Jens Axboe [Thu, 8 Feb 2007 11:59:02 +0000 (12:59 +0100)]
[PATCH] Network engine cleanups and comments
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 7 Feb 2007 13:03:07 +0000 (14:03 +0100)]
[PATCH] Improve dependency checking on builds
From the similar fix for blktrace, from Vasily Tarasov <vtaras@openvz.org>.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 7 Feb 2007 12:59:21 +0000 (13:59 +0100)]
[PATCH] Enable _FORTIFY_SOURCE and stack protector checking
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 7 Feb 2007 12:58:53 +0000 (13:58 +0100)]
[PATCH] We should check system() return value
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 7 Feb 2007 12:55:38 +0000 (13:55 +0100)]
[PATCH] Remove SIGSEGV handler
We don't catch segfaults anymore, so remove that bit from the signal handler.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 7 Feb 2007 12:51:22 +0000 (13:51 +0100)]
[PATCH] aio example should use O_DIRECT IO
Otherwise it isn't async. Since fio defaults to buffered IO now, we need
to correct this example.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 7 Feb 2007 12:14:57 +0000 (13:14 +0100)]
[PATCH] Don't create files for engines that don't need them
Introduce FIO_NULLIO as a way to control that from the io engine.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 7 Feb 2007 10:30:37 +0000 (11:30 +0100)]
[PATCH] Misc fixes
- Change the hour/day eta display to be more clever.
- Style
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 7 Feb 2007 08:54:40 +0000 (09:54 +0100)]
[PATCH] Network engine fixes
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 7 Feb 2007 08:35:29 +0000 (09:35 +0100)]
[PATCH] Network io engine updates
- Use recv/send
- Set MSG_MORE on send if we are going to transmit more data
- Cleanups
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 6 Feb 2007 19:17:21 +0000 (20:17 +0100)]
[PATCH] Document ->queue() in the skeleton engine
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 6 Feb 2007 19:15:38 +0000 (20:15 +0100)]
[PATCH] Support residual io counts from io engines
We need this for requeuing support, the network engine makes this
pretty apparent (it's not unusual to see short tranfers there).
Basically we add an xfer_buf and xfer_buflen member to the io_u,
and these are the fields that the io engine MUST use. That allows
fio to increment and reset these appropriately, and simply requeue
the io_u for service of the next part of it.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 6 Feb 2007 18:27:27 +0000 (19:27 +0100)]
[PATCH] Support for > 1 network connection
Now the net ioengine also supports the nrfiles option to open
multiple connections.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 6 Feb 2007 18:16:01 +0000 (19:16 +0100)]
[PATCH] Network engine updates
A few comments, pass down port short instead of buffer.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 6 Feb 2007 13:43:52 +0000 (14:43 +0100)]
[PATCH] Simple support for networked IO
Adds a new ioengine, net. Use with ioengine=net, it supports only strict
reading or writing (no mixed reads/writes) to/from a single host.
The filename given must contain the host and port to connect to (or
listen from).
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 6 Feb 2007 12:29:55 +0000 (13:29 +0100)]
[PATCH] Rename io engines
Typing fio-engine-foo is annoying, just rename them to their type. They
already reside in a private directory, no point in putting the fio-engine
prefix on the files.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 6 Feb 2007 12:25:33 +0000 (13:25 +0100)]
[PATCH] Fix buffer alignment on non-4kb archs
Retrieve page size through sysconf.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 5 Feb 2007 14:04:46 +0000 (15:04 +0100)]
[PATCH] Add 'description' option
Can help provide a textual description of a job, if it's submitted
to someone for run/review.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 5 Feb 2007 09:52:10 +0000 (10:52 +0100)]
[PATCH] Fix sg ioengine
We need to do the ->init() call after having opened the file,
otherwise sg doesn't know what file type it is dealing with.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 5 Feb 2007 09:47:48 +0000 (10:47 +0100)]
[PATCH] Fix double free in sg engine error path
We need to clear td->io_ops->data if we free the structure,
otherwise the ->cleanup() handler will try to free it again.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 24 Jan 2007 08:21:31 +0000 (09:21 +0100)]
[PATCH] Fix typo
WRITE -> DDIR_WRITE
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 17 Jan 2007 06:42:30 +0000 (17:42 +1100)]
[PATCH] Only generate global bandwidth log if write_bw_log set
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 17 Jan 2007 06:23:11 +0000 (17:23 +1100)]
[PATCH] Add support for writing aggregate bandwidth log
This is the same that is shown while fio is running, it can be useful
for graphing purposes as well.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Sun, 14 Jan 2007 04:56:44 +0000 (05:56 +0100)]
[PATCH] Unlink write file if !overwrite and it exists
Repeatability across runs, unlink the file so that the job
runs like it wasn't there to begin with.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Sat, 13 Jan 2007 23:00:29 +0000 (00:00 +0100)]
[PATCH] Don't make file unlink the default
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Sat, 13 Jan 2007 22:59:00 +0000 (23:59 +0100)]
[PATCH] The last io depth entry is actually >= 64, not > 32
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Sat, 13 Jan 2007 22:56:33 +0000 (23:56 +0100)]
[PATCH] Include io depth distribution statistics
Calculate the distribution of io depths for the job run.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Sat, 13 Jan 2007 14:17:22 +0000 (15:17 +0100)]
[PATCH] Remove the time init check
We cannot control constructor ordering easily, hence the check
code was also in fio_gettime().
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Sat, 13 Jan 2007 14:15:47 +0000 (15:15 +0100)]
[PATCH] Shorten 'commitrate' to 'CR'
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Sat, 13 Jan 2007 11:29:13 +0000 (12:29 +0100)]
[PATCH] Fixup rate usec cycle calculation for small values
If someone set rate to 2KiB/sec or something low in that range,
we would integer divide by zero. Do calculation in reads per msec
instead of full seconds and fix the zero division.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Sat, 13 Jan 2007 11:20:30 +0000 (12:20 +0100)]
[PATCH] Improve rate functionality
usec_sleep() now takes into account the granularity of nanosleep().
Also stop using the remainder returned by nanosleep(), it doesn't
appear to always be accurate. Instead just check the elapsed time
manually.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 12 Jan 2007 08:04:41 +0000 (09:04 +0100)]
[PATCH] Properly shutdown in case of serious errors
Introduce fio_abort to do a forced shutdown of everything for things
like segfaults.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jan 2007 18:25:52 +0000 (19:25 +0100)]
[PATCH] Update HOWTO
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jan 2007 18:24:55 +0000 (19:24 +0100)]
[PATCH] Cleanup some output code
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jan 2007 14:59:35 +0000 (15:59 +0100)]
Merge branch 'master' of ssh://router/data/git/fio
Jens Axboe [Thu, 11 Jan 2007 14:59:26 +0000 (15:59 +0100)]
[PATCH] Update irange documentation
Also allow ':' as the range delimeter, and '/' as the
direction delimeter.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jan 2007 13:06:35 +0000 (14:06 +0100)]
[PATCH] Help: thinktime is in microseconds
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jan 2007 13:04:47 +0000 (14:04 +0100)]
[PATCH] Abort early for bad option given to --cmdhelp
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jan 2007 13:01:27 +0000 (14:01 +0100)]
[PATCH] Really fixup disk util init
Include disk io ticks, if the job itself creates the file. Don't
include it, if we create the file before running the job.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jan 2007 12:58:32 +0000 (13:58 +0100)]
Revert "[PATCH] Move disk utilization setup to thread itself"
This reverts commit
d52fb0c44567cf4fdc2a2c307c32672cc5a57170.
Jens Axboe [Thu, 11 Jan 2007 13:44:40 +0000 (14:44 +0100)]
[PATCH] Move disk utilization setup to thread itself
We don't want it to include potential file creation before the job runs.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jan 2007 13:40:27 +0000 (14:40 +0100)]
[PATCH] Warn if both option variable offset and callback given
If the callback is present, it is preferred to the offsets. So
giving offsets with a callback specified doesn't make sense.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jan 2007 13:36:56 +0000 (14:36 +0100)]
[PATCH] Pretty up option looping
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jan 2007 13:24:29 +0000 (14:24 +0100)]
[PATCH] Change output dev= to stdev=
It's really standard deviation, so make it a little easier to
understand.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jan 2007 12:24:44 +0000 (13:24 +0100)]
[PATCH] Change O_DIRECT vs buffered setup
Change the default from O_DIRECT IO to normal buffered IO. That makes
more sense, as O_DIRECT is a special case and should be manually
enabled as such.
Do this by adding a option negate switch, so we don't need two sets
of parameters to control these options.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 11 Jan 2007 10:04:31 +0000 (11:04 +0100)]
[PATCH] Change timeout to runtime and rearrange option entries
Runtime is a more descriptive name - the documentation is updated
as well, however the parsing has added an alias functionality to
support older names as well. So 'timeout' still works.
Also rearrange option entries, now they matter because the --cmdhelp
option will print them in order.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jan 2007 19:30:53 +0000 (20:30 +0100)]
[PATCH] Thread option didn't work
We stored to the wrong location, woops.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jan 2007 18:56:37 +0000 (19:56 +0100)]
[PATCH] Implement a better mean and standard deviation calculation
Based on Knuth's, volume 2.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jan 2007 15:02:32 +0000 (16:02 +0100)]
[PATCH] strtoul() -> strtol() conversion
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jan 2007 14:41:07 +0000 (15:41 +0100)]
Merge branch 'master' of ssh://git.kernel.dk/data/git/fio
Jens Axboe [Wed, 10 Jan 2007 14:39:51 +0000 (15:39 +0100)]
[PATCH] Fix standard deviation calculation
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jan 2007 12:41:26 +0000 (13:41 +0100)]
[PATCH] Be strict in checking option ranges
Fail if given value falls outside of range instead of adjusting
it automatically.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jan 2007 12:20:09 +0000 (13:20 +0100)]
[PATCH] Option minval should be signed
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jan 2007 12:14:02 +0000 (13:14 +0100)]
[PATCH] Introduce bool option type
We can automatically flag those with min/max values.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jan 2007 11:26:18 +0000 (12:26 +0100)]
[PATCH] Show allowed option values for string matching
Helpful to avoid looking in the documentation.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jan 2007 10:41:11 +0000 (11:41 +0100)]
[PATCH] Pretty up command help output
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jan 2007 10:33:14 +0000 (11:33 +0100)]
[PATCH] Reversed order of thinktime_blocks modulo operation
We wish to find the remainder of block count vs thinktime_blocks,
not the other way around. This makes thinktime_blocks work correctly.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jan 2007 10:27:48 +0000 (11:27 +0100)]
[PATCH] Fixup ratecycle option
It set .name twice, the last one was supposed to be .help instead.
Also kill options->opt_set, it's not used.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 10 Jan 2007 10:23:16 +0000 (11:23 +0100)]
[PATCH] Add default option values
It's handy for the help text, and it also cleans up init.c by removing
a bunch of default option defines.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jan 2007 20:22:18 +0000 (21:22 +0100)]
[PATCH] Move timeout structure where it is used
Also makes sure that it's always initialized to 0's.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jan 2007 20:22:02 +0000 (21:22 +0100)]
[PATCH] Complain if bad option given to 'cmdhelp'
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jan 2007 20:20:13 +0000 (21:20 +0100)]
[PATCH] Add per-option help
It's hard to remember all the options that fio supports, so add a
'cmdhelp' option that either prints all possible options (if argument
is 'all'), or specific option help (if argument is the command).
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 9 Jan 2007 07:19:01 +0000 (08:19 +0100)]
[PATCH] Fio version 1.11
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 8 Jan 2007 21:27:05 +0000 (22:27 +0100)]
[PATCH] Fix busy-looping with aio engine and depth > 1
It's ok for io_getevents() to return 0, if min_nr is 0. Check
first if io_getevents() returns >= min_nr, if so then just break
and return.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 8 Jan 2007 09:57:27 +0000 (10:57 +0100)]
[PATCH] Add extra disk util name check
If it's the same device name, don't print utilization more than once.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 8 Jan 2007 09:49:38 +0000 (10:49 +0100)]
[PATCH] Debug printf
Kill it.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 8 Jan 2007 09:47:43 +0000 (10:47 +0100)]
[PATCH] Unlink and directory fix
Make 'unlink' a bool and fix bug in directory setup.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Mon, 8 Jan 2007 09:25:51 +0000 (10:25 +0100)]
[PATCH] Non-overwrite random write fix
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jan 2007 20:02:41 +0000 (21:02 +0100)]
[PATCH] Make note of mailing list address
Yay, there's a list.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jan 2007 19:45:09 +0000 (20:45 +0100)]
[PATCH] Make sure the fio_options[] members are always zero filled
Make sure the remaining members are cleared, to avoid confusion.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jan 2007 19:43:19 +0000 (20:43 +0100)]
[PATCH] Add thinktime_blocks
Right now 'thinktime' waits after every IO. This may not be what you
need, sometimes you want to wait only after 'x' number of ios. This
options allows you to set the number of ios to complete, before waiting.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jan 2007 13:21:13 +0000 (14:21 +0100)]
[PATCH] Fix deviation overflow
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 3 Jan 2007 10:10:27 +0000 (11:10 +0100)]
[PATCH] Fix division by zero in deviance calculation
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Pascal Bleser [Tue, 2 Jan 2007 08:17:03 +0000 (09:17 +0100)]
[PATCH] Split CFLAGS into OPTFLAGS + CFLAGS
Makes it easier to control rpm generation.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 2 Jan 2007 08:15:36 +0000 (09:15 +0100)]
[PATCH] Don't create libdir
Since we link in io engines, we don't need to create the libdir anymore.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 22 Dec 2006 20:02:44 +0000 (21:02 +0100)]
[PATCH] Don't dump optarg on unknown option
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 22 Dec 2006 19:54:06 +0000 (20:54 +0100)]
[PATCH] Version 1.10
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 21 Dec 2006 08:50:00 +0000 (09:50 +0100)]
[PATCH] Allow mem=mmap to also use a file backing
We have the stuff in place for huge page backed memory, so it's little
extra code to support io buffers inside mmap file backed memory.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 Dec 2006 19:28:33 +0000 (20:28 +0100)]
[PATCH] Warning on hugepage-size setting in HOWTO
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 Dec 2006 19:27:36 +0000 (20:27 +0100)]
[PATCH] Add hugepage-size option
We default to 4MiB, but export the setting so it can be changed.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 Dec 2006 19:15:41 +0000 (20:15 +0100)]
[PATCH] Document how to setup/use huge pages
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 Dec 2006 13:42:17 +0000 (14:42 +0100)]
[PATCH] Remove debug fprintfs
Oops, leftover from the mmaphuge commit.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 Dec 2006 13:40:44 +0000 (14:40 +0100)]
[PATCH] Support for mmap of hugetlb files as memory backing
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 Dec 2006 11:54:25 +0000 (12:54 +0100)]
[PATCH] Fix character + O_DIRECT
Just clear ->odirect if it's a character device, it doesn't apply.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 Dec 2006 11:48:23 +0000 (12:48 +0100)]
[PATCH] Fix hugetlb problems
Alignment was bad, and we need to check the shmhuge string before shm,
otherwise it'll match the latter.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Wed, 20 Dec 2006 09:58:07 +0000 (10:58 +0100)]
[PATCH] Must use ->real_file_size
->file_size is the size of the file from offset, while ->real_file_size
is the total size of the file. So when comparing EOF, we should look
at ->real_file_size against the offset.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 19 Dec 2006 14:18:14 +0000 (15:18 +0100)]
[PATCH] Support for hugetlb backed shared memory
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 15 Dec 2006 15:01:54 +0000 (16:01 +0100)]
[PATCH] Only unlink for created files
If the file was already there, don't unlink it.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 15 Dec 2006 12:11:24 +0000 (13:11 +0100)]
[PATCH] Make unlink and overwrite the default
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 5 Dec 2006 10:56:56 +0000 (11:56 +0100)]
[PATCH] Fix FreeBSD/Solaris linkage
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Tue, 5 Dec 2006 10:44:16 +0000 (11:44 +0100)]
[PATCH] Fix warnings from icc
icc spewed a bunch of warnings on building fio, but it did actually build
and work. Some of them are real bugs, most are just "helpful" warnings.
icc doesn't like pointer arithmetic, however these are not fixed up. It
works as-is, just ignore those class of warnings.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 24 Nov 2006 12:29:20 +0000 (13:29 +0100)]
[PATCH] rusage stat
Sum up total over all loops, not for each one.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 24 Nov 2006 12:12:56 +0000 (13:12 +0100)]
[PATCH] Improve time accounting for a job
We need to account loop time in usecs, or the null io engine can
get the timing wrong. For the system usage stats, don't use summed
up loop counts, but calculate end-start time instead.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 24 Nov 2006 11:43:49 +0000 (12:43 +0100)]
[PATCH] Update ioengine help text
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 24 Nov 2006 11:37:34 +0000 (12:37 +0100)]
[PATCH] 'null' ioengine
Add a null io engine, which can be used to exercise and test/debug fio
itself. It doesn't transfer any data, just pretends to.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 24 Nov 2006 09:54:06 +0000 (10:54 +0100)]
[PATCH] Fix time debug compile
If FIO_TIME_DEBUG gets enabled, we need hash.h.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 24 Nov 2006 09:47:40 +0000 (10:47 +0100)]
[PATCH] Comment typo
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Fri, 24 Nov 2006 09:42:00 +0000 (10:42 +0100)]
[PATCH] Time and seek optimizations
We did too many gettimeofday() calls, this patch cuts the number by
40%. Use clock_gettime() MONOTONIC instead, it is faster on my system
at least.
This patch also optimizes calling lseek() only when necessary for the
sync io engine.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 23 Nov 2006 14:11:53 +0000 (15:11 +0100)]
[PATCH] Get rid of FIO_INST_PREFIX
We don't need it anymore, since we link in known io engines and unknown
ones have to be given with a full path.
Also fixup Makefiles for Solaris and FreeBSD, they were a bit outdated.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Jens Axboe [Thu, 23 Nov 2006 14:01:19 +0000 (15:01 +0100)]
[PATCH] Close files on error
The error path doesn't close already opened files. Also set
O_SYNC regardless of file type.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>