Jens Axboe [Thu, 11 Apr 2013 16:44:33 +0000 (18:44 +0200)]
Makefile: clean $(FIO_OBJS) too
Otherwise fio.o gets left behind after a clean.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 13:03:58 +0000 (15:03 +0200)]
README: update
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 12:20:33 +0000 (14:20 +0200)]
Fix a few 4.8 extra anal warnings
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 11:48:31 +0000 (13:48 +0200)]
Get rid of cast between enums
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 11:48:13 +0000 (13:48 +0200)]
libaio: fix double setting of option category
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 11:46:37 +0000 (13:46 +0200)]
net: fix double setting of option category
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 11:45:51 +0000 (13:45 +0200)]
client: move jobs_eta to the end of client_eta
It has a variably length field.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 10:55:16 +0000 (12:55 +0200)]
client/server: pass ->unit_base properly
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 10:43:05 +0000 (12:43 +0200)]
client/server: pass je->rate[] and je->iops[] over the net
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 10:31:29 +0000 (12:31 +0200)]
Merge branch 'gfio' into gfio-int
Conflicts:
backend.c
fio.c
fio.h
init.c
libfio.c
options.c
thread_options.h
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 10:26:55 +0000 (12:26 +0200)]
to_kmg() -> fio_uint_to_kmg()
It's being exported.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 10:25:17 +0000 (12:25 +0200)]
Move t/jobs/ to .fio format
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 10:24:36 +0000 (12:24 +0200)]
Make io_sample word size agnostic
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 14 Mar 2012 10:39:13 +0000 (11:39 +0100)]
iolog: note type of log
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Conflicts:
backend.c
init.c
iolog.h
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 10:11:34 +0000 (12:11 +0200)]
init: kill unused mlock_size
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 10:04:38 +0000 (12:04 +0200)]
Make nr_thread/nr_process private to backend
No need export that knowledge. Merge from gfio.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 10:01:48 +0000 (12:01 +0200)]
backend: merge in td->o to o-> thread_main() change
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 09:54:08 +0000 (11:54 +0200)]
Rename job files
gfio uses a .fio or .ini ending.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 09:51:35 +0000 (11:51 +0200)]
Move init code to libfio
More merge from gfio.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 11 Apr 2013 06:53:24 +0000 (08:53 +0200)]
ioengine: import whitespace changes
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 28 Mar 2012 18:50:15 +0000 (20:50 +0200)]
Make lockmem a per job option
We need to get rid of per job options that fiddle with global
state. It's confusing, and it breaks remote option handling.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Conflicts:
backend.c
fio.h
init.c
options.c
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 20:23:40 +0000 (22:23 +0200)]
Merge in crc32c-intel probe
Then we can kill the option callback.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 20:19:46 +0000 (22:19 +0200)]
Merge in zipf differences from gfio
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 20:16:01 +0000 (22:16 +0200)]
Merge missing content to man page
We missed the two CPU engine private options.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 20:13:30 +0000 (22:13 +0200)]
README: remove partial option list
The HOWTO or man page should be used, since either of those
has the full list of option keywords.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 18:48:31 +0000 (20:48 +0200)]
Move rbtree to lib/
That's where it was moved in gfio, one further step towards unifying
the branches.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 17:50:32 +0000 (19:50 +0200)]
Merge branch 'master' into gfio
Conflicts:
engines/libaio.c
engines/net.c
iolog.h
options.c
options.h
parse.c
parse.h
thread_options.h
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 17:45:30 +0000 (19:45 +0200)]
tiobench: remove duplicate profile group
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 17:43:59 +0000 (19:43 +0200)]
Categorize engine and profile options
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 17:35:25 +0000 (19:35 +0200)]
parse: fix spelling error
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 17:32:20 +0000 (19:32 +0200)]
parse: bring in more stuff from gfio
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 17:30:50 +0000 (19:30 +0200)]
Unify options with gfio
Add the parser features and unify the option table.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 13:01:42 +0000 (15:01 +0200)]
Add thread_options.h
The options are always problematic to merge between the master
repo and gfio. So import this bit from gfio.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 12:41:08 +0000 (14:41 +0200)]
configure: gfio wording
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 11:08:02 +0000 (13:08 +0200)]
Add group/category to numa/mem node options
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 11:07:43 +0000 (13:07 +0200)]
parse: set invalid group on option with no group set
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 11:05:23 +0000 (13:05 +0200)]
parse: catch all options without a group/category set
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 11:01:30 +0000 (13:01 +0200)]
gfio: add option group/category to filename_format
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 09:21:33 +0000 (11:21 +0200)]
parse: don't print min/max option ranges if posval[] is set
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 09:12:21 +0000 (11:12 +0200)]
Merge branch 'master' into gfio
Conflicts:
options.c
parse.c
server.h
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 09:06:59 +0000 (11:06 +0200)]
Convert unit_base to posval[]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Wed, 10 Apr 2013 09:06:34 +0000 (11:06 +0200)]
parse: for option with posval, dump allowed values on failure
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Tue, 9 Apr 2013 19:05:14 +0000 (21:05 +0200)]
options: make unit_base be posval[] based
We can kill the verifier and ensure we only get the
possible values we need.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Tue, 9 Apr 2013 19:02:15 +0000 (21:02 +0200)]
parse: add posval support to FIO_OPT_INT
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Tue, 9 Apr 2013 19:02:15 +0000 (21:02 +0200)]
parse: add posval support to FIO_OPT_INT
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Tue, 9 Apr 2013 18:49:45 +0000 (20:49 +0200)]
options: make unit_base be posval[] based
We can kill the verifier and ensure we only get the
possible values we need.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Tue, 9 Apr 2013 18:49:34 +0000 (20:49 +0200)]
gfio: add support for FIO_OPT_INT options with posval
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Tue, 9 Apr 2013 18:38:32 +0000 (20:38 +0200)]
gfio: update new options for gfio opt format
And fix gclient.c updates for num2str().
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Tue, 9 Apr 2013 18:30:04 +0000 (20:30 +0200)]
server: bump protocol version
Commit
ad705bcb adds member to protocol structures, bump it.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Tue, 9 Apr 2013 18:29:29 +0000 (20:29 +0200)]
server: bump protocol version
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Tue, 9 Apr 2013 18:29:16 +0000 (20:29 +0200)]
Merge branch 'master' into gfio
Conflicts:
eta.c
fio.h
init.c
options.c
stat.c
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Steven Noonan [Mon, 8 Apr 2013 22:40:13 +0000 (15:40 -0700)]
stat.c: make 'bw' summary line respect 'unit_base' option
Signed-off-by: Steven Noonan <steven@uplinklabs.net>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Steven Noonan [Mon, 8 Apr 2013 22:25:54 +0000 (15:25 -0700)]
stats: show summary bandwidth in terms of kb_base
Signed-off-by: Steven Noonan <steven@uplinklabs.net>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Steven Noonan [Mon, 8 Apr 2013 22:05:25 +0000 (15:05 -0700)]
implement 'unit_base' option to select between KB and Kbit et. al.
With network testing, it's often desirable to measure in terms of
kilobits/megabits rather than kilobytes/megabytes. This adds an option named
'unit_base' which can be set to either '1' or '8', where '1' means represent in
terms of bits and '8' means to represent rate in terms of bytes.
Signed-off-by: Steven Noonan <steven@uplinklabs.net>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Steven Noonan [Fri, 5 Apr 2013 23:57:38 +0000 (16:57 -0700)]
num2str: add arguments to represent values in terms of bytes/bits
This allows for representing I/O rates in terms of e.g. megabits (Mb) versus
megabytes (MB).
Signed-off-by: Steven Noonan <steven@uplinklabs.net>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Mon, 8 Apr 2013 07:46:54 +0000 (09:46 +0200)]
Merge branch 'next'
Jens Axboe [Mon, 8 Apr 2013 07:44:38 +0000 (09:44 +0200)]
Fio 2.0.15
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 5 Apr 2013 20:13:42 +0000 (22:13 +0200)]
Makefile: use ginstall on Solaris
Regular install just doesn't seem to like man pages?!
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 5 Apr 2013 20:13:42 +0000 (22:13 +0200)]
Makefile: use ginstall on Solaris
Regular install just doesn't seem to like man pages?!
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 5 Apr 2013 14:32:33 +0000 (16:32 +0200)]
Man page formatting fixups
- Fix extra .B in the filename_format section
- Fix missing .RE for lockfile
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 5 Apr 2013 12:44:51 +0000 (14:44 +0200)]
Use strcasestr() for matching filename_prefix keywords
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 5 Apr 2013 12:44:03 +0000 (14:44 +0200)]
Add strcasestr()
We'll need it for the next commit, if the platform does not
provide it.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 5 Apr 2013 10:55:42 +0000 (12:55 +0200)]
configure: check for zlib
It's required for gfio, even if gfio isn't enabled. The client
and server uses compression for larger messages.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 5 Apr 2013 10:54:03 +0000 (12:54 +0200)]
Fix merge error
Fails link because we attempt to link in fio.o twice.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 5 Apr 2013 10:46:54 +0000 (12:46 +0200)]
Merge branch 'master' into gfio
Conflicts:
Makefile
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 5 Apr 2013 09:09:20 +0000 (11:09 +0200)]
Add filename_format option
Allows the user to set how fio should generate the file names
used by jobs. The default is $jobname.$jobnumber.$filenumber
as it always has been, but now it's configurable as well. See
the included documentation.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Steven Noonan [Thu, 4 Apr 2013 07:05:56 +0000 (09:05 +0200)]
gitignore: ignore configure script outputs
Signed-off-by: Steven Noonan <steven@uplinklabs.net>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Shaohua Li [Wed, 3 Apr 2013 06:40:17 +0000 (08:40 +0200)]
Fixup json bandwidth output
The final version sent in for inclusion had the check mixed
up, so it doesn't output bandwidth unless the calculation
failed.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 29 Mar 2013 18:20:51 +0000 (12:20 -0600)]
parse: fix misparse of bs=64k-128k
We failed parsing the postfix for this case, resulting in
64 and 128k being the result (instead of 64k and 128k).
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 29 Mar 2013 18:08:40 +0000 (12:08 -0600)]
axmap: get rid of old debug ->fail_ok checking
Don't assert() on not being able to set a bit. We are returning
the number of bits we set anyway, so just have it return 0.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bruce Cran [Fri, 29 Mar 2013 16:17:25 +0000 (16:17 +0000)]
Workaround pthreads-win32 pthread_rwlock_init limitation.
On Windows, pthreads-win32 returns EINVAL if you specify an attribute when
calling pthread_rwlock_init. Work around this by only specifying the attribute
when FIO_HAVE_PSHARED_MUTEX is defined.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bruce Cran [Fri, 29 Mar 2013 16:02:53 +0000 (16:02 +0000)]
Fix rwlock error messages to specify the correct function names.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 29 Mar 2013 15:14:23 +0000 (09:14 -0600)]
t/axmap: add test for multi bit sets
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 29 Mar 2013 14:56:09 +0000 (08:56 -0600)]
axmap: fix bug with max_bs/min_bs ratio being > 64
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 29 Mar 2013 14:55:32 +0000 (08:55 -0600)]
configure: fixup build so that it's more resilient
Allow people to clear/set EXTFLAGS/OPTFLAGS at will.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 28 Mar 2013 21:08:49 +0000 (15:08 -0600)]
Fix usr/sys/ctx/majf/minf for -USR1 usage
Let the threads themselves update their stats.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 28 Mar 2013 18:52:17 +0000 (12:52 -0600)]
Fix build if just one of libverbs or librdma is installed
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Shaohua Li [Thu, 28 Mar 2013 14:28:51 +0000 (08:28 -0600)]
mutex: set pshared attributes on rw semaphore
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Mon, 25 Mar 2013 19:20:08 +0000 (13:20 -0600)]
configure: check for v2 of libnuma
We use options that require 2.x.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Sat, 23 Mar 2013 04:56:55 +0000 (22:56 -0600)]
Consider the maximum block size difference the minimum for loop exit
For mixed block sizes, we can hit the condition where we decided
to stop even if we could have done a small block size. Don't do
another loop for those cases.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Fri, 22 Mar 2013 21:23:20 +0000 (15:23 -0600)]
posixaio: restart suspend list after we have used aio_suspend()
Otherwise we could both end up putting entries in there twice,
if there are still in flight, or having done entries in there too.
So restart the suspend list after aio_suspend(), so we only put
truly still-pending ios in there.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jianpeng Ma [Fri, 22 Mar 2013 13:16:32 +0000 (07:16 -0600)]
mutex: fix the expression for checking the mutext magic
Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 21 Mar 2013 16:06:58 +0000 (10:06 -0600)]
Only attempt file unlock if we use locking
Fixes a segfault on exit, if file locking isn't used.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 21 Mar 2013 16:05:07 +0000 (10:05 -0600)]
mutex: add magic checks
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 21 Mar 2013 12:45:27 +0000 (06:45 -0600)]
Merge branch 'master' into gfio
Conflicts:
Makefile
fio.h
options.c
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 21 Mar 2013 12:23:14 +0000 (06:23 -0600)]
Kill now unused lockfile_batch variable
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Thu, 21 Mar 2013 01:57:01 +0000 (19:57 -0600)]
Rework lockfile= file lock handling
Get rid of the hand rolled rw semaphores, just use pthread
rwlocks instead. Kill the batching too, it was broken by
default, so nobody could have been using it.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Tue, 19 Mar 2013 16:47:15 +0000 (10:47 -0600)]
lib/getopt_long: clear 'optarg' every time
We don't want a stale optarg argument to be available.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Tue, 19 Mar 2013 13:49:35 +0000 (07:49 -0600)]
Fix re-run issue when job size is set to "unknown"
Assert that td->o.size is valid before comparing the IO amount
we did to it. This prevents re-running a job because we think
there's more work to do. But if we haven't set ->o.size to a
valid value, then that must not determine whether to re-run
the loop or not.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Tue, 19 Mar 2013 13:48:46 +0000 (07:48 -0600)]
A few ULL -> uint64_t conversions
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jens Axboe [Mon, 18 Mar 2013 21:34:41 +0000 (15:34 -0600)]
Fixup commit error in 56c796
I hand-fondled it, forgot to revert.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Aaron Carroll [Thu, 14 Mar 2013 05:57:41 +0000 (16:57 +1100)]
Android: remove static Android configuration
Android can now be configured dynamically, so remove the static config.
For this we need to modify the socklen_t check #includes which appear to
be wrong anyway.
Signed-off-by: Aaron Carroll <aaronc@cse.unsw.edu.au>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Aaron Carroll [Thu, 14 Mar 2013 05:57:40 +0000 (16:57 +1100)]
configure: endianness check for cross compile
- Add a cross compile check: assume yes if we can't run a test binary
- If cross compiling, revert to a compile-time endianess check. This tries
a few possible ways to detect big endian, but otherwise assumes little. We
rely on the run-time check to save us if the build-time check is wrong.
Signed-off-by: Aaron Carroll <aaronc@cse.unsw.edu.au>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Aaron Carroll [Thu, 14 Mar 2013 05:57:39 +0000 (16:57 +1100)]
configure: compile-time word size detection
Signed-off-by: Aaron Carroll <aaronc@cse.unsw.edu.au>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Aaron Carroll [Thu, 14 Mar 2013 05:57:38 +0000 (16:57 +1100)]
Propagate target OS from configure to Makefile
Propagate target OS from configure to Makefile (CONFIG_TARGET_OS) to avoid
duplicating the check in Makefile. Also allow CROSS_COMPILE for the toolchain
prefix, which is common (e.g. Linux).
Signed-off-by: Aaron Carroll <aaronc@cse.unsw.edu.au>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Aaron Carroll [Wed, 13 Mar 2013 09:09:10 +0000 (10:09 +0100)]
android: fix compile by not defining FIO_NO_HAVE_SHM_H twice
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Martin Steigerwald [Mon, 11 Mar 2013 15:22:56 +0000 (16:22 +0100)]
Make test for gnuplot work with empty strings.
When I launch fio_generate_plots on a system where gnuplot is not installed I
get this error :
$ fio_generate_plots test
Making bw logs
/usr/bin/fio_generate_plots: 42: /usr/bin/fio_generate_plots: -: not found
That's because the test checking whether gnuplot is installed is failing
because of an empty variable :
GNUPLOT=$(which gnuplot)
if [ ! -x $GNUPLOT ]; then
echo You need gnuplot installed to generate graphs
exit 1
fi
Indeed the command "test -x" is exiting with code 0.
To correct this we should enclose the variable with quotes :
if [ ! -x "$GNUPLOT" ]; then
Then the script is going into the test properly :
$ fio_generate_plots test
You need gnuplot installed to generate graphs
I copied problem description by Hervé from the Debian bug report below.
Fixes:
Bug#700580: /usr/bin/fio_generate_plots: -: not found
http://bugs.debian.org/700580
Reported-By: Hervé Werner <hwerner@score-md.com>
Tested-By: Hervé Werner <hwerner@score-md.com>
Tested-By: Martin Steigerwald <ms@teamix.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Aaron Carroll [Tue, 12 Mar 2013 04:46:10 +0000 (15:46 +1100)]
Android: add ioprio support
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Alex Pyrgiotis [Tue, 12 Mar 2013 08:28:35 +0000 (10:28 +0200)]
lfsr: fix verification and spin bugs
Changes:
1. Verification now works properly and reports for which value it fails
2. Fix a mishandling of spin incrementation which led to multiple
calculation of the same values
Signed-off-by: Alex Pyrgiotis <apyrgio@grnet.gr>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Oleg [Tue, 12 Mar 2013 07:06:53 +0000 (00:06 -0700)]
Conditionally include <sys/shm.h> based on OS support. Define missing MAP_HUGETLB for Android.
Signed-off-by: Oleg Matcovschi <olegmatcovsky@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>