Changelog for
gfio-2.21-2.1.x86_64.rpm :
Thu Jun 29 14:00:00 2017 ktsamisAATTsuse.com
- Update to version 2.21:
* See http://brick.kernel.dk/snaps/fio-2.21.shortlog.txt
- Enabled tests
Sat May 27 14:00:00 2017 david.bahiAATTdell.com
- Update to version 2.20:
* stat: Re-add output of basic bw information if bw_log is not written
* Fix wrap issue with 64-bit pwritev2/preadv2
* options: force refill_buffers with pattern and any reads
* gettime: make utime_since_now and mtime_since_now consistent in how they record the caller and put this all behind FIO_DEBUG_TIME
* client/server: make sure that all elements in io_u_lat_m[] are transferred and received
* stat: reset_io_stats: fix a problem, rearrange some code
* remove redundant _fio_rbd_disconnect, which is already called in fio_rbd_cleaup
* os/os-android.h: fix shared memory support
* Drop triple X for cpu affinity for OpenBSD
* Implement shm_attach_to_open_removed() for OpenBSD
* Turn off lex by default on OpenBSD
* os/os-linux.h: fix broken byte swap intrinsics
* os/os-android.h: use byte swap intrinsics if available
* os/os-android.h: fix broken shmget() due to ndk-r15
* fio.h: zero old flag bits when setting new ioengine flags
* configure: fix broken test for cuda
* thread_options: kill two unused pads
* server: bump protocol version
* gpu: kill a lot of useless ifdefs
* gpu: ensure that we convert gpu_dev_id options
* GPUDirect RDMA support
* seqlock: add simple user space code for sequence locks
* Drop the only local variable declaration within a for-loop (C99)
* Fix num2str() output when modulo != -1U
* stat: cleanup iops/bw logging functions
* stat: make next log time decision cleaner
* Return non-negtive error in order to print right error msg
* doc: add ftruncate engine documentation and example jobfile
* Fix num2str() output when maxlen <= strlen(tmp)
* dump_td_info() doesn\'t really need to be a function
* Don\'t silently terminate td when no I/O performed due to error
* backend: include data direction in debug poisson rate print
* Split poisson rate control into read/write/trim
* engine: add ftruncate ioengine
* engine: e4defrag fix error reporting
Mon Apr 10 14:00:00 2017 mpluskalAATTsuse.com
- Update to version 2.19:
* See http://brick.kernel.dk/snaps/fio-2.19.shortlog.txt
Fri Mar 24 13:00:00 2017 mpluskalAATTsuse.com
- Update to version 2.18:
* See http://brick.kernel.dk/snaps/fio-2.16.shortlog.txt
* See http://brick.kernel.dk/snaps/fio-2.17.shortlog.txt
Fri Jan 6 13:00:00 2017 mpluskalAATTsuse.com
- Update to version 2.16:
* See http://brick.kernel.dk/snaps/fio-2.16.shortlog.txt
- Add gpg signature
Thu Nov 3 13:00:00 2016 mpluskalAATTsuse.com
- Update to version 2.15:
* See http://brick.kernel.dk/snaps/fio-2.15.shortlog.txt
Sun Aug 28 14:00:00 2016 mpluskalAATTsuse.com
- Update to version 2.13:
* workqueue: rename private to priv for compiling as c++
* verify: Reset verify_state before verification phase.
* HOWTO: remove old use cases for the net IO engine
* gfio: call g_thread_init() for <= 2.31.0
* iolog: flush_log() can be bool
* Merge branch \'fix_verify\' of
https://github.com/charles-jacobsen/fio
* pthread: bump min stack size
* log: fix missing entries
* log: fix averaged latency logging
* samples being added to the pending log were silently dropped
because we failed to set nr_samples in the new log they get
copied into
* drifting in output of interval-averaged values was eventually
causing IOP samples to be dropped.
* plot: add gnuplot 5 support
* plot: indicate that the pattern is a glob
* Add os_trim() support for DragonFlyBSD
* Add os_trim() support for FreeBSD
* Fix wrong cpuio option name in documentation
* Add CPU affinity support for DragonFlyBSD
* Make I/O priority option generic for non-Linux environment
* Add ioprio_set() support for DragonFlyBSD
* Change ARCH_X86_64_h to ARCH_X86_64_H
* Add os/os-linux-syscall.h to separate syscall NR from arch
headers
* Add missing header inclusion for Android from 1c764dbe
* Remove hard-coded precision for printing JSON float values
* helper_thread: remove impossible branch
* Fix \"exitall_on_error\" option \"exitall_on_error\" option
should be enabled without any argument.
Sat Jul 16 14:00:00 2016 mpluskalAATTsuse.com
- Enable support for RBD
Thu Jun 16 14:00:00 2016 mpluskalAATTsuse.com
- Update to version 2.12:
* No changelog available
Fri May 27 14:00:00 2016 mpluskalAATTsuse.com
- Update to version 2.11:
* No changelog available
Sun May 22 14:00:00 2016 mpluskalAATTsuse.com
- Update to version 2.10:
* No changelog available
Mon Apr 4 14:00:00 2016 mpluskalAATTsuse.com
- Update dependencies (enable rdmca)
Wed Mar 16 13:00:00 2016 lmbAATTsuse.com
- Update to 2.8
* verify: clear IO verify state all_io_list before writing
* Revert \"options: move pattern_fmt_desc where we need it\"
* lib/rand: make __init_randX() static
* Use 32-bit rand for parts that use rand_between()
* t/gen-rand: use 32-bit random generator
* Fix compile of test programs on archs that use arch_flags at runtime
* t/gen-rand: remove compile warning on 32-bit
* .gitignore: ignore vim undo files
* Add t/gen-rand to test random generator
* Update documentation on log file formats
* rand: use bools
* Add the sample JESD219 job file
* io_u: speed up __get_next_buflen()
* options: improvements to parse dry run
* options: finish merge of bssplit/rand zone code
* options: unify the bssplit/zone split code
* Fix double free of td zone state index
* Log to parent instead of twice to child td
* options: clean number of zones if we fail parsing
* Add support for zones of random IO, with varying frequency of access
* Update documentation for random_distribution=gauss
* Add support for preadv2/pwritev2
* ioengines: account any queued IO on the engine side
* backend: ensure that fio_io_sync() commits IN on queued status
* Allow IO engine driven allocations of IO structures
* Allow for the include file specification to be relative.
* fio.1: man page fixes
* Update documentation
* server: don\'t make SO_REUSEPORT errors fatal
* diskutil: don\'t print terse disk util twice for json,terse output format
* diskutil: fix segfault for both json and terse output
* io_ddir: io_ddir_name array should be static
* fio: fix copy/paste error in io_ddir.h
* log: add support for logging max instead of averages
* stat: cleanup average logging
* fio: fix ignore_error regression
* libmtd: compile fix for linux header changes.
* Makefile: false positive on _FORTIFY_SOURCE
* Add a note of what is the default value for random_generator
* server: send trigger state inline
* client: fix double removal of client on job file open failure
Tue Feb 9 13:00:00 2016 mpluskalAATTsuse.com
- Update to 2.6
* Fix Windows build: define rand_r()
* powerpc: fix cpus_online() to get correct max CPU number for
powerpc64
* ioengine: remove duplicate queue handler check
* server: use the wrapped MSG_DONTWAIT
* Makefile: fix duplicate strcasestr for windows
* client: remove debug printf() for compressed iolog
* server: disable pdu length check for now
* server: split vectored commands into fragments
- Changes for 2.5/2.4
* use option smaalloc_pool_size
* client: rename send_file() to fio_send_file()
* lib/pattern: include oslib/strcasestr.h
* Fix included getopt locations
* t/genzipf: fix for strcasestr missing
* t/dedupe: silence bogus warning on \'bytes\' being used
uninitialized
* configure: check if lex requires/fails with -o for output
* configure: add --enable-lex
* options: make the groups/categories constant
* options: split out option grouping code
* options: we don\'t need the crc32c verify include
* options: move pattern_fmt_desc where we need it
* server: include name of verify file requested for error message
* configure: lex force off
* server: sk_out exit error handling
* verify: split out state header code
* verify-state: forward declare io_u
* Merge branch \'master\' of git://github.com/ezrapedersen/fio
* t/verify-state: add helper to inspect verify dump state files
* options: fix bs_cmp()
* Always create server sk_out key
* Destroy sk_out server key when we are done
* t/verify-state: fix type for printf
* t/verify-state: one more printf type fix
* io_u: ensure that we align new start offset properly for
time_based
* Fix issue with termination before io_size has been reached
* client: drain text output when exiting
* client: fix non-text command leak
* init: increment stat count if parent == default thread
* init: fix compile for FIO_INC_DEBUG not being set
* mutex: fix double unlock in fio_mutex_down_timeout()
* Disable libaio for ESXi build - bug#80
- Enable parallel building
Sun Jan 3 13:00:00 2016 mpluskalAATTsuse.com
- Update to 2.2.13
* no changelog available
Sun Dec 6 13:00:00 2015 mpluskalAATTsuse.com
- Update to 2.2.12
* no changelog available
Sun Sep 20 14:00:00 2015 mpluskalAATTsuse.com
- Update to 2.2.10
* no changelog available
Mon Jul 6 14:00:00 2015 mpluskalAATTsuse.com
- Update to 2.2.9
* see http://brick.kernel.dk/snaps/fio-2.2.9.shortlog.txt
Sat Apr 4 14:00:00 2015 mpluskalAATTsuse.com
- Update to 2.2.6
* make per-thread IOPS more accurate
* man page updates
* misc bugfixes
- Cleanup spec file with spec-cleaner
- Update dependencies
* enable numa
* enable glusterfs
Wed Jan 28 13:00:00 2015 dsterbaAATTsuse.cz
- 2.2.5
- rate limiting improvements
- verify, checksumming fixes
- shm support can be made optional
- lots of statbility fixes
- 2.2.4
- bugfixes in axmap, cleanups
- 2.2.3
- cpumask handling fixed
- 2.2.2
- fixed buffer scrambling
- 2.2.1
- btrace2fio: -a passes specific options
- minor bugfixes
- 2.2.0
- verify_state triggers
- buffer_pattern to specify the pattern manually
- make suggestions for unknown options
- time accounting improvements
- documentation updates
Thu Nov 20 13:00:00 2014 dsterbaAATTsuse.cz
- 2.1.14
- arithmetic expressions in jobfiles
- support for --remote-config
- net engine updates
- rbd engine updates
Fri Oct 3 14:00:00 2014 dsterbaAATTsuse.cz
- 2.1.13
- btrace2fio: new tool to turn trace into fio job file
- basic dedupe support, job config, standalone tool
Fri Sep 12 14:00:00 2014 dsterbaAATTsuse.cz
- 2.1.12
- Add support for HDFS IO engine
- misc ioengine fixes
Wed Jul 30 14:00:00 2014 dsterbaAATTsuse.cz
- 2.1.7
* file append support
- 2.1.8
* mixed commandline and global options support
* fixes for s390
- 2.1.9
* assorted bugfixes
- 2.1.10
* glusterfs support
* aarch64 support
- 2.1.11
* configurable support for gfapi/rbd
* server protocol update
* add log_offset
* removed --latency-log
* direct io fo glusterfs
* iolog: logging improvements (compression) and fixes
Mon Mar 10 13:00:00 2014 bahidAATTemc.com
- update to 2.1.6.1:
* Full support for IPv6 for the net I/O engine.
* Windows threading fixes.
* Support for ranges in the \'startdelay\' option.
* Adds rbd (rados block device) engine support.
* Various fixes for blktrace playback on Linux.
* Adds OpenBSD support.
* An option added to manually seed random generators.
* Various client/server fixes.
* A fix for a crash on some file locking options.
* Adds support for fast software-based xxhash, and support for testing checksum speeds.
* A fix for not always honoring the \'openfiles\' setting.
* Support for setting policy on how to distribute CPUs set for jobs.
* Support for data retention testing with verify
Tue Dec 17 13:00:00 2013 termimAATTgmail.com
- Add HOWTO (job file description) to the docs
Thu Dec 5 13:00:00 2013 termimAATTgmail.com
- update to 2.1.4:
* adds JSON output for client/server
* fix for Windows build breakage in the previous release
* fix for ioengine specific options
* adds support for O_ATOMIC atomic writes on Linux
* failsafe check for realloc() failure during logging
* fix for a Solaris thread problem
* various man page and documentation updates
* fix for a crash on PowerPC without the mfspr instruction
Wed Aug 7 14:00:00 2013 pascal.bleserAATTopensuse.org
- update to 2.1.2:
* adds basic support for an ACT (aerospike) profile
* fixes atomic write support
* adds various parsing fixes
* makes a big revamp of the included graphing support
* makes the IOPS log includes bandwidth
* adds support for specific randomness settings for both reads and writes
* adds support for random vs. sequential blocksize settings
* fixes a potential verification failure
* adds multicast support for the net ioengine
Fri May 24 14:00:00 2013 Greg.FreemyerAATTgmail.com
- update to 2.1
* Integration of the gfio GTK frontend.
* Better Android integration.
* The initial file fill honors data compression settings.
* An option for controlling randomness of I/O (percentage_random) instead of being fully random or fully sequential.
* Fixes a bug in uneven pattern sizes for verification.
* Various other important fixes
- start calling ./configure as now described in README
- put the newly included gfio into it\'s own dedicated sub-package
- add BuildRequres zlib-devel to support compression capability for gfio support
- add BuildRequres gtk2-devel for gfio support
Mon Apr 8 14:00:00 2013 Greg.FreemyerAATTgmail.com
- update to 2.0.15
* adds an --eta-newline option to insert ETA new line at fixed intervals.
* Fixes for the CPU clock, notably PPC.
* Fixes for workloads with read/write file locking.
* A fix for skewed latencies with rated IOPS.
* Fixes a bug with block sizes varying more than 64x.
* Improves the LFSR implementation.
* Fixes missing JSON bandwidth output.
* Various fixes for the Android platform.
* Fixes man page installation.
* Other bugfixes and improvements.
- remove fio-2.0.14-fix-configure.patch, now in upstream
- change Recommends gnuplot to Suggests gnuplot, based on end user feedback
Mon Mar 25 13:00:00 2013 Greg.FreemyerAATTgmail.com
- update to 2.0.14
* Autodetect features with a new configure script when building.
* Improve ETA for some workloads.
* Support for verify mode that requires no metadata, greatly reducing memory requirements for big device verify workloads.
* Ensures fsync is always done, even for direct/raw I/O (if asked to).
* Support for unified read/write/trim reporting.
* A few axmap and cpuclock bugfixes.
* Fixes zones and numjobs used together.
* Adds support for idle profiling, showing idle system CPU when running a workload.
* Many other fixes; all users should upgrade
- added patch to fix new configure logic introduced by upstream in 2.0.14
* fio-2.0.14-fix-configure.patch
Mon Feb 4 13:00:00 2013 bahidAATTemc.com
- update to 2.0.13:
* Fix potential null pointer dereference on verify and requeue events
* server: move fio_net_cmd to the end of fio_net_int_cmd
* Use variable[] instead of GCC variable[0] syntax
* Enable -ffast-math
* time: convert to uint64_t
* Use uintptr_t for engine data
* Start of transition from unsigned long long to uint64_t
* zipf/pareto: use min of file size and io size for init
* zipf/pareto: use size= if given, not always device size
* libaio: use container_of() instead of silly casting
* gettime: even rounding, don\'t always round up
* Move \'tsc_reliable\' outside of ARCH_HAVE_CPU_CLOCK
* implement get_cpu_clock() for s390/s390x
* clock: turn expensive division into multiply + cheap division
* t/axmap: update tester for lfsr_init() taking a seed argument
* Include
before checking __WORDSIZE
- Fio 2.0.12.2
* Add return statements to arch_init functions for ia64 and ppc.
* Use clock_gettime() for CPU clock calibration
* Consistently indent using tabs in Windows files and remove trailing spaces.
* Create a new Windows installer product GUID for 2.0.12.
* Fix $(CC) override: use system compiler except on HP-UX and Solaris.
- Fio 2.0.12.1
* Fixup CPUID for 32-bit x86
- Fio 2.0.12
* windowsaio: fix typo
* gettime: make last_cycles thread local too
* gettime: use pthread_{set,get}specific() for TLS
* gettime: fix race/bug with threads and time keeping
* Add --cpuclock-test and clocksource= option help
* gettime: include per-cpu clock calibration in cpu clock test
* gettime: locking fix and debug check for identical sequence
* cpu clock: add independent test for monotonic/sane TSC
* clock: ensure that we re-init if the clocksource changes from the default
* obinject: fixup ->file_data breakage
* file: unify ->file_data and ->file_pos
* file hash: don\'t close file fd on lookup/add race
* cpu clock: round up when dividing by samples
* Define TSC arch_init() for PPC and IA64
* gettime: fix CPU calibration reported mean
* gettime: calibration rounding error
* Increase CPU clock calibration accuracy
* Add check for invariant TSC on x86 and use TSC is default clock if reliable
* windowsaio: create a single completion port during init, associate files during Document the ioengine=net pingpong= option
* Fix windows out-of-memory handling
* Free io_u related structures before killing IO engine
* windowsaio: initialize and map windowsaio IO structure to io_u
* parser: always match the correct option length for posval options
* Fix man page indentation
* lfsr: add HOWTO and man page documentation
* lfsr: ensure that the cycle follows the randrepeat= setting
* Name the various random offsets we use
* Revert \"Change preferred default clocksource to gettimeofday()\"
* engines/net.c: fix compilation error due to missing signal.h include
* Wire up SIGUSR2 to kill blocking threads
* Change preferred default clocksource to gettimeofday()
* Cache layout improvements
* net: fix receiver start time
* net: add basic ping/pong type workload support
* verify: treat as failure if given verify type is different from media
* net: sent udp open messages
* Merge branch \'rand-map\'
* solaris: enable fadvise()
* Fix broken \'norandommap\' rand-map
* parse: fix wrong \"might be used unitialized\" warning on some compilers
* Wire up _weak posix_fadvise()
* net: set runstate to SETTING_UP while waiting for a connection
* net: exit on WAITALL and !ret
* net: fix segfault with receiver, tcp, and no hostname set
* libfio: don\'t sent KILL signal to own process from signal handler
* Rename the bitmap to axmap
* bitmap: fix off-by-8 allocation error
* Fix windowsaio IO error handling and document device write access issue on WindoFix compile for older Linux without mmap MAP_HUGETLB
* lfsr: crank it 128 times before using the sequence
* Ensure that lfsr_init() is always called
* Add LFSR generator
* Restore BUSY_OK bypassing of bitmap
* Kill now unused bitmap defines from legacy code
* bitmap: kill debug code
* bitmap: fix bit_masks[] for 32-bit compiles
* Rework file random map
* Add support for Android
* memory: wire up mmap huge page support
* Use unique seed for zipf/pareto init if rand_repeat is not set
* Generate a new Windows installer product code for 2.0.11.
* Add more examples to the Windows installer.
- Fio 2.0.11
* Document random distribution
* Disable random map automatically if a non-uniform random distribution is given
* genzipf: add help/usage screen (-h)
* Get rid of -fno-omit-frame-pointer
* t/genzfip: add -c option for csv output
* server: properly reset stat_number in fio_reset_state()
* client/server: fixup \"All clients\" reporting
Thu Oct 18 14:00:00 2012 pascal.bleserAATTopensuse.org
- update to 2.0.10:
* support for trim as a primary IO directive
* addition of the JSON output format for easy parsing by both humans and
computers
* fixes for engine starvation for extraordinarily long latencies
* fallocate() fixes
* fixes for netio compatibility
* add support for SIGBREAK
* the critical error list is now configurable
* three new IO engines were provided: falloc, e4defrag, and Fusion-io atomic
write engine
* a new sample job file was provided for SSD steady state performance
Tue Oct 16 14:00:00 2012 Greg.FreemyerAATTgmail.com
- correct License to GPLv2, not GPLv2+ (bnc#783876)
Fri Oct 5 14:00:00 2012 Greg.FreemyerAATTgmail.com
- update to 2.0.9
* A fix for premature freeing of disk utilization structures, and missing
locking of the same.
* fio dumps current progress if sent a USR1 signal.
* Includes git version in version output.
* A fix for a crash with more than ~500 jobs.
* Man page updates.
* A fix for wrap-around behavior for time-based sequential read jobs.
* A fix for a \"file hash not empty\" warning on blktrace replay.
* A fix for compilation on the MIPS architecture.
* Improved display of commit-rate while running.
* A fix for fallocate failing on some Linux systems.
* Includes an end-of-run timestamp in output
Sat Apr 14 14:00:00 2012 pascal.bleserAATTopensuse.org
- update to 2.0.7:
* fixes the terse output when used with a description field
* enhances display of reaped threads (indicating whether they exited
normally, were killed, or died due to a signal)
* removes the assert which could occasionally trigger on verify overwrites
* fixes the \"Group run status\" aggrb field
Fri Mar 30 14:00:00 2012 pascal.bleserAATTopensuse.org
- update to 2.0.6:
* a new offset_increment option to more easily be able to have different
threads use different regions of a file/device
* a new buffer_compression_percentage option enabling the user to specify the
compression ratio to expect from the data written by fio
* a fix for a segfault caused by invalid verification job files
* a few client/server bugfixes
* a fix for a bug that caused early termination of fio with verify_backlog
* a fix for a problem with iodepth_batch_complete=0 and accounting
* enables read/write sequential offset to be a non-multiple of the block size
- changes from 2.0.5:
* a bug affecting IOPS logging was fixed
Sun Feb 26 13:00:00 2012 pascal.bleserAATTopensuse.org
- update to 2.0.4: too many changes to list, consult documentation
Fri Sep 30 14:00:00 2011 pascal.bleserAATTopensuse.org
- update to 1.58:
* adds an RDMA IO engine
* adds support for direct userspace reaping of completed events with libaio
IO engine
* ensures that bool options also work if given a parameter
* adds support for \"holed\" IO patterns
* fixes a latency reporting bug
* fixes a disk utilization bug for delayed jobs
* adds resolution support to fio_generate_plots
* fixes summed computation of standard deviation
* adds completion latency percentile support
* fixes lots of other minor bugs
Fri Jul 22 14:00:00 2011 pascal.bleserAATTopensuse.org
- update to 1.57:
* fixes an infinite loop on platforms with very limited number of AIO request
depth support
* fixes an issue with killing of threads that haven\'t started yet
Mon May 30 14:00:00 2011 pascal.bleserAATTopensuse.org
- update to 1.55:
* an issue was fixed with the splice IO engine, in which it didn\'t work
correctly on file systems that did not support O_DIRECT
* an issue was fixed where IO could span two separate regions, causing write
verify failures
Fri May 13 14:00:00 2011 pascal.bleserAATTopensuse.org
- update to 1.54:
* a 32-bit randomness problem has been fixed
* a crash with giving a bad --ioengine command line option has been fixed
Fri May 6 14:00:00 2011 pascal.bleserAATTopensuse.org
- update to 1.53:
* this is a maintenance release, fixing a few important bugs
* the runtime option was broken due to a bad change
- fix compiling with optflags
Fri Apr 29 14:00:00 2011 pascal.bleserAATTopensuse.org
- update to 1.52:
* an end-of-job slowdown for big random workloads was fixed
* support was added for running multiple sections at the time with the
- -section option
* a parser bug with range options and size postfix was fixed
* an internal random generator is now used for IO offsets
- changes from 1.50:
* new features for blktrace replay
* fixes a slowdown bug in the posixaio engine
* adds a binject I/O engine
* adds support for TRIM/DISCARD operations
* extends I/O verification to dump good and bad data on failure for better
post mortem analysis
* various speedups and optimizations
* a lot of obscure bugs have been found and fixed
Fri Jun 18 14:00:00 2010 pascal.bleserAATTopensuse.org
- update to 1.41:
* adds support for explicit clocksources
* speeds random buffer fills
* improves ETA calculation for workloads using \"fill_device\"
* fixes a bug with --section matching
* adds improvements for verifying workloads
* add support for offloading \"verify\" to dedicated threads
* fixes strict non-overlap random workloads
* fixes a bug in matching strings in the parser
- update to 1.38:
* fixes a bug in sha256 that prevented it from doing correct verifies
* fixes a bug with mixed block size ranges and verify
* now uses the minimum block size as the verify interval
* command size length was increased
* initial support for benchmark profiles was added (\'tiobench\' is now such a
profile)
* a few parser bugs were fixed
* support for \'sync_file_range()\' on Linux
* a few places where errors would not get reported were fixed
* overhead of the IO path was reduced in a few places
Tue Feb 23 13:00:00 2010 pascal.bleserAATTopensuse.org
- update to 1.37:
* a bug in sha256 calculation was fixed
* bad use of lstat(), causing problems with links to devices, was fixed
* a specific fallocate() control switch was added
* CPU utilization of sequential sync IO was reduced
- changes from 1.36:
* simple math support and reserved keywords were added to job
file
* initial support for benchmark profiles was added
* support for cgroup based IO controllers on Linux was added
* verify failure logging was improved
* a bug with rate IO and queuing IO engines was fixed
Thu Nov 5 13:00:00 2009 pascal.bleserAATTopensuse.org
- update to 1.35:
* fixes some parser bugs
* fixes the shared memory allocator
* reserved keywords are now supported in job files
* a few fixes were made for verifying workloads
* fixed a hang when resolving component devices of md/dm arrays
* IO log replay fixes and additions were made
* fixed a hang due to an incorrect mutex clock source
* added support for wider than 32-bit pattern verifies
* fixed a few races when using async verifier threads
* fixed a hang with the async engine and IO depth batch settings
Thu Jul 9 14:00:00 2009 pascal.bleserAATTopensuse.org
- update to 1.31:
* fixes completion latency numbers (they would read 0)
* fixes an alignment bug
* improves the speed of the mmap engine on 64-bit architectures
* adds support for asynchronous \"verify\" offload threads to speed \"verify\"
workloads
- changes from 1.29:
* fixes and improvements for the IO limiting rate option
* fix for submission latency numbers in some cases
* fix for a bug in the job parser that would sometimes truncate the postfix
value when multiple values were given
* a continue_on_error option was added to allow a job to keep running if it
sees non-fatal errors
* a bug in the mmap engine with large files and 32-bit machines was fixed
* support for fdatasync() was added with the fdatasync option
* an iomem_align option was added for fine-grained control over IO buffer
memory alignment
* the pre_read option was fixed and no longer attempts to pre-read on IO
engines that cannot seek
Tue Jun 2 14:00:00 2009 pascal.bleserAATTopensuse.org
- update to 1.27:
* corruption of submission/completion latency logs for shared files was fixed
* blocksizes are included in latency logs
* a \"pre_read\" option was added to pre-read files into memory before starting
the job
* exec_prerun/postrun now invokes a shell
* accuracy of disk util stats was improved
* large files are supported with an mmap engine on 32-bit architectures
* ETA stats were improved with better throughput reporting, and IOPS
reporting was added there too
* the man page was fixed based on lintian
Tue Apr 21 14:00:00 2009 pascal.bleserAATTopensuse.org
- update to 1.26:
* a crash with multiple files for a job was fixed
* reporting of throughput was corrected, as it was sometimes off by about
2.4%
* IO logging is allowed for diskless engines
* support was added for different weights for reads and writes with the
bssplit= option
* a crash with engines not supporting get_file_size() was fixed
* a sample SSD performance analysis job was added
* a problem was fixed with IO offsets sometimes being too large, causing
early exit of failed IO verification