Jens Axboe [Fri, 3 Jul 2009 06:30:15 +0000 (08:30 +0200)]
Fix O_DIRECT memory alignment
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 2 Jul 2009 08:20:26 +0000 (10:20 +0200)]
Fio 1.29
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 2 Jul 2009 06:31:30 +0000 (08:31 +0200)]
Merge branch 'master' of ssh://router/data/git/fio
Radha Ramachandran [Thu, 2 Jul 2009 06:30:49 +0000 (08:30 +0200)]
Fix issue with fill_device=1 and continue_on_error
I added a change so when both the continue_on_error and fill_device
option is on, when it hits ENOSPC issue the code ignores the error.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 1 Jul 2009 21:02:10 +0000 (23:02 +0200)]
Add code to detect a task that exited prior to up'ing the startup mutex
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 1 Jul 2009 10:26:28 +0000 (12:26 +0200)]
Don't allow pre_read on IO engines that cannot seek
We cannot pre-read files, if we cannot seek back and read the same
data again. So just disable it with a warning to the user.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 1 Jul 2009 08:45:31 +0000 (10:45 +0200)]
Fio 1.29-rc1
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 1 Jul 2009 08:43:05 +0000 (10:43 +0200)]
Fix compile error
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 1 Jul 2009 08:42:18 +0000 (10:42 +0200)]
Move gtod_cpu affinity check to verify callback
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 1 Jul 2009 08:38:35 +0000 (10:38 +0200)]
Move rw= write && read_only check to ->verify callback
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 1 Jul 2009 08:33:03 +0000 (10:33 +0200)]
Add support for the iomem_align option
This allows detailed control of the alignment of the IO buffers
that fio uses.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 1 Jul 2009 07:24:05 +0000 (09:24 +0200)]
Add verify hook to options
This allows us to individually verify options, instead of having
a big function that attempts to fix things up.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 1 Jul 2009 07:06:12 +0000 (09:06 +0200)]
Only show maxval in --cmdhelp if it is given
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 1 Jul 2009 07:00:52 +0000 (09:00 +0200)]
Cleanup IO buffer memory allocation
This is in preparation for adding an option that controls the
alignment of the buffers.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 29 Jun 2009 09:54:56 +0000 (11:54 +0200)]
mmap engine: fix for large files on 32-bit archs
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 16 Jun 2009 21:23:01 +0000 (23:23 +0200)]
sg engine: IO should be done sync of O_SYNC and O_DIRECT
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 16 Jun 2009 20:40:26 +0000 (22:40 +0200)]
Add support for fdatasync()
Adds a new option, fdatasync=. It's identical to the fsync= option,
but uses fdatasync() instead.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 16 Jun 2009 06:31:58 +0000 (08:31 +0200)]
Merge branch 'master' of ssh://router/data/git/fio
Radha Ramachandran [Tue, 16 Jun 2009 06:31:29 +0000 (08:31 +0200)]
Alter the order of checking time exceeded vs getting an io_u
I was hitting a case where fio would loop for ever even though the
runtime was exceeded if it had difficulty getting the io_u struct, so
I suggest this change to the code so it wld not loop forever.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 15 Jun 2009 21:29:54 +0000 (23:29 +0200)]
Default print pretty up
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 15 Jun 2009 12:45:43 +0000 (14:45 +0200)]
Cmd help type fix
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 15 Jun 2009 10:37:30 +0000 (12:37 +0200)]
Fix typo on continue_on_error patch
Also expand the error print a bit.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Radha Ramachandran [Mon, 15 Jun 2009 06:40:16 +0000 (08:40 +0200)]
Add a 'continue_on_error' option to fio
Add option to make fio continue on non-fatal errors.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jun 2009 20:06:40 +0000 (22:06 +0200)]
Fio 1.28
Should be good enough for another major release.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jun 2009 20:06:23 +0000 (22:06 +0200)]
Yet another rate fixup
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jun 2009 08:55:39 +0000 (10:55 +0200)]
Improve rate limiting
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jun 2009 08:41:28 +0000 (10:41 +0200)]
Get rid of (now) unused rate_throttle()
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jun 2009 07:56:36 +0000 (09:56 +0200)]
Fio 1.27.4
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jun 2009 07:56:21 +0000 (09:56 +0200)]
Remove a reference check
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jun 2009 07:05:13 +0000 (09:05 +0200)]
File close fixes
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jun 2009 06:55:35 +0000 (08:55 +0200)]
Merge branch 'master' of ssh://router/data/git/fio
Jens Axboe [Wed, 10 Jun 2009 06:55:21 +0000 (08:55 +0200)]
More documentation updates
fio.1 should now be in sync with HOWTO.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jun 2009 06:39:16 +0000 (08:39 +0200)]
Documentation update
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jun 2009 04:58:47 +0000 (06:58 +0200)]
Fio 1.27.3
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jun 2009 04:57:47 +0000 (06:57 +0200)]
Fix slat numbers
After commit
306ddc9752eef70c3fbb111af63d197a3a6d447f, it was logging
the blocksize since arguments 3 and 4 were switched.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jun 2009 04:35:48 +0000 (06:35 +0200)]
Fix bad min rate check
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jun 2009 13:43:25 +0000 (15:43 +0200)]
parser: get rid of FIO_OPT_STR_VAL_INT
It's treated the same as FIO_OPT_INT since we got rid of the
int vs siint distinction.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jun 2009 13:23:38 +0000 (15:23 +0200)]
Use __e for errno
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jun 2009 12:38:06 +0000 (14:38 +0200)]
Fio 1.27.2
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jun 2009 12:34:35 +0000 (14:34 +0200)]
Fix problem with too many opened files
If using raw block devices, we would leak the fd from the bdev_size()
function. Also add some debug triggers to catch this in the future.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jun 2009 12:14:20 +0000 (14:14 +0200)]
Bad indent
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jun 2009 11:05:32 +0000 (13:05 +0200)]
Fix fd leak in cache line checking
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jun 2009 10:58:37 +0000 (12:58 +0200)]
Missed int to pointer conversion check for ->ratemin
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jun 2009 10:48:56 +0000 (12:48 +0200)]
Merge branch 'master' of ssh://router/data/git/fio
Jens Axboe [Tue, 9 Jun 2009 10:47:16 +0000 (12:47 +0200)]
Add support for limiting only rate in only one direction
So now you can say 'limit writes to 10MB/sec' and have reads go
full throttle, for instance.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jun 2009 10:28:57 +0000 (12:28 +0200)]
Fix bug in parser with postfix and truncated direction values
If you pass in foo=500k, then the 'k' multiplier will go unnoticed.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 5 Jun 2009 20:12:19 +0000 (22:12 +0200)]
Fio 1.27.1
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 5 Jun 2009 20:11:52 +0000 (22:11 +0200)]
Fix fill_device with read/randread
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 5 Jun 2009 20:10:51 +0000 (22:10 +0200)]
Move td_* ddir defines
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 10:14:08 +0000 (12:14 +0200)]
Add pre_read to man page
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 09:36:29 +0000 (11:36 +0200)]
Fio 1.27
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 08:53:45 +0000 (10:53 +0200)]
Improve throughput ETA display
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 08:47:44 +0000 (10:47 +0200)]
Add iops rate to ETA display
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 08:30:33 +0000 (10:30 +0200)]
Fix bug with rwmix and multiple loops over the same set
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 07:29:28 +0000 (09:29 +0200)]
More fio.h cleanups
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 07:25:57 +0000 (09:25 +0200)]
Split off diskutil include
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 06:54:00 +0000 (08:54 +0200)]
Move file service type defines to file.h
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 06:53:28 +0000 (08:53 +0200)]
Add iolog defines
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 06:50:36 +0000 (08:50 +0200)]
Move IO engine flags
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 06:49:39 +0000 (08:49 +0200)]
Add ioengine.h header
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 06:46:38 +0000 (08:46 +0200)]
Move more file stuff to file.h
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 06:45:40 +0000 (08:45 +0200)]
Cleanup verify headers
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jun 2009 06:41:15 +0000 (08:41 +0200)]
Clean up file flags
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 2 Jun 2009 19:40:09 +0000 (21:40 +0200)]
Fix problem with fill_device and read workloads
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 2 Jun 2009 12:19:25 +0000 (14:19 +0200)]
Make pre_read imply invalidate=0
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 2 Jun 2009 12:18:41 +0000 (14:18 +0200)]
mmap engine: don't invalidate cache on segment remap
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 2 Jun 2009 11:41:27 +0000 (13:41 +0200)]
mmap engine: remove 'mmap too large' check for 32-bit
We now support large files regardless of architecture, so just kill
the check again.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 2 Jun 2009 11:06:01 +0000 (13:06 +0200)]
mmap engine: allow large files on 32-bit archs
Map chunks of 2GB at the time in total, remapping as we go
along.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 2 Jun 2009 10:15:51 +0000 (12:15 +0200)]
Fio man page still refers to old mailing list
Point it at the vger list instead.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 2 Jun 2009 09:27:26 +0000 (11:27 +0200)]
Fix problem with fadvise() and fill_device option
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 27 May 2009 11:12:14 +0000 (13:12 +0200)]
Merge branch 'master' of ssh://router/data/git/fio
Jens Axboe [Wed, 27 May 2009 11:11:20 +0000 (13:11 +0200)]
Increase accuracy of disk utility percentage
If a job file contains more than one job that works on separate
disks, then the job that finishes first will get more idle time
accounted to it.
Fix this by only doing disk util stats when we have files open
on that device.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 22 May 2009 07:51:20 +0000 (09:51 +0200)]
Remove io_u->endpos
It's only set in one location and never read.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 May 2009 10:10:02 +0000 (12:10 +0200)]
Make sure that exec_prerun/postrun work for shell commands
Pass the arguments to sh for execution.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 May 2009 09:56:54 +0000 (11:56 +0200)]
Fix old printf() in pre_read()
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 May 2009 09:53:04 +0000 (11:53 +0200)]
Fio 1.26.4
pre_read option was buggy, so tag a new release now.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 May 2009 09:52:15 +0000 (11:52 +0200)]
pre_read fixes
1) Add a specific runstate for pre-read, so we can see that this
is what the job is currently doing.
2) open/close files for pre-read
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 May 2009 09:37:15 +0000 (11:37 +0200)]
Fio 1.26.3
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 May 2009 09:36:39 +0000 (11:36 +0200)]
Add fio_generate_plots.1 man page from debian package
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Martin Steigerwald [Wed, 20 May 2009 09:33:49 +0000 (11:33 +0200)]
man page fix to make lintian happy
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Zhang, Yanmin [Wed, 20 May 2009 09:30:55 +0000 (11:30 +0200)]
Add pre_read option
With this option set, files will be pre-read into memory before
starting the given IO operation(s).
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 May 2009 08:49:44 +0000 (10:49 +0200)]
Add appropriate warning in mmap engine for large maps failing
On 32-bit archs, this can trigger for large files. This should be fixed by
mapping lazily, for now just catch the error.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 May 2009 08:45:12 +0000 (10:45 +0200)]
Clear f on error get_next_file_rr()
Otherwise we could return an invalid non-open file and trigger and
assertion failure later in the process.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 19 May 2009 08:35:38 +0000 (10:35 +0200)]
Return EILSEQ on verify errors instead of EIO
Suggested by Randy Dunlap <randy.dunlap@oracle.com>
This makes it easier to differentiate between an error to read/write
from disk versus getting the wrong data back.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 18 May 2009 11:08:12 +0000 (13:08 +0200)]
Include transfer size in the clat/slat measurements
It's useful to know the block size that matches the given latency, so
add that to the end.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Fabio Checconi [Mon, 18 May 2009 10:54:50 +0000 (12:54 +0200)]
Fix garbling of clat/slat logs if a shared log file is used
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 13 May 2009 05:51:19 +0000 (07:51 +0200)]
Merge branch 'master' of ssh://router/data/git/fio
Randy Dunlap [Wed, 13 May 2009 05:51:05 +0000 (07:51 +0200)]
fio howto typo fixes
Fix a few typos in the fio HOWTO file.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Randy Dunlap [Wed, 13 May 2009 05:50:38 +0000 (07:50 +0200)]
fio docs corrections
Correct fio mailing list address and unit of time for cpuchunks.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 11 May 2009 08:37:33 +0000 (10:37 +0200)]
net engine: fix compilation warnings on Solaris
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 7 May 2009 12:21:59 +0000 (14:21 +0200)]
Fio 1.26.2
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 30 Apr 2009 06:07:43 +0000 (08:07 +0200)]
Merge branch 'master' of ssh://brick.kernel.dk/data/git/fio
Jens Axboe [Wed, 29 Apr 2009 07:52:10 +0000 (09:52 +0200)]
'V' is a running state, not an idle state. Correct HOWTO.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 29 Apr 2009 07:48:04 +0000 (09:48 +0200)]
Align io units to processor cache line size
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Radha Ramachandran [Wed, 29 Apr 2009 05:23:26 +0000 (07:23 +0200)]
Fix name of the slat log file
Currently the slat and clat logs go to the same clat log file when the
latency log file is specified.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Radha Ramachandran [Tue, 28 Apr 2009 06:12:56 +0000 (08:12 +0200)]
Don't populate random data for verify reads
I was seeing some performance drop during the read verification phase
of a test, and from the code in io_u.c in get_io_u function, we
prepare/populate the buffer in the io_u structure based on the verify
patterns/options.
This makes sense when we are doing writes, but I dont understand why
we do this for the read phase when this data is going to be
overwritten anyways(and in case of truncated reads, we do modify the
buf_len).
So based on that I changed the code to populate the buffer(io_u->buf)
only if its a write with verify enabled.
This works for my tests, but I do not know if there was a reason why
this was populated for reads as well to begin with.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 27 Apr 2009 06:05:45 +0000 (08:05 +0200)]
Fio 1.26.1
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 27 Apr 2009 06:00:48 +0000 (08:00 +0200)]
Filename must be parsed after directory option
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 24 Apr 2009 12:23:59 +0000 (14:23 +0200)]
Update bssplit documentation in fio.1
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Kedar Sovani [Thu, 23 Apr 2009 11:47:41 +0000 (13:47 +0200)]
fio: Include support for ARM
Assembly instructions picked from include/asm-arm/system.h
Signed-off-by: Kedar Sovani <kedars@marvell.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>