Doug Kwan [Fri, 6 Jun 2014 22:29:11 +0000 (15:29 -0700)]
Add aarch64 support.
This patch adds support for aarch64 for fio. This is tested by building
FIO on an aarch64 Ubuntu system and running some samepls.
Signed-off-by: Doug Kwan <dougkwan@google.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Tue, 3 Jun 2014 14:36:20 +0000 (08:36 -0600)]
glusterfs: remove glfs-handles.h include
We don't seem to need it, and newer packages don't include
it.
Signed-off-by: Jens Axboe <axboe@fb.com>
Haomai Wang [Wed, 21 May 2014 08:30:55 +0000 (16:30 +0800)]
Make rbd engine use aio flush
The fio's option "fsync" can make rbd engine issue sync request. In practice,
we would like to use aio_rbd_flush to make it non-block. So it will be more
reliable to use fio to evaluate actual io scene like qemu.
Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Rik Faith [Thu, 29 May 2014 20:13:39 +0000 (14:13 -0600)]
ETA no longer works when -o specified
Regarding fio version fio-2.1.9-17-gd9b1 (after April 11, 2014):
A recent change (
5e1d8745471d1f315113c5c3fd8b93809d9ddff9) caused stdout
to be closed when -o was specified, which prevents ETA output from showing
up on stdout even when, e.g., json is being saved in the file.
This patch continues to close f_out so that there is no leak if multiple -o
options are specified, but doesn't close stdout. Hence, ETA works again
when it should.
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Tue, 20 May 2014 01:57:05 +0000 (19:57 -0600)]
Add ->invalidate() IO engine ops
Allow IO engines to plug in their own invalidate cache handler.
This fixes an issue on rbd, where we attempt to invalidate the
cache, but fail because it's not a valid file descriptor.
Reported-by: xan.peng@gmail.com
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Tue, 20 May 2014 01:53:03 +0000 (19:53 -0600)]
gfapi: indentify
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 19 May 2014 16:12:56 +0000 (10:12 -0600)]
Run the gluster engines through indent
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 19 May 2014 15:50:54 +0000 (09:50 -0600)]
Merge branch 'master' of git://github.com/cvubrugier/fio
Jens Axboe [Mon, 19 May 2014 15:49:56 +0000 (09:49 -0600)]
Merge branch 'master' of git://github.com/rootfs/fio
Signed-off-by: Jens Axboe <axboe@fb.com>
Conflicts:
configure
Xan Peng [Mon, 19 May 2014 15:39:00 +0000 (09:39 -0600)]
Fix issue with dangling pointer in the rbd engine
Without this patch, after _fio_rbd_connect() failed, fio will hang for a few seconds, display the following logs and then quit:
====console output start====
rbd_iodepth32: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=rbd, iodepth=32
fio-2.1.9
Starting 1 process
rbd engine: RBD version: 0.1.8
rbd engine: RBD version: 0.1.8
rbd_open failed.
fio_rbd_connect failed.
common/Mutex.cc: In function 'void Mutex::Lock(bool)' thread
7f6e337aa780 time 2014-05-17 19:23:23.417285
common/Mutex.cc: 93: FAILED assert(r == 0)
ceph version 0.80 (
b78644e)
1: (()+0x254093) [0x7f6e32502093]
2: (librados::RadosClient::shutdown()+0x1f) [0x7f6e324741df]
3: (rados_shutdown()+0x9) [0x7f6e32459209]
4: ./fio() [0x456a46]
5: (setup_files()+0xc2) [0x420872]
6: ./fio() [0x449c15]
7: (fio_backend()+0x115) [0x44a515]
8: (__libc_start_main()+0xf5) [0x7f6e311f3455]
9: ./fio() [0x40ad3c]
NOTE: a copy of the executable, or objdump -rdS is needed to interpret this.
terminate called after throwing an instance of 'ceph::FailedAssertion'
Aborted
====console output end====
With this patch, after _fio_rbd_connect() failed, fio will quit at once:
====console output start====
rbd_iodepth32: (g=0): rw=randwrite, bs=4K-4K/4K-4K/4K-4K, ioengine=rbd, iodepth=32
fio-2.1.9
Starting 1 process
rbd engine: RBD version: 0.1.8
rbd_open failed.
fio_rbd_connect failed.
Run status group 0 (all jobs):
====console output end====
Signed-off-by: Xan Peng <xanpeng@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 12 May 2014 23:18:13 +0000 (17:18 -0600)]
Fio 2.1.9
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Wed, 7 May 2014 01:06:51 +0000 (19:06 -0600)]
verify: fix verification of uncompleted or errored ios
Stoo Davies reports:
------
I'm doing some powerfail recovery testing on a storage array over iSCSI.
Host is RHEL 6.4 kernel 2.6.32-358.el6.x86_64.
With fio 2.1.2 -> 2.1.4 the job file below rides through the disks going
away, and continues I/O after they come back, without reporting any
errors. With fio 2.1.5 -> 2.1.8 when the disks come back fio
immediately reports a meta verification error.
I captured a trace with an finisar analyzer, and can see that after the
disks come back and the host logs back in, a read is issued for an lba
which was never written to. Since I don't see verification errors
outside of the powerfail testing, I suspect fio isn't correctly handling
failed writes during the time the disks are unavailable.
------
The issue is caused by a change between 2.1.4 and 2.1.5 that started
logging IO for verification at issue time, instead of at completion
time. If the IO only partially completes or completes in error, then
we need to adjust the verification unit so that we verify only what
fio actually wrote.
Signed-off-by: Jens Axboe <axboe@fb.com>
Daniel Gollub [Thu, 1 May 2014 15:07:49 +0000 (17:07 +0200)]
Avoid buildenv conditional in thread_option struct
Managed to run into issues with an external ioengine
which got build with CONFIG_LIBNUMA not defined. Fio
itself got build with CONFIG_LIBNUMA this resulted
in different struct members offsets in the two different
ELF objects. Causing crashes due to invalidate offsets
inside the thread_data structure (e.g. td->io_ops->data).
Ideally all structs which might be used by external
ioengines should be independent of buildenv conditionals
like CONFIG_LIBNUMA or others.
Removed the CONFIG_LIBNUMA in thread_options.h and replaced
the libnuma specific "struct bitmask" members with strings
which hold the option's input value. This should also make
the marshaling/demarshaling in cconv.c easier.
(Note: the NUMA bits are not handled in cconv.c at the
moment. And not part of the thread_options_packed struct)
Signed-off-by: Daniel Gollub <daniel.gollub@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Daniel Gollub [Thu, 1 May 2014 12:07:04 +0000 (14:07 +0200)]
Avoid double declaration of leXX_to_cpu
for external ioengines. leXX_to_cpu is a very common
declaration - e.g. Ceph is using this as well.
Signed-off-by: Daniel Gollub <daniel.gollub@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Daniel Gollub [Thu, 1 May 2014 12:07:03 +0000 (14:07 +0200)]
Avoid double-declaration of ARRAY_SIZE
for external ioengines. ARRAY_SIZE is a very common
declaration - e.g. Ceph is using this as well.
Signed-off-by: Daniel Gollub <daniel.gollub@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Daniel Gollub [Thu, 1 May 2014 12:07:02 +0000 (14:07 +0200)]
distinguish internal/ext. builds with FIO_INTERNAL
To distinguish between internal and external includes of
fio.h and other headers FIO_INTERNAL get applied only for
the fio (and gfio) internal build itself.
This helps to prevent double-declaration issues by making
ambigous declaration with common names coditional based on
the FIO_INTERNAL define.
Signed-off-by: Daniel Gollub <daniel.gollub@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Daniel Gollub [Thu, 1 May 2014 12:07:01 +0000 (14:07 +0200)]
Replace CONFIG_CPU_COUNT with plain CPU_COUNT
Fixes double-declartion issue of CPU_COUNT when fio.h is included by
third-party projects.
Checking for a define of CPU_COUNT seems to be a common approach
to deal with different versions of sched.h
Signed-off-by: Daniel Gollub <daniel.gollub@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Danny Al-Gaaf [Wed, 30 Apr 2014 18:48:16 +0000 (20:48 +0200)]
crc/test.c: fix include of time.h
Replace include of time.h with renamed fio_time.h to fix build.
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Wed, 30 Apr 2014 21:01:00 +0000 (15:01 -0600)]
Makefile: clean all *.d dependency files on make clean
Signed-off-by: Jens Axboe <axboe@fb.com>
Daniel Gollub [Wed, 30 Apr 2014 09:25:04 +0000 (11:25 +0200)]
Rename time.h for third-party include of fio.h
External project might build external ioengines and need to
include there for fio.h. If a project set the include path
to the fio root source directory and the third-party source
holds an "time.h" as well things get complicated.
Signed-off-by: Daniel Gollub <d.gollub@telekom.de>
Cc: Daniel Gollub <daniel.gollub@t-online.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Daniel Gollub [Wed, 30 Apr 2014 06:27:53 +0000 (08:27 +0200)]
Fix gfio build due to disk_stat struct seperation
Signed-off-by: Daniel Gollub <d.gollub@telekom.de>
Cc: Daniel Gollub <daniel.gollub@t-online.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Bruce Cran [Mon, 28 Apr 2014 21:54:43 +0000 (15:54 -0600)]
Windows: always open files in binary mode
On Windows, open() defaults to text mode, causing control characters
to be interpreted. Avoid this by passing the _O_BINARY flag.
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 28 Apr 2014 18:08:47 +0000 (12:08 -0600)]
Add option for io_limit
'size' denotes both the size of the region for IO, as well as the
amount of IO to transfer. Sometimes it's useful to be able to set
them separately instead. With this patch, you can do:
size=100G
io_limit=10G
and have fio do 10G of IO between 0..100G of the address space.
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Wed, 23 Apr 2014 15:59:05 +0000 (09:59 -0600)]
Merge branch 'master' of ssh://git.kernel.dk/data/git/fio
Jens Axboe [Wed, 23 Apr 2014 15:58:39 +0000 (09:58 -0600)]
init: don't set did_arg just for an engine option
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Wed, 16 Apr 2014 21:51:27 +0000 (23:51 +0200)]
Error out if the set blocksize is larger than the file size
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Tue, 15 Apr 2014 15:07:44 +0000 (09:07 -0600)]
sg/binject: only restore file flags if they have been set
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Tue, 15 Apr 2014 14:30:06 +0000 (08:30 -0600)]
sg: check for error in fcntl() restore of flags
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Tue, 15 Apr 2014 14:28:51 +0000 (08:28 -0600)]
binject: check for error in fcntl() restore of flags
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Tue, 15 Apr 2014 14:26:57 +0000 (08:26 -0600)]
backend: fix potential division-by-zero
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Tue, 15 Apr 2014 14:25:16 +0000 (08:25 -0600)]
idletime: fix another missing unlock on error
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Tue, 15 Apr 2014 14:18:36 +0000 (08:18 -0600)]
Merge branch 'master' of ssh://git.kernel.dk/data/git/fio
Jens Axboe [Tue, 15 Apr 2014 14:18:08 +0000 (08:18 -0600)]
strcpy: kill last of the suspect ones
Signed-off-by: Jens Axboe <axboe@fb.com>
Bruce Cran [Tue, 15 Apr 2014 00:33:22 +0000 (18:33 -0600)]
Windows: set sizeof(sun_path) to 260 (MAX_PATH).
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 19:37:46 +0000 (13:37 -0600)]
init: improve command line check for when to print usage/help
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 19:29:52 +0000 (13:29 -0600)]
init: fix leak bug fix from commit
60c5ec73
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 19:25:12 +0000 (13:25 -0600)]
disk util: fix bug in strncpy() conversion
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 19:20:38 +0000 (13:20 -0600)]
Fix potential NULL dereference in alloc_mem_mmap()
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 19:19:13 +0000 (13:19 -0600)]
diskutil: make it clear that parent != NULL for json output
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 19:16:10 +0000 (13:16 -0600)]
parse: make it clear that 'cp' can't be NULL for o->off1
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 19:13:21 +0000 (13:13 -0600)]
Fix is_power_of_2() for longs
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 19:11:19 +0000 (13:11 -0600)]
idletime: unlock ->start_lock in error cases
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 19:09:04 +0000 (13:09 -0600)]
Avoid overwriting allocated pid_file
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 18:20:04 +0000 (12:20 -0600)]
gettime: init 'failed' before use
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 18:15:35 +0000 (12:15 -0600)]
stat: warn on failure to detach DU thread
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 18:14:17 +0000 (12:14 -0600)]
gettime: handle pthread_create() failure
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 18:07:36 +0000 (12:07 -0600)]
init: silence bogus td == NULL warning
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 18:05:22 +0000 (12:05 -0600)]
filesetup: silence bogus warning on close(-1)
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 18:03:22 +0000 (12:03 -0600)]
e4defrag: pass 'error' into td_verror()
Don't use -1 or -Exxxx
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 18:01:45 +0000 (12:01 -0600)]
td_verror() needs errno style error
put_file_log() could pass in -1, we need the actual error instead.
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 17:57:05 +0000 (11:57 -0600)]
net engine: get rid of unused struct sockaddr_in6
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 17:45:34 +0000 (11:45 -0600)]
server: ensure that incoming iolog name is NULL terminated
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 17:43:47 +0000 (11:43 -0600)]
server: sanity check incoming command size (and payload)
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 17:36:01 +0000 (11:36 -0600)]
backend: better NULL termination fix for IO scheduler switching
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 16:21:11 +0000 (10:21 -0600)]
num2str: ensure we never access postfix[] out-of-bounds
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 16:13:46 +0000 (10:13 -0600)]
make_filename: another off-by-one buffer issue
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 16:04:21 +0000 (10:04 -0600)]
Error check fcntl() calls
Add a helper to mark an fd as non-blocking, so we can check
and complain in one place.
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 15:52:50 +0000 (09:52 -0600)]
init: remove dead code
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 15:51:10 +0000 (09:51 -0600)]
backend: fix potential divide-by-zero
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 15:49:38 +0000 (09:49 -0600)]
stat: fix potential divide-by-zero
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 15:47:14 +0000 (09:47 -0600)]
idletime: fix potential divide-by-zero
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 15:45:19 +0000 (09:45 -0600)]
eta: fix potential divide by zero
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 15:07:30 +0000 (09:07 -0600)]
server: fix missing comma in fio_server_ops[] strings
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 15:05:46 +0000 (09:05 -0600)]
gettime: use unsigned loop counter
We're comparing with an unsigned exit condition.
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 15:04:48 +0000 (09:04 -0600)]
Make td_verror() use an unsigned type
Get rid of warnings stating the input can't be negative.
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 14:59:17 +0000 (08:59 -0600)]
Fix missing sign conversion in ignore_error_type()
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 14:57:35 +0000 (08:57 -0600)]
stat: fix wrong check for whether we have a description or not
It's an array, hence never NULL. Check for size instead.
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 14:55:13 +0000 (08:55 -0600)]
server: fix potential buffer overrun in socket setup path
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 14:54:09 +0000 (08:54 -0600)]
server: fix potential buffer overrun in bind string
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 14:50:33 +0000 (08:50 -0600)]
client: fix potential buffer overrun in socket path
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 14:49:04 +0000 (08:49 -0600)]
net engine: fix potential buffer overrun in socket path
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 14:45:13 +0000 (08:45 -0600)]
file: fix potential buffer overrun in get_fs_free_counts()
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 14:43:55 +0000 (08:43 -0600)]
verify: fix potential buffer overrun in dump_buf()
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 14:34:43 +0000 (08:34 -0600)]
init: fix potential buffer overrun in make_filename()
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 14:28:08 +0000 (08:28 -0600)]
server: ensure that dus->name is always NULL terminated
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 14:23:23 +0000 (08:23 -0600)]
stat: check that TEMP environment variable is half-way sane
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Mon, 14 Apr 2014 14:17:30 +0000 (08:17 -0600)]
init: fix more cases of leaking pid_file
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 18:17:53 +0000 (12:17 -0600)]
diskutil: separate out stats from name
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:42:13 +0000 (11:42 -0600)]
init: remove bogus check for NULL parent
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:39:46 +0000 (11:39 -0600)]
Don't leak f_out file desciptor if specified more than once
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:38:38 +0000 (11:38 -0600)]
Don't leak memory if profile option is given twice
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:37:34 +0000 (11:37 -0600)]
server: ensure that server pid file mem is always freed
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:34:17 +0000 (11:34 -0600)]
Ensure that fread() string is always NULL terminated
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:32:49 +0000 (11:32 -0600)]
Fix potentially unterminated string after strncpy() call
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:31:17 +0000 (11:31 -0600)]
Fix leak of directory handle in find_add_disk_slaves() error case
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:29:50 +0000 (11:29 -0600)]
net: fix leak of fd in bind() error path
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:28:54 +0000 (11:28 -0600)]
Fix leak on error path in extend_file()
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:27:34 +0000 (11:27 -0600)]
Fix leak of file descriptor in error path of bc_calc()
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:25:32 +0000 (11:25 -0600)]
Fix leak of string when doing parse dryrun
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:24:00 +0000 (11:24 -0600)]
Fix directory leak in recurse_dir() on error
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Fri, 11 Apr 2014 17:20:29 +0000 (11:20 -0600)]
Fix leak of 'error' if we don't assign it
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Thu, 10 Apr 2014 22:50:39 +0000 (16:50 -0600)]
Fio 2.1.8
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Thu, 10 Apr 2014 22:08:41 +0000 (00:08 +0200)]
blktrace: fix missing line return in dprint()
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Thu, 10 Apr 2014 22:03:32 +0000 (00:03 +0200)]
iolog: check for thread termination in long delays
If we are doing a long delay now, fio won't be able to kill
the thread when you ctrl-c the job. Split the delay into .5s
intervals, and check for signal in between.
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Thu, 10 Apr 2014 14:59:43 +0000 (08:59 -0600)]
Make cache invalidation a soft error
On some platforms, we don't have reliable means of flushing
eg block device caches. Just warn of this fact instead of
erroring out, since the user can't do anything to stop this.
Signed-off-by: Jens Axboe <axboe@fb.com>
Christian Ehrhardt [Thu, 10 Apr 2014 13:09:47 +0000 (15:09 +0200)]
fio: fix s390 timing for older machines/binutils
Older gcc/binutils or one with very low default march will refuse to use stckf.
On top of that machines older than a z196 don not even have the store clock
fast facility.
To fix all of those issues, check for the facilities in configure step and make
the usage of stckf and a matching march dependent to it.
Signed-off-by: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Wed, 9 Apr 2014 19:58:43 +0000 (13:58 -0600)]
Merge branch 'master' of ssh://git.kernel.dk/data/git/fio
Jens Axboe [Wed, 9 Apr 2014 19:57:38 +0000 (13:57 -0600)]
Add exit_on_io_done option to the CPU IO engine
The CPU IO engine is most often used to saturate the system,
while running an IO load on it. As such, it's useful to have
CPU engine threads exit automatically, when IO has completed.
Add exit_on_io_done as a CPU IO engine option for that purpose.
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Wed, 9 Apr 2014 03:18:37 +0000 (21:18 -0600)]
iolog: skip max_bs[rw] increment for rw == DDIR_INVAL
Signed-off-by: Jens Axboe <axboe@fb.com>
Jens Axboe [Wed, 9 Apr 2014 03:15:09 +0000 (21:15 -0600)]
diskutil: fix potential out-of-bounds array write
Signed-off-by: Jens Axboe <axboe@fb.com>