breaks the log file parsing for logs
obtained using Darshan-3.0.0-pre2 & Darshan-3.0.0-pre1
* bug fix in regression test scripts for setting proper environment
variables to use MPI profiling configuration for Fortran apps
* bug fix in bzip2 log writing implementation in darshan-logutils
* possible race conditions resolved in each module's shutdown code
* general code, comment, and documentation cleanup
* addition of module-specific counter descriptions printed prior
to parsing a modules I/O data in darshan-parser
Darshan-3.0.0-pre2
=============
* add fix to install appropriate headers for linking external
applications with darshan-util (reported by Matthieu Dorier)
* add darshan-util Ruby bindings for the new modularized version
of Darshan (3.0) (Matthieu Dorier)
* add enhancement to darshan-runtime to allow per-module instrumentation
memory to be user configurable using a configure option or a runtime
environment variable
Darshan-3.0.0-pre1
=============
* new version of Darshan with the following features/improvements:
- hooks for developers to add their own instrumentation module
implementations to capture new I/O characterization data
- these instrumentation modules can be used to instrument new
I/O interfaces or gather system-specific parameters, for instance
- modularized log format allows new module-specific utilities to
access their I/O characterization data independently
- this new format also allows new counters to be added to existing
instrumentation modules without breaking existing utilities
- Darshan logs now contain a mapping of Darshan's unique record
identifiers to full file names, instead of fix-sized file name
suffixes
- a new instrumentation module for capturing BG/Q-specific parameters
(BG/Q environment is automatically detected at configure time)
(implemented by Kevin Harms)
- new darshan-parser and darshan-job-summary output to utilize the
new modularized log format
* updated documentation outlining changes in this release, as well as
steps for adding new instrumentation modules is given in the top-level
'doc' directory.
- documentation for configuring and using the darshan-runtime and
darshan-util components are mostly the same and still located in
their respective directories ('darshan-runtime/doc' and 'darshan-util/doc')
darshan-2.3.2-pre1
=============
* Fix gnuplot version number check to allow darshan-job-summary.pl to work
with gnuplot 5.0 (Kay Thust)
* Fix function pointer mapping typo in lio_listio64 wrapper (Shane Snyder)
* Fix faulty logic in extracting I/O data from the aio_return
wrapper (Shane Snyder)
* Fix bug in common access counter logic (Shane Snyder)
* Expand and clarify darshan-parser documentation (Huong Luu)
darshan-2.3.1
=============
* added documentation and example configuration files for using the -profile
or $MPICC_PROFILE hooks to add instrumentation to MPICH-based MPI
implementations without generating custom wrapper scripts
* Add wrappers for mkstemp(), mkostemp(), mkstemps(), and mkostemps()
(reported by Tom Peterka)
* Change OPEN_TIMESTAMP field to report timestamp right before open() is
invoked rather than after timestamp after open is completed.
NOTE: updated log format version to 2.06 to reflect this change.
* Change start_time and end_time fields in job record to use min and max
(respectively) across all ranks
* Fix bug in write volume data reported in file system table in
darshan-job-summary.pl (reported by Matthieu Dorier)
* Clean up autoconf test for zlib and make zlib mandatory (reported by Kalyana
Chadalavada)
* add --start-group and --end-group notation to Darshan libraries for Cray PE
2.x environment to fix link-time corner cases (Yushu Yao)
* improve y axis labels on time interval graphs in darshan-job-summary.pl
(reported by Tom Peterka)
* misc. improvements to darshan-parser --perf output (reported by Shane
Snyder)
- indicate which rank was slowest in unique file results
- label I/O vs. meta time more clearly
- include unique file meta time in agg_perf_by_slowest calculation
* added regression test script framework in darshan-test/regression/
- currently support platforms include:
- Linux with static linking and generated compiler wrappers
- Linux with static linking and profiler configuration files
- Linux with dynamic linking and LD_PRELOAD
- Blue Gene/Q with static linking and profiler configuration files
* update darshan-gen-fortran.pl and darshan-gen-cxx.pl to support new library
naming conventions in MPICH 3.1.1 and higher
* update documentation to reflect known issues with some versions of MPICH
* Cray platforms: modify darshan-runtime so that link-time instrumentation
options are only used when statically linking via Libs.private.
(reported by Kalyana Chadalavada)
darshan-2.3.0
=============
* Fix incorrect version numbering in darshan-runtime component of Darshan
2.2.9, reported by Jean-Guillaume Piccinali
darshan-2.2.9
=============
* Bug fixes:
- Fix mnt table overflow if a large number of file systems are mounted,
reported by David Shrader.
- Fix argument parsing for darshan-convert, reported by Mouhamed Gueye.
- Fix metadata annotation overflow in darshan-convert, reported by
Mouhamed Gueye.
- Fix const-correctness in dynamic library when built against MPI 3.x
libraries.
- Fix "undefined symbol: dlsym" error when using preloaded dynamic library
on some platforms, reported by Florin Isaila.
- Normalize timestamps to always be relative to MPI_Init().
- Better library name matching in compiler wrappers to handle more MPICH
variations on Blue Gene systems.
- Fixed header dependency problems when using darshan-logutils.h in external
programs (Shane Snyder).
* Enhancements:
- Add support (both in documentation and in provided module files) for
Cray PE 2.x.
- Honor CC variable to allow darshan-util to be built with other compilers
besides gcc.
- Update documentation to clarify the interaction between rpath and
LD_PRELOAD when instrumenting dynamic libraries, issue reported and
investigated by Davide Del Vento.
- Ability to disable shared-file reduction by setting the
DARSHAN_DISABLE_SHARED_REDUCTION environment variable.
- More thorough output from darshan-parser --perf, suggested by Huong Luu.
- Increased metadata annotation room from 64 bytes to 1KiB in header.
- CP_F_{FASTEST/SLOWEST}_RANK_TIME counters now take MPI-IO time into
account, not just POSIX time, issue reported by Huong Luu.
- Better handling of systems with many mounted file systems (after which
point Darshan will assume file resides on / file system), issue reported
and investigated by David Shrader:
- Track up to 64 rather than 32 mounted file systems at runtime.
- Increase header space available for storing mount point information in
log file from approximately 1 KiB to approximately 3 KiB.
- Prioritize storing information about non-NFS volumes over NFS volumes
if too many file systems are mounted to record them all.
- Added darshan-util pkgconfig file (Shane Snyder).
- Added --enable-shared configure option to darshan-util to build and
a shared library version of libdarshan-util
darshan-2.2.8
=============
* WARNING: please note that the Darshan module file for Cray environments has
been updated, especially in the DARSHAN_POST_LINK_OPTS variable. Please
update your module file accordingly when upgrading from 2.2.7 or earlier
on Cray platforms.
* Improved ability to analyze I/O activity related to particular files
opened by an application
- darshan-summary-per-file.sh script to generate a separate pdf summary for
each file opened by an application. Developed by Rob Latham.
- Added --file-list and --file-list-detailed options to darshan-parser to
list files opened by an application along with brief statistics
- Added --file option to darshan-convert to filter out activity for a
specific file from a Darshan log
* Add wrappers for POSIX AIO operations (fixes tracking of underlying POSIX
operations resulting from nonblocking MPI-IO operations. Bug reported by
David Shrader.)
* Fix compile-time errors when Darshan is configured for use with MPICH 3.x
installations with const correct MPI function prototypes. Also affects
MPICH 1.5+ installations with optional const support. Reported by Yushu Yao.
* Fix segmentation fault when using LD_PRELOAD instrumentation on programs
that use MPI_Init_thread() rather than MPI_Init(). Reported by Myriam
Botalla.
darshan-2.2.7
=============
* Updated Cray installation documentation for cleaner integration in Cray
environments
* Fix bug that recorded incorrect device ID (and therefore incorrect mount
point mapping) if stat() was called before open() on a file
* Store version number of the darshan runtime library in the log file
metadata (see lib_ver in darshan-parser output)
darshan-2.2.6
=============
* Bug fixes:
- make sure to honor user-specified hints passed in at runtime
via the DARSHAN_LOGHINTS env variable.
- include fread and fwrite in darshan-job-summary.pl read and write counts
- fix segmentation fault on invalid arguments to darshan-parser
* Performance enhancements:
- collect mount point information at rank 0 and broadcast to all
processes to avoid excess file system traffic on startup
- change default MPI-IO hints for writing log file to
romio_no_indep_rw=true and cb_nodes=4 to improve log creation performance
* Install libdarshan-util and headers during installation process for
darshan-util; patch contributed by Matthieu Dorier
* Detect PMPI support at link time when using compiler scripts produced
by the darshan-gen-* utilities. This avoids link problems when Darshan
compiler wrappers are used with the ADIOS dummy MPI library; reported
by Jingqing Mu
* Support MPICH_{CC/CXX/F77} environment variables in compiler scripts
produced by the darshan-gen-* utilities
* Add instrumentation of startup time, enabled via DARSHAN_INTERNAL_TIMING
environment variable
* Rename cp-shutdown-bench test utility to darshan-shutdown-bench and enable
benchmarking hooks in library by default so that darshan-shutdown-bench
can be used with any Darshan installation
* Remove deprecated --enable-st-dev-workaround configure option
* Misc. documentation updates
darshan-2.2.5
=============
* Fix bug in mount point identification when --enable-stat-at-open option is
not used. In Darshan 2.2.4, some file entries were recorded as using the "/"
file system regardless of their location.
* Update patches and documentation for Cray xt-asyncpe environment 5.12 or
higher; contributed by Yushu Yao.
* Update documentation to no longer recommend --enable-stat-at-open option on
any systems
* Fix unresolved symbols error when linking non-MPI programs with Cray
compiler. Reported by Yushu Yao.
* Use fstat() in place of stat() for detecting file information when
--enable-stat-at-open option is used
* Use statfs() at startup to detect default block size for supported file
systems
* Added --enable-group-readable-logs configure option, which will cause
Darshan to to generate log files with the group read permission bit set.
This option is useful in conjunction with deployments that set the setgid
bit on log directories.
darshan-2.2.4
=============
* Disable extra stat() of newly opened files by default. This improves
performance on shared files for some platforms. Reported by Yushu Yao.
- prevents Darshan from reporting SIZE_AT_OPEN, FILE_ALIGNENT, and
DARSHAN_FILE_NOT_ALIGNED values
- features can still be enabled via --enable-stat-at-open option
* Fix missing -lz in post ld flags reported by Yushu Yao. Fixes a link-time
error for some corner-case applications.
* Fix bug in Cray compiler script patches that was setting compiler flags
incorrectly. Reported by Yushu Yao.
WARNING: if you are using a previous Darshan release (2.2.3 through
2.2.4-pre4) on a Cray platform, please re-patch your compiler scripts.
* Update darshan-gen-* scripts to support the potential for additonal LDFLAGS
or CFLAGS that appear before the object or source file in the compile or
link commands. This fixes compatibility with some mvapich2 installations,
reported by Dragos Constantin.
darshan-2.2.3
=============
* improved Cray XE6 support
- support for GNU, PGI, Cray, Pathscale, and Intel compilers
- patch adding Darshan capability to system compiler scripts
- software module, including testing and features contributed
by Yushu Yao and Katie Antypas of The National Energy Research
Scientific Computing Center (NERSC)
- improved documentation
* minor bug fixes:
- properly detect cxx library name when generating BG/Q compiler wrappers
- improve timing mechanism for cp-shutdown-bench benchmark
- improve hashing to avoid log file name collisions
darshan-2.2.2
=============
* bug fixes:
- remove debugging message that was inadvertently included in
MPI_File_sync() wrapper
- fix potential hang if the --with-log-path-by-env argument was used at
configure time but the environment variable was not set at run time
darshan-2.2.1
=============
* significant improvements to how counters are handled in multi-threaded
applications
* initial (rough) documentation for using Darshan in Cray
programming environments with static linking
* bug fixes:
- escape special characters in mount point paths in darshan-job-summary.pl
(reported by Mouhamed Gueye)
- workarounds for various runtime problems with cuserid() and stat() in
Cray environments
- build problems with darshan-utils on some versions of OSX
- accurate shared file statistics for libraries that use deferred opens
darshan-2.2.0
=============
* split darshan into separate packages:
- darshan-runtime: for runtime instrumentation
- darshan-utils: for processing darshan log files
* changed default output file name for darshan-job-summary.pl to be based on
input file name rather than summary.pdf
* reorganized init and finalize routines so that can be linked separately (to
allow for easier integration with other instrumentation tools)
* add -cc, -cxx, -f77, -f90, and -fc support to compiler scripts generated by
the darshan-gen-*.pl scripts
* bug fixes:
- potential MAX_BYTE overflow on 32 bit systems
- incorrect pread and pwrite offset tracking
- corrections to darshan-job-summary variance table
- better runtime error handling if bzip or gnuplot tools are insufficient
- improvements to time range in darshan-job-summary graphs
* documentation:
- improved documentation for both the darshan-runtime and darshan-util
portions of Darshan can be found in the respective doc/ subdirectory for
each
darshan-2.1.2
=============
* improved error handling when writing log files. If a write fails then the
log file will be deleted and a warning will be printed to stderr.
darshan-2.1.1
=============
* new darshan-convert command line utility for converting existing log files,
with optional anonymization and optional bzip2 compression
* bzip2 support in command line utilities (but not in the darshan library
itself)
* updated log file format that allows for string key/value pairs to be stored
in the header
* added ability to set MPI-IO hints when writing darshan log
at configure time: --with-log-hints
at run time: DARSHAN_LOGHINTS environment variable
* bug fix contributed by Sandra Schröder: use case-insensitive search for MPI
symbols in Fortran wrapper script
* performance bug fix: remove unecessary call to MPI_File_set_size when
writing log
* added --with-logpath-by-env configure option to allow absolute
log path to be specified via environment variable
darshan-2.1.0
=============
* additinoal environment variables to control log, jobid and
alignment parameters
* additional configure tests to improve portability
* bug fixes for darshan-parser --perf calucations
* support for MPI1.x
* support for OpenMPI
* support for PGI, Intel compilers
* new libdarshan.so dynamic library for use with LD_PRELOAD
darshan-2.0.2
=============
* added a random identifier to job logs (to avoid collisions from multiple
application instances within a single scheduler job)
* improved installation and library path management for darshan-job-summary.pl
* improved error handling in darshan-job-summary.pl
* additional derived statistics categories for darshan-parser output:
--all : all sub-options are enabled
--base : darshan log field data [default]
--file : total file counts
--perf : derived perf data
--total : aggregated darshan field data
darshan-2.0.1
=============
* bug fix to variance/minimum calculations on shared files
* switch to automatic generation of all MPI compiler scripts using
darshan-gen-* tools
* new run time environment variable: DARSHAN_INTERNAL_TIMING. If set at job
execution time, it will cause Darshan to time its own internal data
aggregation routines and print the results to stdout at rank 0.
darshan-2.0.0
=============
* new output file format that is portable across architectures
NOTE: Darshan 1.x output files are incompatible with the tools in this
release unless they were generated on a ppc32 architecture (Blue Gene)
* 8 new counters that record the rank of the fastest and slowest process that
opened each shared file, along with the number of seconds and number of
bytes consumed by those processes. It also reports the variance in both
time and amount of data.
* new --with-jobid-env configure argument to support recording job identifiers
from different schedulers
* job ID is now recorded within the Darshan log in addition to in the file
name
* darshan-job-summary.pl:
* opens output files directly without using intermediate darshan-parser output
* table showing data usage per file system
* table showing I/O variance in shared files
* Fixes for bugs reported by Noah Watkins:
* avoid name collision in hashing function
* divide by zero error in darshan-job-summary.pl
darshan-1.1.14
==============
* fixed erroneous incompatibility warning when opening old logs in darshan-parser
darshan-1.1.13
==============
* updated compiler scripts to fix link order in some cases with -lhdf5
darshan-1.1.12
==============
* improved mapping of file records to mount points
* new page in darshan-job-summary.pl output showing timelines of file access
* bug fix to "fast" mpif90 script
* checkpoint/restart ability in parallel fsstats script
darshan-1.1.11
==============
* Bug fix for lseek, pread, and pwrite when used in 32 applications without large file support
* Improved experimental darshan-gen-cc.pl script
* Added new experimental darshan-gen-cxx.pl script
darshan-1.1.10
==============
* Bug fix for files that are accessed with stat() but never opened
* Workaround zlib problems with 64 bit offsets on 32 bit architectures
darshan-1.1.9
=============
* Added "fast" version of each BG/P compiler
* Added experimental scripts in test directory to run fsstats in parallel
* Kevin Harms: Added experimental utilities for loading darshan results into SQL
* Use exclusive flag when opening output file (to protect against file name collision)
* Rob Ross: updates to allow command line tools to build on Darwin
* Bug fix for pnetcdf configure problem reported by Rob Latham; darshan now always pulls in MPI_Wtime() symbol at link time
darshan-1.1.8
=============
* Added tracking of file system type and mount point for each file
* Added tracking of file size at open time (CP_SIZE_AT_OPEN)
* Moved sync cost to be counted in cumulative write time rather than cumulative metadata time
* Added sync as a separate category in darshan-job-summar.pl counters
* Bug fix to most frequent access size table in darshan-job-summary.pl
* Converted all utilities to use darshan-logutils api for reading output files
* Added backwards compatibility to darshan-logutils routines
* Kevin Harms: Added darshan-analyzer utility to summarize usage of MPI-IO, pNetCDF, HDF5, and shared files across a set of output files
* Fixed bug field listing for darshan-diff utility
darshan-1.1.7
=============
* Minor fix for a compile warning
darshan-1.1.6
=============
* Kevin Harms: bug fix for segfault in apps that use MPI_Init_thread()
darshan-1.1.5
=============
* Limit PMPI usage in library to fewer functions
* Update PMPI detection in compiler scripts to ignore functions unused by Darshan
darshan-1.1.4
=============
* Track files opened via Parallel NetCDF
* Track files opened via HDF5
* Record slowest individual POSIX read and write times along with access size for those operations
* Inspect symbols at compile time to determine whether to enable Darshan or not based on the presence of MPI and PMPI symbols
* Use GNU and IBM compilers from path rather than hard coded location
* Simplify warning message if unable to open log file
* Remove unused internal benchmark routines
darshan-1.1.3
=============
* Updated compiler scripts for V1R4 driver on BlueGene/P
darshan-1.1.2
=============
* Added *_r versions of each IBM compiler script on BlueGene/P
darshan-1.1.1
=============
* Set default permissions to 0400 (user read only) for output files
* Automatically disable Darshan at link time if common PMPI libraries are detected in the command line
* Experimental tool (darshan-gen-cc.pl) to automatically generate Darshan-enabled mpicc scripts
darshan-1.1.0
=============
* Initial public release