rbd_obj_request_wait() should cancel the underlying OSD request if
interrupted. Otherwise libceph will hold onto it indefinitely, causing
assert failures or leaking the original object request.
This also adds an rbd wrapper around ceph_osdc_cancel_request() to
match rbd_obj_request_submit() and rbd_obj_request_wait().
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Alexander reported mkswap on /dev/zram0 is failed if other process is
opening the block device file.
Step is as follows,
0. Reset the unused zram device.
1. Use a program that opens /dev/zram0 with O_RDWR and sleeps
until killed.
2. While that program sleeps, echo the correct value to
/sys/block/zram0/disksize.
3. Verify (e.g. in /proc/partitions) that the disk size is applied
correctly. It is.
4. While that program still sleeps, attempt to mkswap /dev/zram0.
This fails: mkswap: error: swap area needs to be at least 40 KiB
When I investigated, the size get by ioctl(fd, BLKGETSIZE64, xxx) on
mkswap to get a size of blockdev was zero although zram0 has right size by
2.
The reason is zram didn't revalidate disk after changing capacity so that
size of blockdev's inode is not uptodate until all of file is close.
This patch should fix the BUG.
Signed-off-by: Minchan Kim <minchan@kernel.org>
Reported-by: Alexander E. Patrakov <patrakov@gmail.com>
Tested-by: Alexander E. Patrakov <patrakov@gmail.com>
Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Acked-by: Jerome Marchand <jmarchan@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Firstly this patch supports more than one virtual queues for virtio-blk
device.
Secondly this patch maps the virtual queue to blk-mq's hardware queue.
With this approach, both scalability and performance can be improved.
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Pull block fixes from Jens Axboe:
"A small collection of fixes/changes for the current series. This
contains:
- Removal of dead code from Gu Zheng.
- Revert of two bad fixes that went in earlier in this round, marking
things as __init that were not purely used from init.
- A fix for blk_mq_start_hw_queue() using the __blk_mq_run_hw_queue(),
which could place us wrongly. Make it use the non __ variant,
which handles cases where we are called from the wrong CPU set.
From me.
- A fix for drbd, which allocates discard requests without room for
the SCSI payload. From Lars Ellenberg.
- A fix for user-after-free in the blkcg code from Tejun.
- Addition of limiting gaps in SG lists, if the hardware needs it.
This is the last pre-req patch for blk-mq to enable the full NVMe
conversion. Could wait until 3.17, but it's simple enough so would
be nice to have everything we need for the NVMe port in the 3.17
release. From me"
* 'for-linus' of git://git.kernel.dk/linux-block:
drbd: fix NULL pointer deref in blk_add_request_payload
blk-mq: blk_mq_start_hw_queue() should use blk_mq_run_hw_queue()
block: add support for limiting gaps in SG lists
bio: remove unused macro bip_vec_idx()
Revert "block: add __init to elv_register"
Revert "block: add __init to blkcg_policy_register"
blkcg: fix use-after-free in __blkg_release_rcu() by making blkcg_gq refcnt an atomic_t
floppy: format block0 read error message properly
Discards don't have any payload.
But the scsi layer still expects a bio_vec it can use internally,
see sd_setup_discard_cmnd() and blk_add_request_payload().
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
The following check in rbd_img_obj_request_submit()
rbd_dev->parent_overlap <= obj_request->img_offset
allows the fall through to the non-layered write case even if both
parent_overlap and obj_request->img_offset belong to the same RADOS
object. This leads to data corruption, because the area to the left of
parent_overlap ends up unconditionally zero-filled instead of being
populated with parent data. Suppose we want to write 1M to offset 6M
of image bar, which is a clone of foo@snap; object_size is 4M,
parent_overlap is 5M:
rbd_data.<id>.0000000000000001
---------------------|----------------------|------------
| should be copyup'ed | should be zeroed out | write ...
---------------------|----------------------|------------
4M 5M 6M
parent_overlap obj_request->img_offset
4..5M should be copyup'ed from foo, yet it is zero-filled, just like
5..6M is.
Given that the only striping mode kernel client currently supports is
chunking (i.e. stripe_unit == object_size, stripe_count == 1), round
parent_overlap up to the next object boundary for the purposes of the
overlap check.
Cc: stable@vger.kernel.org # 3.10+
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Pull block fixes from Jens Axboe:
"A smaller collection of fixes for the block core that would be nice to
have in -rc2. This pull request contains:
- Fixes for races in the wait/wakeup logic used in blk-mq from
Alexander. No issues have been observed, but it is definitely a
bit flakey currently. Alternatively, we may drop the cyclic
wakeups going forward, but that needs more testing.
- Some cleanups from Christoph.
- Fix for an oops in null_blk if queue_mode=1 and softirq completions
are used. From me.
- A fix for a regression caused by the chunk size setting. It
inadvertently used max_hw_sectors instead of max_sectors, which is
incorrect, and causes hangs on btrfs multi-disk setups (where hw
sectors apparently isn't set). From me.
- Removal of WQ_POWER_EFFICIENT in the kblockd creation. This was a
recent addition as well, but it actually breaks blk-mq which relies
on strict scheduling. If the workqueue power_efficient mode is
turned on, this breaks blk-mq. From Matias.
- null_blk module parameter description fix from Mike"
* 'for-linus' of git://git.kernel.dk/linux-block:
blk-mq: bitmap tag: fix races in bt_get() function
blk-mq: bitmap tag: fix race on blk_mq_bitmap_tags::wake_cnt
blk-mq: bitmap tag: fix races on shared ::wake_index fields
block: blk_max_size_offset() should check ->max_sectors
null_blk: fix softirq completions for queue_mode == 1
blk-mq: merge blk_mq_drain_queue and __blk_mq_drain_queue
blk-mq: properly drain stopped queues
block: remove WQ_POWER_EFFICIENT from kblockd
null_blk: fix name and description of 'queue_mode' module parameter
block: remove elv_abort_queue and blk_abort_flushes
In case reading of block 0 fails, line without trailing newline
is printed causing dmesg to look horrible.
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Only blk-mq completions have payload attached to the request, for
request_fn mode we have stored it in req->special. This fixes an
oops with queue_mode=1 and softirq completions.
Signed-off-by: Jens Axboe <axboe@fb.com>
Pull NVMe update from Matthew Wilcox:
"Mostly bugfixes again for the NVMe driver. I'd like to call out the
exported tracepoint in the block layer; I believe Keith has cleared
this with Jens.
We've had a few reports from people who're really pounding on NVMe
devices at scale, hence the timeout changes (and new module
parameters), hotplug cpu deadlock, tracepoints, and minor performance
tweaks"
[ Jens hadn't seen that tracepoint thing, but is ok with it - it will
end up going away when mq conversion happens ]
* git://git.infradead.org/users/willy/linux-nvme: (22 commits)
NVMe: Fix START_STOP_UNIT Scsi->NVMe translation.
NVMe: Use Log Page constants in SCSI emulation
NVMe: Define Log Page constants
NVMe: Fix hot cpu notification dead lock
NVMe: Rename io_timeout to nvme_io_timeout
NVMe: Use last bytes of f/w rev SCSI Inquiry
NVMe: Adhere to request queue block accounting enable/disable
NVMe: Fix nvme get/put queue semantics
NVMe: Delete NVME_GET_FEAT_TEMP_THRESH
NVMe: Make admin timeout a module parameter
NVMe: Make iod bio timeout a parameter
NVMe: Prevent possible NULL pointer dereference
NVMe: Fix the buffer size passed in GetLogPage(CDW10.NUMD)
NVMe: Update data structures for NVMe 1.2
NVMe: Enable BUILD_BUG_ON checks
NVMe: Update namespace and controller identify structures to the 1.1a spec
NVMe: Flush with data support
NVMe: Configure support for block flush
NVMe: Add tracepoints
NVMe: Protect against badly formatted CQEs
...
This patch contains several fixes for Scsi START_STOP_UNIT. The previous
code did not account for signed vs. unsigned arithmetic which resulted
in an invalid lowest power state caculation when the device only supports
1 power state.
The code for Power Condition == 2 (Idle) was not following the spec. The
spec calls for setting the device to specific power states, depending
upon Power Condition Modifier, without accounting for the number of
power states supported by the device.
The code for Power Condition == 3 (Standby) was using a hard-coded '0'
which is replaced with the macro POWER_STATE_0.
Signed-off-by: Dan McLeran <daniel.mcleran@intel.com>
Reviewed-by: Vishal Verma <vishal.l.verma@linux.intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
The nvme-scsi file defined its own Log Page constant. Use the
newly-defined one from the header file instead.
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
There is a potential dead lock if a cpu event occurs during nvme probe
since it registered with hot cpu notification. This fixes the race by
having the module register with notification outside of probe rather
than have each device register.
The actual work is done in a scheduled work queue instead of in the
notifier since assigning IO queues has the potential to block if the
driver creates additional queues.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Pull Ceph updates from Sage Weil:
"This has a mix of bug fixes and cleanups.
Alex's patch fixes a rare race in RBD. Ilya's patches fix an ENOENT
check when a second rbd image is mapped and a couple memory leaks.
Zheng fixes several issues with fragmented directories and multiple
MDSs. Josh fixes a spin/sleep issue, and Josh and Guangliang's
patches fix setting and unsetting RBD images read-only.
Naturally there are several other cleanups mixed in for good measure"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client: (23 commits)
rbd: only set disk to read-only once
rbd: move calls that may sleep out of spin lock range
rbd: add ioctl for rbd
ceph: use truncate_pagecache() instead of truncate_inode_pages()
ceph: include time stamp in every MDS request
rbd: fix ida/idr memory leak
rbd: use reference counts for image requests
rbd: fix osd_request memory leak in __rbd_dev_header_watch_sync()
rbd: make sure we have latest osdmap on 'rbd map'
libceph: add ceph_monc_wait_osdmap()
libceph: mon_get_version request infrastructure
libceph: recognize poolop requests in debugfs
ceph: refactor readpage_nounlock() to make the logic clearer
mds: check cap ID when handling cap export message
ceph: remember subtree root dirfrag's auth MDS
ceph: introduce ceph_fill_fragtree()
ceph: handle cap import atomically
ceph: pre-allocate ceph_cap struct for ceph_add_cap()
ceph: update inode fields according to issued caps
rbd: replace IS_ERR and PTR_ERR with PTR_ERR_OR_ZERO
...
'use_mq' is not the name of the module parameter, 'queue_mode' is.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Pull block layer fixes from Jens Axboe:
"Final small batch of fixes to be included before -rc1. Some general
cleanups in here as well, but some of the blk-mq fixes we need for the
NVMe conversion and/or scsi-mq. The pull request contains:
- Support for not merging across a specified "chunk size", if set by
the driver. Some NVMe devices perform poorly for IO that crosses
such a chunk, so we need to support it generically as part of
request merging avoid having to do complicated split logic. From
me.
- Bump max tag depth to 10Ki tags. Some scsi devices have a huge
shared tag space. Before we failed with EINVAL if a too large tag
depth was specified, now we truncate it and pass back the actual
value. From me.
- Various blk-mq rq init fixes from me and others.
- A fix for enter on a dying queue for blk-mq from Keith. This is
needed to prevent oopsing on hot device removal.
- Fixup for blk-mq timer addition from Ming Lei.
- Small round of performance fixes for mtip32xx from Sam Bradshaw.
- Minor stack leak fix from Rickard Strandqvist.
- Two __init annotations from Fabian Frederick"
* 'for-linus' of git://git.kernel.dk/linux-block:
block: add __init to blkcg_policy_register
block: add __init to elv_register
block: ensure that bio_add_page() always accepts a page for an empty bio
blk-mq: add timer in blk_mq_start_request
blk-mq: always initialize request->start_time
block: blk-exec.c: Cleaning up local variable address returnd
mtip32xx: minor performance enhancements
blk-mq: ->timeout should be cleared in blk_mq_rq_ctx_init()
blk-mq: don't allow queue entering for a dying queue
blk-mq: bump max tag depth to 10K tags
block: add blk_rq_set_block_pc()
block: add notion of a chunk size for request merging
rbd_open(), called every time the device is opened, calls
set_device_ro(). There's no reason to set the device read-only or
read-write every time it is opened. Just do this once during device
setup, using set_disk_ro() instead because the struct block_device
isn't available to us there.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
Reviewed-by: Alex Elder <elder@linaro.org>
get_user() and set_disk_ro() may allocate memory, leading to a
potential deadlock if theye are called while a spin lock is held.
Move the acquisition and release of rbd_dev->lock from rbd_ioctl()
into rbd_ioctl_set_ro(), so it can occur between get_user() and
set_disk_ro().
Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
Reviewed-by: Alex Elder <elder@linaro.org>
When running the following commands:
[root@ceph0 mnt]# blockdev --setro /dev/rbd1
[root@ceph0 mnt]# blockdev --getro /dev/rbd1
0
The block setro didn't take effect, it is because
the rbd doesn't support ioctl of block driver.
This resolves:
http://tracker.ceph.com/issues/6265
Signed-off-by: Guangliang Zhao <guangliang@unitedstack.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Len field is already set to zero, but not the from field which is sent
as 0xfffffffffffffe00. This makes no sense, and may cause confuse
server implementations doing sanity checks (qemu-nbd is an example.)
Signed-off-by: Hani Benhabiles <hani@linux.com>
Cc: Paul Clements <paul.clements@us.sios.com>
Cc: Paul Clements <Paul.Clements@steeleye.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch adds the following:
1) Compiler hinting in the fast path.
2) A prefetch of port->flags to eliminate moderate cpu stalling later
in mtip_hw_submit_io().
3) Eliminate a redundant rq_data_dir().
4) Reorder members of driver_data to eliminate false cacheline sharing
between irq_workers_active and unal_qdepth.
With some workload and topology configurations, I'm seeing ~1.5%
throughput improvement in small block random read benchmarks as well
as improved latency std. dev.
Signed-off-by: Sam Bradshaw <sbradshaw@micron.com>
Add include of <linux/prefetch.h>
Signed-off-by: Jens Axboe <axboe@fb.com>
With the optimizations around not clearing the full request at alloc
time, we are leaving some of the needed init for REQ_TYPE_BLOCK_PC
up to the user allocating the request.
Add a blk_rq_set_block_pc() that sets the command type to
REQ_TYPE_BLOCK_PC, and properly initializes the members associated
with this type of request. Update callers to use this function instead
of manipulating rq->cmd_type directly.
Includes fixes from Christoph Hellwig <hch@lst.de> for my half-assed
attempt.
Signed-off-by: Jens Axboe <axboe@fb.com>
ida_destroy() needs to be called on module exit to release ida caches.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Each image request contains a reference count, but to date it has
not actually been used. (I think this was just an oversight.) A
recent report involving rbd failing an assertion shed light on why
and where we need to use these reference counts.
Every OSD request associated with an object request uses
rbd_osd_req_callback() as its callback function. That function will
call a helper function (dependent on the type of OSD request) that
will set the object request's "done" flag if the object request if
appropriate. If that "done" flag is set, the object request is
passed to rbd_obj_request_complete().
In rbd_obj_request_complete(), requests are processed in sequential
order. So if an object request completes before one of its
predecessors in the image request, the completion is deferred.
Otherwise, if it's a completing object's "turn" to be completed, it
is passed to rbd_img_obj_end_request(), which records the result of
the operation, accumulates transferred bytes, and so on. Next, the
successor to this request is checked and if it is marked "done",
(deferred) completion processing is performed on that request, and
so on. If the last object request in an image request is completed,
rbd_img_request_complete() is called, which (typically) destroys
the image request.
There is a race here, however. The instant an object request is
marked "done" it can be provided (by a thread handling completion of
one of its predecessor operations) to rbd_img_obj_end_request(),
which (for the last request) can then lead to the image request
getting torn down. And this can happen *before* that object has
itself entered rbd_img_obj_end_request(). As a result, once it
*does* enter that function, the image request (and even the object
request itself) may have been freed and become invalid.
All that's necessary to avoid this is to properly count references
to the image requests. We tear down an image request's object
requests all at once--only when the entire image request has
completed. So there's no need for an image request to count
references for its object requests. However, we don't want an
image request to go away until the last of its object requests
has passed through rbd_img_obj_callback(). In other words,
we don't want rbd_img_request_complete() to necessarily
result in the image request being destroyed, because it may
get called before we've finished processing on all of its
object requests.
So the fix is to add a reference to an image request for
each of its object requests. The reference can be viewed
as representing an object request that has not yet finished
its call to rbd_img_obj_callback(). That is emphasized by
getting the reference right after assigning that as the image
object's callback function. The corresponding release of that
reference is done at the end of rbd_img_obj_callback(), which
every image object request passes through exactly once.
Cc: stable@vger.kernel.org
Signed-off-by: Alex Elder <elder@linaro.org>
Reviewed-by: Ilya Dryomov <ilya.dryomov@inktank.com>
osd_request, along with r_request and r_reply messages attached to it
are leaked in __rbd_dev_header_watch_sync() if the requested image
doesn't exist. This is because lingering requests are special and get
an extra ref in the reply path. Fix it by unregistering linger request
on the error path and split __rbd_dev_header_watch_sync() into two
functions to make it maintainable.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Given an existing idle mapping (img1), mapping an image (img2) in
a newly created pool (pool2) fails:
$ ceph osd pool create pool1 8 8
$ rbd create --size 1000 pool1/img1
$ sudo rbd map pool1/img1
$ ceph osd pool create pool2 8 8
$ rbd create --size 1000 pool2/img2
$ sudo rbd map pool2/img2
rbd: sysfs write failed
rbd: map failed: (2) No such file or directory
This is because client instances are shared by default and we don't
request an osdmap update when bumping a ref on an existing client. The
fix is to use the mon_get_version request to see if the osdmap we have
is the latest, and block until the requested update is received if it's
not.
Fixes: http://tracker.ceph.com/issues/8184
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Merge misc updates from Andrew Morton:
- a few fixes for 3.16. Cc'ed to stable so they'll get there somehow.
- various misc fixes and cleanups
- most of the ocfs2 queue. Review is slow...
- most of MM. The MM queue is pretty huge this time, but not much in
the way of feature work.
- some tweaks under kernel/
- printk maintenance work
- updates to lib/
- checkpatch updates
- tweaks to init/
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (276 commits)
fs/autofs4/dev-ioctl.c: add __init to autofs_dev_ioctl_init
fs/ncpfs/getopt.c: replace simple_strtoul by kstrtoul
init/main.c: remove an ifdef
kthreads: kill CLONE_KERNEL, change kernel_thread(kernel_init) to avoid CLONE_SIGHAND
init/main.c: add initcall_blacklist kernel parameter
init/main.c: don't use pr_debug()
fs/binfmt_flat.c: make old_reloc() static
fs/binfmt_elf.c: fix bool assignements
fs/efs: convert printk(KERN_DEBUG to pr_debug
fs/efs: add pr_fmt / use __func__
fs/efs: convert printk to pr_foo()
scripts/checkpatch.pl: device_initcall is not the only __initcall substitute
checkpatch: check stable email address
checkpatch: warn on unnecessary void function return statements
checkpatch: prefer kstrto<foo> to sscanf(buf, "%<lhuidx>", &bar);
checkpatch: add warning for kmalloc/kzalloc with multiply
checkpatch: warn on #defines ending in semicolon
checkpatch: make --strict a default for files in drivers/net and net/
checkpatch: always warn on missing blank line after variable declaration block
checkpatch: fix wildcard DT compatible string checking
...
We want to skip the physical block(PAGE_SIZE) which is partially covered
by the discard bio, so we check the remaining size and subtract it if
there is a need to goto the next physical block.
The current offset usage in zram_bio_discard is incorrect, it will cause
its upper filesystem breakdown. Consider the following scenario:
On some architecture or config, PAGE_SIZE is 64K for example, filesystem
is set up on zram disk without PAGE_SIZE aligned, a discard bio leads to a
offset = 4K and size=72K, normally, it should not really discard any
physical block as it partially cover two physical blocks. However, with
the current offset usage, it will discard the second physical block and
free its memory, which will cause filesystem breakdown.
This patch corrects the offset usage in zram_bio_discard.
Signed-off-by: Weijie Yang <weijie.yang@samsung.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Acked-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Bob Liu <bob.liu@oracle.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
brd is effectively a thinly provisioned device. Thinly provisioned
devices return -ENOSPC when they can't write a new block. -ENOMEM is an
implementation detail that callers shouldn't know.
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Acked-by: Dave Chinner <david@fromorbit.com>
Cc: Dheeraj Reddy <dheeraj.reddy@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We currently pass in the hardware queue, and get the tags from there.
But from scsi-mq, with a shared tag space, it's a lot more convenient
to pass in the blk_mq_tags instead as the hardware queue isn't always
directly available. So instead of having to re-map to a given
hardware queue from rq->mq_ctx, just pass in the tags structure.
Signed-off-by: Jens Axboe <axboe@fb.com>
It's positively immoral to have a global variable called 'io_timeout'.
Keep the module parameter called io_timeout, though.
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
After skipping right-padded spaces, use the last four bytes of the
firmware revision when reporting the Inquiry Product Revision. These
are generally more indicative to what is running.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Acked-by: Vishal Verma <vishal.l.verma@linux.intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Recently, a new sysfs control "iostats" was added to selectively
enable or disable io statistics collection for request queues. This
patch hooks that control.
IO statistics collection is rather expensive on large, multi-node
machines with drives pushing millions of iops. Having the ability to
disable collection if not needed can improve throughput significantly.
As a data point, on a quad E5-4640, I see more than 50% throughput
improvement when io statistics accounting is disabled during heavily
multi-threaded small block random read benchmarks where device
performance is in the million iops+ range.
Signed-off-by: Sam Bradshaw <sbradshaw@micron.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
The routines to get and lock nvme queues required the caller to "put"
or "unlock" them even if getting one returned NULL. This patch fixes that.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
This define isn't used, and any code that wanted to use it should use
NVME_FEAT_TEMP_THRESH instead.
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
This was originally set to 4 times the IO timeout, but that was when
the IO timeout was 5 seconds instead of 30. 20 seconds for total time
to failure seemed more reasonable than 2 minutes for most, but other
users have requested to make this a module parameter instead.
Signed-off-by: Keith Busch <keith.busch@intel.com>
[renamed the module parameter to retry_time]
[made retry_time static]
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Pull scheduler updates from Ingo Molnar:
"The main scheduling related changes in this cycle were:
- various sched/numa updates, for better performance
- tree wide cleanup of open coded nice levels
- nohz fix related to rq->nr_running use
- cpuidle changes and continued consolidation to improve the
kernel/sched/idle.c high level idle scheduling logic. As part of
this effort I pulled cpuidle driver changes from Rafael as well.
- standardized idle polling amongst architectures
- continued work on preparing better power/energy aware scheduling
- sched/rt updates
- misc fixlets and cleanups"
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (49 commits)
sched/numa: Decay ->wakee_flips instead of zeroing
sched/numa: Update migrate_improves/degrades_locality()
sched/numa: Allow task switch if load imbalance improves
sched/rt: Fix 'struct sched_dl_entity' and dl_task_time() comments, to match the current upstream code
sched: Consolidate open coded implementations of nice level frobbing into nice_to_rlimit() and rlimit_to_nice()
sched: Initialize rq->age_stamp on processor start
sched, nohz: Change rq->nr_running to always use wrappers
sched: Fix the rq->next_balance logic in rebalance_domains() and idle_balance()
sched: Use clamp() and clamp_val() to make sys_nice() more readable
sched: Do not zero sg->cpumask and sg->sgp->power in build_sched_groups()
sched/numa: Fix initialization of sched_domain_topology for NUMA
sched: Call select_idle_sibling() when not affine_sd
sched: Simplify return logic in sched_read_attr()
sched: Simplify return logic in sched_copy_attr()
sched: Fix exec_start/task_hot on migrated tasks
arm64: Remove TIF_POLLING_NRFLAG
metag: Remove TIF_POLLING_NRFLAG
sched/idle: Make cpuidle_idle_call() void
sched/idle: Reflow cpuidle_idle_call()
sched/idle: Delay clearing the polling bit
...
kmalloc() used by the nvme_alloc_iod() to allocate memory for 'iod'
can fail. So check the return value.
Signed-off-by: Santosh Y <santosh.sy@samsung.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
In GetLogPage the buffer size passed to device is a 0's based value.
Signed-off-by: Indraneel M <indraneel.m@samsung.com>
Reported-by: Shiro Itou <shiro.itou@outlook.com>
Reviewed-by: Vishal Verma <vishal.l.verma@linux.intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Pull block driver changes from Jens Axboe:
"Now that the core bits are in, here's the pull request for the driver
related changes for 3.16. Nothing out of the ordinary here, mostly
business as usual. There are a few pulls of for-3.16/core into this
branch, which were done when the blk-mq was modified after the
mtip32xx conversion was put in.
The pull request contains:
- skd and cciss converted to use pci_enable_msix_exact(). From
Alexander Gordeev.
- A few mtip32xx fixes from Asai @ Micron.
- The conversion of mtip32xx from make_request_fn to blk-mq, and a
later small fix for that conversion on quiescing for non-queued IO.
From me.
- A fix for bsg to use an exported function to check whether this
driver is request based or not. Needed updating for blk-mq, which
is request based, but does not have a request_fn hook. From me.
- Small floppy bug fix from Jiri.
- A series of cleanups for the cdrom uniform layer from Joe Perches.
Gets rid of various old ugly macros, making the code conform more
to the modern coding style.
- A series of patches for drbd from the drbd crew (Lars Ellenberg and
Philipp Reisner).
- A use-after-free fix for null_blk from Ming Lei.
- Also from Ming Lei is a performance patch for virtio-blk, which can
net us a 3x win on kvm platforms where world notification is
expensive.
- Ming Lei also fixed a stall issue in virtio-blk, due to a race
between queue start/stop and resource limits.
- A small batch of fixes for xen-blk{back,front} from Olaf Hering and
Valentin Priescu"
* 'for-3.16/drivers' of git://git.kernel.dk/linux-block: (54 commits)
block: virtio_blk: don't hold spin lock during world switch
xen-blkback: defer freeing blkif to avoid blocking xenwatch
xen blkif.h: fix comment typo in discard-alignment
xen/blkback: disable discard feature if requested by toolstack
xen-blkfront: remove type check from blkfront_setup_discard
floppy: do not corrupt bio.bi_flags when reading block 0
mtip32xx: move error handling to service thread
virtio_blk: fix race between start and stop queue
mtip32xx: stop block hardware queues before quiescing IO
mtip32xx: blk_mq_init_queue() returns an ERR_PTR
mtip32xx: convert to use blk-mq
cdrom: Remove unnecessary prototype for cdrom_get_disc_info
cdrom: Remove unnecessary prototype for cdrom_mrw_exit
cdrom: Remove cdrom_count_tracks prototype
cdrom: Remove cdrom_get_next_writeable prototype
cdrom: Remove cdrom_get_last_written prototype
cdrom: Move mmc_ioctls above cdrom_ioctl to remove unnecessary prototype
cdrom: Remove unnecessary sanitize_format prototype
cdrom: Remove unnecessary check_for_audio_disc prototype
cdrom: Remove prototype for open_for_data
...
Pull block core updates from Jens Axboe:
"It's a big(ish) round this time, lots of development effort has gone
into blk-mq in the last 3 months. Generally we're heading to where
3.16 will be a feature complete and performant blk-mq. scsi-mq is
progressing nicely and will hopefully be in 3.17. A nvme port is in
progress, and the Micron pci-e flash driver, mtip32xx, is converted
and will be sent in with the driver pull request for 3.16.
This pull request contains:
- Lots of prep and support patches for scsi-mq have been integrated.
All from Christoph.
- API and code cleanups for blk-mq from Christoph.
- Lots of good corner case and error handling cleanup fixes for
blk-mq from Ming Lei.
- A flew of blk-mq updates from me:
* Provide strict mappings so that the driver can rely on the CPU
to queue mapping. This enables optimizations in the driver.
* Provided a bitmap tagging instead of percpu_ida, which never
really worked well for blk-mq. percpu_ida relies on the fact
that we have a lot more tags available than we really need, it
fails miserably for cases where we exhaust (or are close to
exhausting) the tag space.
* Provide sane support for shared tag maps, as utilized by scsi-mq
* Various fixes for IO timeouts.
* API cleanups, and lots of perf tweaks and optimizations.
- Remove 'buffer' from struct request. This is ancient code, from
when requests were always virtually mapped. Kill it, to reclaim
some space in struct request. From me.
- Remove 'magic' from blk_plug. Since we store these on the stack
and since we've never caught any actual bugs with this, lets just
get rid of it. From me.
- Only call part_in_flight() once for IO completion, as includes two
atomic reads. Hopefully we'll get a better implementation soon, as
the part IO stats are now one of the more expensive parts of doing
IO on blk-mq. From me.
- File migration of block code from {mm,fs}/ to block/. This
includes bio.c, bio-integrity.c, bounce.c, and ioprio.c. From me,
from a discussion on lkml.
That should describe the meat of the pull request. Also has various
little fixes and cleanups from Dave Jones, Shaohua Li, Duan Jiong,
Fengguang Wu, Fabian Frederick, Randy Dunlap, Robert Elliott, and Sam
Bradshaw"
* 'for-3.16/core' of git://git.kernel.dk/linux-block: (100 commits)
blk-mq: push IPI or local end_io decision to __blk_mq_complete_request()
blk-mq: remember to start timeout handler for direct queue
block: ensure that the timer is always added
blk-mq: blk_mq_unregister_hctx() can be static
blk-mq: make the sysfs mq/ layout reflect current mappings
blk-mq: blk_mq_tag_to_rq should handle flush request
block: remove dead code in scsi_ioctl:blk_verify_command
blk-mq: request initialization optimizations
block: add queue flag for disabling SG merging
block: remove 'magic' from struct blk_plug
blk-mq: remove alloc_hctx and free_hctx methods
blk-mq: add file comments and update copyright notices
blk-mq: remove blk_mq_alloc_request_pinned
blk-mq: do not use blk_mq_alloc_request_pinned in blk_mq_map_request
blk-mq: remove blk_mq_wait_for_tags
blk-mq: initialize request in __blk_mq_alloc_request
blk-mq: merge blk_mq_alloc_reserved_request into blk_mq_alloc_request
blk-mq: add helper to insert requests from irq context
blk-mq: remove stale comment for blk_mq_complete_request()
blk-mq: allow non-softirq completions
...
Pull m68k updates from Geert Uytterhoeven:
"Highlights:
- support for running kernels in fast TT-RAM instead of slow ST-RAM
on Atari
- multi-platform EARLY_PRINTK
- better support for machines with lots of RAM (think ARAnyM), and
for running kernels larger than 4 MiB (think multi-platform)"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k:
m68k/hp300: Convert printk to pr_foo()
m68k/apollo: Convert printk to pr_foo()
m68k/amiga: Convert printk(foo to pr_foo()
m68k: Increase initial mapping to 8 or 16 MiB if possible
m68k: Update defconfigs for v3.15-rc2
m68k/atari: fix SCC initialization for debug console
m68k/mvme16x: Adopt common boot console
m68k: Multi-platform EARLY_PRINTK
m68k: Toward platform agnostic framebuffer debug logging
m68k/atari - atari_scsi: use correct virt/phys translation for DMA buffer
m68k/atari - ataflop: use correct virt/phys translation for DMA buffer
m68k/atari - atafb: convert allocation of fb ram to new interface
m68k/atari - stram: alloc ST-RAM pool even if kernel not in ST-RAM
Firstly, it isn't necessary to hold lock of vblk->vq_lock
when notifying hypervisor about queued I/O.
Secondly, virtqueue_notify() will cause world switch and
it may take long time on some hypervisors(such as, qemu-arm),
so it isn't good to hold the lock and block other vCPUs.
On arm64 quad core VM(qemu-kvm), the patch can increase I/O
performance a lot with VIRTIO_RING_F_EVENT_IDX enabled:
- without the patch: 14K IOPS
- with the patch: 34K IOPS
fio script:
[global]
direct=1
bsrange=4k-4k
timeout=10
numjobs=4
ioengine=libaio
iodepth=64
filename=/dev/vdc
group_reporting=1
[f1]
rw=randread
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: virtualization@lists.linux-foundation.org
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: stable@kernel.org # 3.13+
Signed-off-by: Jens Axboe <axboe@fb.com>
Konrad writes:
Please git pull the following branch:
git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip.git stable/for-jens-3.16
which has a bunch of fixes to the Xen block frontend and backend driver
and a new parameter for Xen backend driver - an override (set by the toolstack)
whether to expose the discard support (if disk of course supports it) or not.
Currently xenwatch blocks in VBD disconnect, waiting for all pending I/O
requests to finish. If the VBD is attached to a hot-swappable disk, then
xenwatch can hang for a long period of time, stalling other watches.
INFO: task xenwatch:39 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
ffff880057f01bd0 0000000000000246 ffff880057f01ac0 ffffffff810b0782
ffff880057f01ad0 00000000000131c0 0000000000000004 ffff880057edb040
ffff8800344c6080 0000000000000000 ffff880058c00ba0 ffff880057edb040
Call Trace:
[<ffffffff810b0782>] ? irq_to_desc+0x12/0x20
[<ffffffff8128f761>] ? list_del+0x11/0x40
[<ffffffff8147a080>] ? wait_for_common+0x60/0x160
[<ffffffff8147bcef>] ? _raw_spin_lock_irqsave+0x2f/0x50
[<ffffffff8147bd49>] ? _raw_spin_unlock_irqrestore+0x19/0x20
[<ffffffff8147a26a>] schedule+0x3a/0x60
[<ffffffffa018fe6a>] xen_blkif_disconnect+0x8a/0x100 [xen_blkback]
[<ffffffff81079f70>] ? wake_up_bit+0x40/0x40
[<ffffffffa018ffce>] xen_blkbk_remove+0xae/0x1e0 [xen_blkback]
[<ffffffff8130b254>] xenbus_dev_remove+0x44/0x90
[<ffffffff81345cb7>] __device_release_driver+0x77/0xd0
[<ffffffff81346488>] device_release_driver+0x28/0x40
[<ffffffff813456e8>] bus_remove_device+0x78/0xe0
[<ffffffff81342c9f>] device_del+0x12f/0x1a0
[<ffffffff81342d2d>] device_unregister+0x1d/0x60
[<ffffffffa0190826>] frontend_changed+0xa6/0x4d0 [xen_blkback]
[<ffffffffa019c252>] ? frontend_changed+0x192/0x650 [xen_netback]
[<ffffffff8130ae50>] ? cmp_dev+0x60/0x60
[<ffffffff81344fe4>] ? bus_for_each_dev+0x94/0xa0
[<ffffffff8130b06e>] xenbus_otherend_changed+0xbe/0x120
[<ffffffff8130b4cb>] frontend_changed+0xb/0x10
[<ffffffff81309c82>] xenwatch_thread+0xf2/0x130
[<ffffffff81079f70>] ? wake_up_bit+0x40/0x40
[<ffffffff81309b90>] ? xenbus_directory+0x80/0x80
[<ffffffff810799d6>] kthread+0x96/0xa0
[<ffffffff81485934>] kernel_thread_helper+0x4/0x10
[<ffffffff814839f3>] ? int_ret_from_sys_call+0x7/0x1b
[<ffffffff8147c17c>] ? retint_restore_args+0x5/0x6
[<ffffffff81485930>] ? gs_change+0x13/0x13
With this patch, when there is still pending I/O, the actual disconnect
is done by the last reference holder (last pending I/O request). In this
case, xenwatch doesn't block indefinitely.
Signed-off-by: Valentin Priescu <priescuv@amazon.com>
Reviewed-by: Steven Kady <stevkady@amazon.com>
Reviewed-by: Steven Noonan <snoonan@amazon.com>
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Newer toolstacks may provide a boolean property "discard-enable" in the
backend node. Its purpose is to disable discard for file backed storage
to avoid fragmentation. Recognize this setting also for physical
storage. If that property exists and is false, do not advertise
"feature-discard" to the frontend.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
In its initial implementation a check for "type" was added, but only phy
and file are handled. This breaks advertised discard support for other
type values such as qdisk.
Fix and simplify this function: If the backend advertises discard
support it is supposed to implement it properly, so enable
feature_discard unconditionally. If the backend advertises the need for
a certain granularity and alignment then propagate both properties to
the blocklayer. The discard-secure property is a boolean, update the code
to reflect that.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Pull in core changes (again), since we got rid of the alloc/free
hctx mq_ops hooks and mtip32xx then needed updating again.
Signed-off-by: Jens Axboe <axboe@fb.com>
There is no need for drivers to control hardware context allocation
now that we do the context to node mapping in common code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
mtip32xx uses blk_mq_alloc_reserved_request(), so pull in the
core changes so we have a properly merged end result.
Signed-off-by: Jens Axboe <axboe@fb.com>
Commit 41a55b4de3 ("floppy: silence warning during disk test") caused
bio.bi_flags being overwritten, and its initialization to BIO_UPTODATE
in bio_init() to be lost.
This was unnoticed until 7b7b68bba5 ("floppy: bail out in open() if
drive is not responding to block0 read"), because the error value wasn't
checked for in the bio completion callback.
Now we are actually looking at the error, and the loss of BIO_UPTODATE
causes EIO to be wrongly passed to the callback, which confuses the
FD_OPEN_SHOULD_FAIL_BIT logic.
Fix this by not destroying previous value of bi_flags when setting
BIO_QUIET.
Cc: Stephen Hemminger <shemminger@vyatta.com>
Reported-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Drivers currently have to figure this out on their own, and they
are missing information to do it properly. The ones that did
attempt to do it, do it wrong.
So just pass in the suggested node directly to the alloc
function.
Signed-off-by: Jens Axboe <axboe@fb.com>
Quiesce and shutdown the device prior to reset, then restart the device and
resume IO after.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
When there isn't enough vring descriptor for adding to vq,
blk-mq will be put as stopped state until some of pending
descriptors are completed & freed.
Unfortunately, the vq's interrupt may come just before
blk-mq's BLK_MQ_S_STOPPED flag is set, so the blk-mq will
still be kept as stopped even though lots of descriptors
are completed and freed in the interrupt handler. The worst
case is that all pending descriptors are freed in the
interrupt handler, and the queue is kept as stopped forever.
This patch fixes the problem by starting/stopping blk-mq
with holding vq_lock.
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Cc: stable@kernel.org
Signed-off-by: Jens Axboe <axboe@fb.com>
Conflicts:
drivers/block/virtio_blk.c
With the kernel running from FastRAM instead of ST-RAM, none of ST-RAM is
mapped by mem_init, and DMA-addressable buffer must be mapped by ioremap.
Use platform specific virt/phys translation helpers for this case.
Signed-off-by: Michael Schmitz <schmitz@debian.org>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Move error handling to service thread, and use mtip_set_timeout()
to set timeouts for HDIO_DRIVE_TASK and HDIO_DRIVE_CMD IOCTL commands.
Signed-off-by: Selvan Mani <smani@micron.com>
Signed-off-by: Asai Thambi S P <asamymuthupa@micron.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
When there isn't enough vring descriptor for adding to vq,
blk-mq will be put as stopped state until some of pending
descriptors are completed & freed.
Unfortunately, the vq's interrupt may come just before
blk-mq's BLK_MQ_S_STOPPED flag is set, so the blk-mq will
still be kept as stopped even though lots of descriptors
are completed and freed in the interrupt handler. The worst
case is that all pending descriptors are freed in the
interrupt handler, and the queue is kept as stopped forever.
This patch fixes the problem by starting/stopping blk-mq
with holding vq_lock.
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
We need to stop the block layer queues to prevent new "normal"
IO from entering the driver, while we wait for existing commands
to finish.
Signed-off-by: Jens Axboe <axboe@fb.com>
We changed this from blk_alloc_queue_node() to blk_mq_init_queue() so
the check needs to be updated as well.
Fixes: ffc771b3ca ('mtip32xx: convert to use blk-mq')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
This rips out timeout handling, requeueing, etc in converting
it to use blk-mq instead.
Acked-by: Asai Thambi S P <asamymuthupa@micron.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Since _nvme_check_size() wasn't being called from anywhere, the compiler
was optimising it away ... along with all the link-time build failures
that would result if any of the structures were the wrong size. Call it
from nvme_exit() for no particular reason.
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
It is possible a filesystem may send a flush flagged bio with write
data. There is no such composite NVMe command, so the driver sends flush
and write separately.
The device is allowed to execute these commands in any order, so it was
possible the driver ends the bio after the write completes, but while the
flush is still active. We don't want to let a filesystem believe flush
succeeded before it really has; this could cause data corruption on a
power loss between these events. To fix, this patch splits the flush
and write into chained bios.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
This configures an nvme request_queue as flush capable if the device
has a volatile write cache present.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Do not leak kernel-only floppy_raw_cmd structure members to userspace.
This includes the linked-list pointer and the pointer to the allocated
DMA space.
Signed-off-by: Matthew Daley <mattd@bugfuzz.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Always clear out these floppy_raw_cmd struct members after copying the
entire structure from userspace so that the in-kernel version is always
valid and never left in an interdeterminate state.
Signed-off-by: Matthew Daley <mattd@bugfuzz.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Adding tracepoints for bio_complete and block_split into nvme to help
with gathering IO info using blktrace and blkparse.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
If a misbehaving device posts a CQE with a command id < depth but for
one that was never allocated, the command info will have a callback
function set to NULL and we don't want to try invoking that.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Help people diagnose what is going wrong at initialisation time by
printing out which command has gone wrong and what the device returned.
Also fix the error message printed while waiting for reset.
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Make the copyright dates accurate and remove the final paragraph that
includes the address of the FSF.
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
entry(cmd->ll_list) may belong to new request once end_cmd()
returns, so fix the bug with the patch.
Without the change, it is easy to observe oops when
doing null_blk(timer) test.
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
If there are no peer_devices or connections, I'd rather have NULL
than some "arbitrary" address pretending to point to a struct.
Helps to avoid hard to debug symptoms, in case we ever try to use
and dereference a drbd_connection or drbd_peer_device
where we in fact don't have any connection at all.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
A newly created device was never exposed before, i.e. has a
exposed_data_uuid of 0. Then it is valid to attach to any current_uuid
of a backing device (of course also to a newly created one (4))
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
In case a connection transitions into C_TIMEOUT within the timer
function (request_timer_fn()) we need to make sure that the receiver
thread (potentially running on a different CPU) sees the updated
cstate later on.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Just because it is the oldest not yet completed request
does not make it the oldest request waiting for disk.
Or waiting for the peer.
And we completely missed already completed requests
that would still hold references to activity log extents,
waiting only for the barrier ack.
Find two oldest not yet completely processed requests,
one that is still waiting for local completion,
and one that is still waiting for some response from the peer.
These may or may not be the same request object.
Then separately apply the network and disk timeouts, respectively.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
In the implementation as it was, the two peers sent each other
a challenge, and expects the challenge hashed with the shared
secret back.
A attacker could simply wait for the challenge of the peer, and
send the same challenge back. Then it waits for the response, and
sends the same response back.
Prevent this by not accepting a challenge from the peer that is
the same as the challenge sent to the peer.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Once our sender thread needs to wait_for_work(),
and actually needs to schedule(), just before we do that,
we already check if it is useful to implicitly close the last epoch.
The condition was too strict: only implicitly close the epoch,
if there have been no new (write) requests at all.
The assumption was that if there were new requests, they would
always be communicated one way or another, and would send necessary
epoch separating barriers explicitly.
This is not always true, e.g. when becoming diskless,
or while explicitly starting a full resync.
The last communicated epoch could stay open for a long time,
locking down corresponding activity log extents.
It is safe to always implicitly send that last barrier, as soon as we
determin that there cannot be more requests in the last communicated
epoch, even if there have been (uncommunicated) new requests in new
epochs meanwhile.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
When batching more updates to the activity log into single transactions,
we lost the ability for new requests to force themselves into the active
set: all preparation steps became non-blocking, and if all currently
hot extents keep busy, they could starve out new incoming requests
to cold extents for quite a while.
This can only happen if your IO backend accepts more IO operations per
average DRBD replication round trip time than you have al-extents
configured.
If we have incoming requests to cold extents,
at least do one blocking update per transaction.
In an artificial worst-case workload on SSD with an asynchronous 600 ms
replication link, with al-extents = 7 (the minimum we allow), and
concurrent full resynch, without this patch, some write requests have
been observed to be starved for 40 seconds.
With this patch, application observed a worst case latency of twice the
replication round trip time.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
We want to store in persistent meta data what the peer DRBD can handle,
which, due to spreading requests to multiple bios,
may be more than its backing device can handle.
Otherwise, if a disconnected Primary temporarily loses access to its local data
as well, we may accidentally shrink the max-bio setting, portentially causing
already assembled, but not yet processed, application bios to be spuriously
failed due to device limits.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
In the drbd make request function, specifically in
drbd_send_and_submit(), we decide whether we want to send the actual
write request, or only a "set this block out of sync" information.
We do so based on the current connection state, while holding the req_lock.
The connection state is not supposed to change while holding the req_lock.
But in drbd_start_resync, we did change that state anyways,
while only holding the global_state_lock, which is enough to change
sync-after dependencies (paused vs active resync), but
not good enough to change the connection state.
Fix: in drbd_start_resync, first grab the req_lock to serialize with
drbd_send_and_submit(), before grabbing the global_state_lock
to be able to evaluate the sync-after dependencies.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Allow the user of REQ_DISCARD.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Note that I do NOT call __drbd_chk_io_error for failed REQ_DISCARD.
That may be wrong, though, or needs to differ between EOPNOTSUPP and
other errors...
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
If the receiver needs to serve a discard request on a queue that does
not announce to be discard cabable, it falls back to do synchronous
blkdev_issue_zeroout().
We expect only "reasonably" large (up to one activity log extent?)
discard requests.
We do this to not to not block the receiver for too long in this
fallback code path, and to not set/clear too many bits inside one
spinlock_irq_save() in drbd_set_in_sync/drbd_set_out_of_sync,
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
We plan to use genl_family->parallel_ops = true in the future,
but need to review all possible interactions first.
For now, only selectively drop genl_lock() in drbd_set_role(),
instead serializing on our own internal resource->conf_update mutex.
We now can be promoted/demoted on many resources in parallel,
which may significantly improve cluster failover times
when fencing is required.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Because all administrative requests via genetlink have been globally
serialized via genl_lock(), we used to have one static struct
drbd_config_context "admin context".
Move this on-stack to the respective callback functions.
This will allow us to selectively drop the genl_lock()
(or use genl_family->parallel_ops) in the future.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
When a 'cluster wide' disconnect executes, the result comes back
from the peer, and immediately after that the connection breaks
then _conn_rq_cond() reported back SS_CW_SUCCESS.
Therefore _conn_request_state() calls conn_set_state(), which
has a BUG() in it.
The BUG() is hit because conn_is_valid_transition() does not like
the transaction. Which goes back to is_valid_soft_transition()
returning SS_OUTDATE_WO_CONN.
This fix is to consider an error reported by is_valid_soft_transition()
even when the peer agreed to the transaction.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Before, application IO could pre-empt resync activity
for up to hardcoded 20 seconds per resync request.
A very busy server could throttle the effective resync bandwidth
down to one request per 20 seconds.
Now, we only let application IO pre-empt resync traffic
while the current resync rate estimate is above c-min-rate.
If you disable the c-min-rate throttle feature (set c-min-rate = 0),
application IO will no longer pre-empt resync traffic at all.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
If max-buffers and socket buffer sizes are "too small" for the chosen
resync rate, this could lead potentially lead to a distributed deadlock,
which may or may not resolve itself via the "ko-count" and request
timeout mechanism, or could be resolved by forced disconnect.
One option to deal with this is proper configuration:
use larger max-buffer and socket buffers settings,
or reduce the resync rate.
But even with bad configuration we should not deadlock,
but "gracefully" recover.
The issue is avoided by using only up to max-buffers/2 for resync
requests, and by using max-buffers not as a hard limit for data buffer
allocations, but as a throttle threshold only.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
While merging adjacent dirty blocks into resync requests,
the resync rate throttle was disregarded.
For very low resync rates, the effective rate may have exceeded
the intended rate by a larger margin.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
If we don't make resync or verify progress for "too long",
we want to flag it as "stalled".
Since 2010, "use rolling marks for resync speed calculation"
this "too long" was wrong by a factor of HZ.
With HZ 250, it would have been flagged as stalled
after 100 minutes.
Hardcode 3 minutes instead.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
If a user forces the operation he takes the blame in case
the peer does not have enough space. No reason to dey this...
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Actually we are clearing the susp_fen flag if we are not going
to call a fencing handler.
For setting the susp_fen flag needs to be edge-triggerd, and not
level triggered.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
When we need to outdate the peer while being promoted to primary,
and the connection gets established at the same time, we deadlock
in drbd_try_outdate_peer() when trying to clear the susp_fen
bit.
Fix this by setting the STATE_SENT bit while holding the mutex.
Using drbd_change_state(.. , CS_HARD, ..) which does not block
until STATE_SENT is cleared, is only for clearness. It does
not contribute anything to the fix.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
A hardware quirk in P320h/P420m interfere with PCIe transactions on some
AMD chipsets, making P320h/P420m unusable. This workaround is to disable
ERO and NoSnoop bits in the parent and root complex for normal
functioning of these devices
NOTE: This workaround is specific to AMD chipset with a PCIe upstream
device with device id 0x5aXX
Signed-off-by: Asai Thambi S P <asamymuthupa@micron.com>
Signed-off-by: Sam Bradshaw <sbradshaw@micron.com>
Cc: stable@kernel.org
Signed-off-by: Jens Axboe <axboe@fb.com>
In module exit, dfs_parent and it's subtree were removed before
unregistering with pci. When debugfs entry for each device is attempted
to remove in pci_remove() context, they don't exist, as dfs_parent and
its children were already ripped apart.
Modified to first unregister with pci and then remove dfs_parent.
Signed-off-by: Asai Thambi S P <asamymuthupa@micron.com>
Cc: stable@kernel.org
Signed-off-by: Jens Axboe <axboe@fb.com>
As result of deprecation of MSI-X/MSI enablement functions
pci_enable_msix() and pci_enable_msi_block() all drivers
using these two interfaces need to be updated to use the
new pci_enable_msi_range() or pci_enable_msi_exact()
and pci_enable_msix_range() or pci_enable_msix_exact()
interfaces.
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Cc: Mike Miller <mike.miller@hp.com>
Cc: iss_storagedev@hp.com
Cc: Jens Axboe <axboe@kernel.dk>
Cc: linux-pci@vger.kernel.org
Signed-off-by: Jens Axboe <axboe@fb.com>
Function pci_enable_msix_exact() is a variation of
pci_enable_msix_range() that allows a device driver
to request a particular number of MSI-X interrupts,
rather than any number within a specified range.
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: linux-pci@vger.kernel.org
Signed-off-by: Jens Axboe <axboe@fb.com>
Store the pointer to the page there, so we can always safely
reference it from end_io context where ->bio may have been
cleared.
Signed-off-by: Jens Axboe <axboe@fb.com>
Add a new blk_mq_tag_set structure that gets set up before we initialize
the queue. A single blk_mq_tag_set structure can be shared by multiple
queues.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Modular export of blk_mq_{alloc,free}_tagset added by me.
Signed-off-by: Jens Axboe <axboe@fb.com>
The current blk_mq_init_commands/blk_mq_free_commands interface has a
two problems:
1) Because only the constructor is passed to blk_mq_init_commands there
is no easy way to clean up when a comman initialization failed. The
current code simply leaks the allocations done in the constructor.
2) There is no good place to call blk_mq_free_commands: before
blk_cleanup_queue there is no guarantee that all outstanding
commands have completed, so we can't free them yet. After
blk_cleanup_queue the queue has usually been freed. This can be
worked around by grabbing an unconditional reference before calling
blk_cleanup_queue and dropping it after blk_mq_free_commands is
done, although that's not exatly pretty and driver writers are
guaranteed to get it wrong sooner or later.
Both issues are easily fixed by making the request constructor and
destructor normal blk_mq_ops methods.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Drivers can reach their private data easily using the blk_mq_rq_to_pdu
helper and don't need req->special. By not initializing it code can
be simplified nicely, and we also shave off a few more instructions from
the I/O path.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
This was used in the olden days, back when onions were proper
yellow. Basically it mapped to the current buffer to be
transferred. With highmem being added more than a decade ago,
most drivers map pages out of a bio, and rq->buffer isn't
pointing at anything valid.
Convert old style drivers to just use bio_data().
For the discard payload use case, just reference the page
in the bio.
Signed-off-by: Jens Axboe <axboe@fb.com>
Pull vfs updates from Al Viro:
"The first vfs pile, with deep apologies for being very late in this
window.
Assorted cleanups and fixes, plus a large preparatory part of iov_iter
work. There's a lot more of that, but it'll probably go into the next
merge window - it *does* shape up nicely, removes a lot of
boilerplate, gets rid of locking inconsistencie between aio_write and
splice_write and I hope to get Kent's direct-io rewrite merged into
the same queue, but some of the stuff after this point is having
(mostly trivial) conflicts with the things already merged into
mainline and with some I want more testing.
This one passes LTP and xfstests without regressions, in addition to
usual beating. BTW, readahead02 in ltp syscalls testsuite has started
giving failures since "mm/readahead.c: fix readahead failure for
memoryless NUMA nodes and limit readahead pages" - might be a false
positive, might be a real regression..."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (63 commits)
missing bits of "splice: fix racy pipe->buffers uses"
cifs: fix the race in cifs_writev()
ceph_sync_{,direct_}write: fix an oops on ceph_osdc_new_request() failure
kill generic_file_buffered_write()
ocfs2_file_aio_write(): switch to generic_perform_write()
ceph_aio_write(): switch to generic_perform_write()
xfs_file_buffered_aio_write(): switch to generic_perform_write()
export generic_perform_write(), start getting rid of generic_file_buffer_write()
generic_file_direct_write(): get rid of ppos argument
btrfs_file_aio_write(): get rid of ppos
kill the 5th argument of generic_file_buffered_write()
kill the 4th argument of __generic_file_aio_write()
lustre: don't open-code kernel_recvmsg()
ocfs2: don't open-code kernel_recvmsg()
drbd: don't open-code kernel_recvmsg()
constify blk_rq_map_user_iov() and friends
lustre: switch to kernel_sendmsg()
ocfs2: don't open-code kernel_sendmsg()
take iov_iter stuff to mm/iov_iter.c
process_vm_access: tidy up a bit
...
Pull NVMe driver updates from Matthew Wilcox:
"Various updates to the NVMe driver. The most user-visible change is
that drive hotplugging now works and CPU hotplug while an NVMe drive
is installed should also work better"
* git://git.infradead.org/users/willy/linux-nvme:
NVMe: Retry failed commands with non-fatal errors
NVMe: Add getgeo to block ops
NVMe: Start-stop nvme_thread during device add-remove.
NVMe: Make I/O timeout a module parameter
NVMe: CPU hot plug notification
NVMe: per-cpu io queues
NVMe: Replace DEFINE_PCI_DEVICE_TABLE
NVMe: Fix divide-by-zero in nvme_trans_io_get_num_cmds
NVMe: IOCTL path RCU protect queue access
NVMe: RCU protected access to io queues
NVMe: Initialize device reference count earlier
NVMe: Add CONFIG_PM_SLEEP to suspend/resume functions
For commands returned with failed status, queue these for resubmission
and continue retrying them until success or for a limited amount of
time. The final timeout was arbitrarily chosen so requests can't be
retried indefinitely.
Since these are requeued on the nvmeq that submitted the command, the
callbacks have to take an nvmeq instead of an nvme_dev as a parameter
so that we can use the locked queue to append the iod to retry later.
The nvme_iod conviently can be used to track how long we've been trying
to successfully complete an iod request. The nvme_iod also provides the
nvme prp dma mappings, so I had to move a few things around so we can
keep those mappings.
Signed-off-by: Keith Busch <keith.busch@intel.com>
[fixed checkpatch issue with long line]
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Some programs require HDIO_GETGEO work, which requires we implement
getgeo.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Done to ensure nvme_thread is not running when there
are no devices to poll.
Signed-off-by: Dan McLeran <daniel.mcleran@intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Increase the default timeout to 30 seconds to match SCSI.
Signed-off-by: Keith Busch <keith.busch@intel.com>
[use byte instead of ushort]
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Registers with hot cpu notification to rebalance, and potentially allocate
additional, io queues.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
The device's IO queues are associated with CPUs, so we can use a per-cpu
variable to map the a qid to a cpu. This provides a convienient way
to optimally assign queues to multiple cpus when the device supports
fewer queues than the host has cpus. The previous implementation may
have assigned these poorly in these situations. This patch addresses
this by sharing queues among cpus that are "close" together and should
have a lower lock contention penalty.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Pull block layer fixes from Jens Axboe:
"A small collection of fixes that should go in before -rc1. The pull
request contains:
- A two patch fix for a regression with block enabled tagging caused
by a commit in the initial pull request. One patch is from Martin
and ensures that SCSI doesn't truncate 64-bit block flags, the
other one is from me and prevents us from double using struct
request queuelist for both completion and busy tags. This caused
anything from a boot crash for some, to crashes under load.
- A blk-mq fix for a potential soft stall when hot unplugging CPUs
with busy IO.
- percpu_counter fix is listed in here, that caused a suspend issue
with virtio-blk due to percpu counters having an inconsistent state
during CPU removal. Andrew sent this in separately a few days ago,
but it's here. JFYI.
- A few fixes for block integrity from Martin.
- A ratelimit fix for loop from Mike Galbraith, to avoid spewing too
much in error cases"
* 'for-linus' of git://git.kernel.dk/linux-block:
block: fix regression with block enabled tagging
scsi: Make sure cmd_flags are 64-bit
block: Ensure we only enable integrity metadata for reads and writes
block: Fix integrity verification
block: Fix for_each_bvec()
drivers/block/loop.c: ratelimit error messages
blk-mq: fix potential stall during CPU unplug with IO pending
percpu_counter: fix bad counter state during suspend
Merge second patch-bomb from Andrew Morton:
- the rest of MM
- zram updates
- zswap updates
- exit
- procfs
- exec
- wait
- crash dump
- lib/idr
- rapidio
- adfs, affs, bfs, ufs
- cris
- Kconfig things
- initramfs
- small amount of IPC material
- percpu enhancements
- early ioremap support
- various other misc things
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (156 commits)
MAINTAINERS: update Intel C600 SAS driver maintainers
fs/ufs: remove unused ufs_super_block_third pointer
fs/ufs: remove unused ufs_super_block_second pointer
fs/ufs: remove unused ufs_super_block_first pointer
fs/ufs/super.c: add __init to init_inodecache()
doc/kernel-parameters.txt: add early_ioremap_debug
arm64: add early_ioremap support
arm64: initialize pgprot info earlier in boot
x86: use generic early_ioremap
mm: create generic early_ioremap() support
x86/mm: sparse warning fix for early_memremap
lglock: map to spinlock when !CONFIG_SMP
percpu: add preemption checks to __this_cpu ops
vmstat: use raw_cpu_ops to avoid false positives on preemption checks
slub: use raw_cpu_inc for incrementing statistics
net: replace __this_cpu_inc in route.c with raw_cpu_inc
modules: use raw_cpu_write for initialization of per cpu refcount.
mm: use raw_cpu ops for determining current NUMA node
percpu: add raw_cpu_ops
slub: fix leak of 'name' in sysfs_slab_add
...
zram is ram based block device and can be used by backend of filesystem.
When filesystem deletes a file, it normally doesn't do anything on data
block of that file. It just marks on metadata of that file. This
behavior has no problem on disk based block device, but has problems on
ram based block device, since we can't free memory used for data block.
To overcome this disadvantage, there is REQ_DISCARD functionality. If
block device support REQ_DISCARD and filesystem is mounted with discard
option, filesystem sends REQ_DISCARD to block device whenever some data
blocks are discarded. All we have to do is to handle this request.
This patch implements to flag up QUEUE_FLAG_DISCARD and handle this
REQ_DISCARD request. With it, we can free memory used by zram if it isn't
used.
[akpm@linux-foundation.org: tweak comments]
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Jerome Marchand <jmarchan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
sysfs.txt documentation lists the following requirements:
- The buffer will always be PAGE_SIZE bytes in length. On i386, this
is 4096.
- show() methods should return the number of bytes printed into the
buffer. This is the return value of scnprintf().
- show() should always use scnprintf().
Use scnprintf() in show() functions.
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When we initialized zcomp with single, we couldn't change
max_comp_streams without zram reset but current interface doesn't show
any error to user and even it changes max_comp_streams's value without
any effect so it would make user very confusing.
This patch prevents max_comp_streams's change when zcomp was initialized
as single zcomp and emit the error to user(ex, echo).
[akpm@linux-foundation.org: don't return with the lock held, per Sergey]
[fengguang.wu@intel.com: fix coccinelle warnings]
Signed-off-by: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Acked-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Instead of returning just NULL, return ERR_PTR from zcomp_create() if
compressing backend creation has failed. ERR_PTR(-EINVAL) for unsupported
compression algorithm request, ERR_PTR(-ENOMEM) for allocation (zcomp or
compression stream) error.
Perform IS_ERR() check of returned from zcomp_create() value in
disksize_store() and set return code to PTR_ERR().
Change suggested by Jerome Marchand.
[akpm@linux-foundation.org: clean up error recovery flow]
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reported-by: Jerome Marchand <jmarchan@redhat.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
While fixing lockdep spew of ->init_lock reported by Sasha Levin [1],
Minchan Kim noted [2] that it's better to move compression backend
allocation (using GPF_KERNEL) out of the ->init_lock lock, same way as
with zram_meta_alloc(), in order to prevent the same lockdep spew.
[1] https://lkml.org/lkml/2014/2/27/337
[2] https://lkml.org/lkml/2014/3/3/32
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reported-by: Minchan Kim <minchan@kernel.org>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Sasha Levin <sasha.levin@oracle.com>
Acked-by: Jerome Marchand <jmarchan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Introduce LZ4 compression backend and make it available for selection.
LZ4 support is optional and requires user to set ZRAM_LZ4_COMPRESS config
option. The default compression backend is LZO.
TEST
(x86_64, core i5, 2 cores + 2 hyperthreading, zram disk size 1G,
ext4 file system, 3 compression streams)
iozone -t 3 -R -r 16K -s 60M -I +Z
Test LZO LZ4
----------------------------------------------
Initial write 1642744.62 1317005.09
Rewrite 2498980.88 1800645.16
Read 3957026.38 5877043.75
Re-read 3950997.38 5861847.00
Reverse Read 2937114.56 5047384.00
Stride read 2948163.19 4929587.38
Random read 3292692.69 4880793.62
Mixed workload 1545602.62 3502940.38
Random write 2448039.75 1758786.25
Pwrite 1670051.03 1338329.69
Pread 2530682.00 5097177.62
Fwrite 3232085.62 3275942.56
Fread 6306880.25 6645271.12
So on my system LZ4 is slower in write-only tests, while it performs
better in read-only and mixed (reads + writes) tests.
Official LZ4 benchmarks available here http://code.google.com/p/lz4/
(linux kernel uses revision r90).
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch allows to change max_comp_streams on initialised zcomp.
Introduce zcomp set_max_streams() knob, zcomp_strm_multi_set_max_streams()
and zcomp_strm_single_set_max_streams() callbacks to change streams limit
for zcomp_strm_multi and zcomp_strm_single, accordingly. set_max_streams
for single steam zcomp does nothing.
If user has lowered the limit, then zcomp_strm_multi_set_max_streams()
attempts to immediately free extra streams (as much as it can, depending
on idle streams availability).
Note, this patch does not allow to change stream 'policy' from single to
multi stream (or vice versa) on already initialised compression backend.
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Existing zram (zcomp) implementation has only one compression stream
(buffer and algorithm private part), so in order to prevent data
corruption only one write (compress operation) can use this compression
stream, forcing all concurrent write operations to wait for stream lock
to be released. This patch changes zcomp to keep a compression streams
list of user-defined size (via sysfs device attr). Each write operation
still exclusively holds compression stream, the difference is that we
can have N write operations (depending on size of streams list)
executing in parallel. See TEST section later in commit message for
performance data.
Introduce struct zcomp_strm_multi and a set of functions to manage
zcomp_strm stream access. zcomp_strm_multi has a list of idle
zcomp_strm structs, spinlock to protect idle list and wait queue, making
it possible to perform parallel compressions.
The following set of functions added:
- zcomp_strm_multi_find()/zcomp_strm_multi_release()
find and release a compression stream, implement required locking
- zcomp_strm_multi_create()/zcomp_strm_multi_destroy()
create and destroy zcomp_strm_multi
zcomp ->strm_find() and ->strm_release() callbacks are set during
initialisation to zcomp_strm_multi_find()/zcomp_strm_multi_release()
correspondingly.
Each time zcomp issues a zcomp_strm_multi_find() call, the following set
of operations performed:
- spin lock strm_lock
- if idle list is not empty, remove zcomp_strm from idle list, spin
unlock and return zcomp stream pointer to caller
- if idle list is empty, current adds itself to wait queue. it will be
awaken by zcomp_strm_multi_release() caller.
zcomp_strm_multi_release():
- spin lock strm_lock
- add zcomp stream to idle list
- spin unlock, wake up sleeper
Minchan Kim reported that spinlock-based locking scheme has demonstrated
a severe perfomance regression for single compression stream case,
comparing to mutex-based (see https://lkml.org/lkml/2014/2/18/16)
base spinlock mutex
==Initial write ==Initial write ==Initial write
records: 5 records: 5 records: 5
avg: 1642424.35 avg: 699610.40 avg: 1655583.71
std: 39890.95(2.43%) std: 232014.19(33.16%) std: 52293.96
max: 1690170.94 max: 1163473.45 max: 1697164.75
min: 1568669.52 min: 573429.88 min: 1553410.23
==Rewrite ==Rewrite ==Rewrite
records: 5 records: 5 records: 5
avg: 1611775.39 avg: 501406.64 avg: 1684419.11
std: 17144.58(1.06%) std: 15354.41(3.06%) std: 18367.42
max: 1641800.95 max: 531356.78 max: 1706445.84
min: 1593515.27 min: 488817.78 min: 1655335.73
When only one compression stream available, mutex with spin on owner
tends to perform much better than frequent wait_event()/wake_up(). This
is why single stream implemented as a special case with mutex locking.
Introduce and document zram device attribute max_comp_streams. This
attr shows and stores current zcomp's max number of zcomp streams
(max_strm). Extend zcomp's zcomp_create() with `max_strm' parameter.
`max_strm' limits the number of zcomp_strm structs in compression
backend's idle list (max_comp_streams).
max_comp_streams used during initialisation as follows:
-- passing to zcomp_create() max_strm equals to 1 will initialise zcomp
using single compression stream zcomp_strm_single (mutex-based locking).
-- passing to zcomp_create() max_strm greater than 1 will initialise zcomp
using multi compression stream zcomp_strm_multi (spinlock-based locking).
default max_comp_streams value is 1, meaning that zram with single stream
will be initialised.
Later patch will introduce configuration knob to change max_comp_streams
on already initialised and used zcomp.
TEST
iozone -t 3 -R -r 16K -s 60M -I +Z
test base 1 strm (mutex) 3 strm (spinlock)
-----------------------------------------------------------------------
Initial write 589286.78 583518.39 718011.05
Rewrite 604837.97 596776.38 1515125.72
Random write 584120.11 595714.58 1388850.25
Pwrite 535731.17 541117.38 739295.27
Fwrite 1418083.88 1478612.72 1484927.06
Usage example:
set max_comp_streams to 4
echo 4 > /sys/block/zram0/max_comp_streams
show current max_comp_streams (default value is 1).
cat /sys/block/zram0/max_comp_streams
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is preparation patch to add multi stream support to zcomp.
Introduce struct zcomp_strm_single and a set of functions to manage
zcomp_strm stream access. zcomp_strm_single implements single compession
stream, same way as current zcomp implementation. This moves zcomp_strm
stream control and locking from zcomp, so compressing backend zcomp is not
aware of required locking.
Single and multi streams require different locking schemes. Minchan Kim
reported that spinlock-based locking scheme (which is used in multi stream
implementation) has demonstrated a severe perfomance regression for single
compression stream case, comparing to mutex-based. see
https://lkml.org/lkml/2014/2/18/16
The following set of functions added:
- zcomp_strm_single_find()/zcomp_strm_single_release()
find and release a compression stream, implement required locking
- zcomp_strm_single_create()/zcomp_strm_single_destroy()
create and destroy zcomp_strm_single
New ->strm_find() and ->strm_release() callbacks added to zcomp, which are
set to zcomp_strm_single_find() and zcomp_strm_single_release() during
initialisation. Instead of direct locking and zcomp_strm access from
zcomp_strm_find() and zcomp_strm_release(), zcomp now calls ->strm_find()
and ->strm_release() correspondingly.
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
ZRAM performs direct LZO compression algorithm calls, making it the one
and only option. While LZO is generally performs well, LZ4 algorithm
tends to have a faster decompression (see http://code.google.com/p/lz4/
for full report)
Name Ratio C.speed D.speed
MB/s MB/s
LZ4 (r101) 2.084 422 1820
LZO 2.06 2.106 414 600
Thus, users who have mostly read (decompress) usage scenarious or mixed
workflow (writes with relatively high read ops number) will benefit from
using LZ4 compression backend.
Introduce compressing backend abstraction zcomp in order to support
multiple compression algorithms with the following set of operations:
.create
.destroy
.compress
.decompress
Schematically zram write() usually contains the following steps:
0) preparation (decompression of partioal IO, etc.)
1) lock buffer_lock mutex (protects meta compress buffers)
2) compress (using meta compress buffers)
3) alloc and map zs_pool object
4) copy compressed data (from meta compress buffers) to object allocated by 3)
5) free previous pool page, assign a new one
6) unlock buffer_lock mutex
As we can see, compressing buffers must remain untouched from 1) to 4),
because, otherwise, concurrent write() can overwrite data. At the same
time, zram_meta must be aware of a) specific compression algorithm memory
requirements and b) necessary locking to protect compression buffers. To
remove requirement a) new struct zcomp_strm introduced, which contains a
compress/decompress `buffer' and compression algorithm `private' part.
While struct zcomp implements zcomp_strm stream handling and locking and
removes requirement b) from zram meta. zcomp ->create() and ->destroy(),
respectively, allocate and deallocate algorithm specific zcomp_strm
`private' part.
Every zcomp has zcomp stream and mutex to protect its compression stream.
Stream usage semantics remains the same -- only one write can hold stream
lock and use its buffers. zcomp_strm_find() turns caller into exclusive
user of a stream (holding stream mutex until zram release stream), and
zcomp_strm_release() makes zcomp stream available (unlock the stream
mutex). Hence no concurrent write (compression) operations possible at
the moment.
iozone -t 3 -R -r 16K -s 60M -I +Z
test base patched
--------------------------------------------------
Initial write 597992.91 591660.58
Rewrite 609674.34 616054.97
Read 2404771.75 2452909.12
Re-read 2459216.81 2470074.44
Reverse Read 1652769.66 1589128.66
Stride read 2202441.81 2202173.31
Random read 2236311.47 2276565.31
Mixed workload 1423760.41 1709760.06
Random write 579584.08 615933.86
Pwrite 597550.02 594933.70
Pread 1703672.53 1718126.72
Fwrite 1330497.06 1461054.00
Fread 3922851.00 3957242.62
Usage examples:
comp = zcomp_create(NAME) /* NAME e.g. "lzo" */
which initialises compressing backend if requested algorithm is supported.
Compress:
zstrm = zcomp_strm_find(comp)
zcomp_compress(comp, zstrm, src, &dst_len)
[..] /* copy compressed data */
zcomp_strm_release(comp, zstrm)
Decompress:
zcomp_decompress(comp, src, src_len, dst);
Free compessing backend and its zcomp stream:
zcomp_destroy(comp)
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
allocate new `zram_meta' in disksize_store() only for uninitialised zram
device, saving a number of allocations and deallocations in case if
disksize_store() was called on currently used device. at the same time
zram_meta stack variable is not necessary, because we can set ->meta
directly. there is also no need in setting QUEUE_FLAG_NONROT queue on
every disksize_store(), set it once during device creation.
[minchan@kernel.org: handle zram->meta alloc fail case]
[minchan@kernel.org: prevent lockdep spew of init_lock]
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Signed-off-by: Minchan Kim <minchan@kernel.org>
Acked-by: Jerome Marchand <jmarchan@redhat.com>
Cc: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
zram accounted but did not report numbers of failed read and write
queries. make these stats available as failed_reads and failed_writes
attrs.
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Acked-by: Jerome Marchand <jmarchan@redhat.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is a preparation patch for stats code duplication removal.
1) use atomic64_t for `pages_zero' and `pages_stored' zram stats.
2) `compr_size' and `pages_zero' struct zram_stats members did not
follow the existing device attr naming scheme: zram_stats.ATTR has
ATTR_show() function. rename them:
-- compr_size -> compr_data_size
-- pages_zero -> zero_pages
Minchan Kim's note:
If we really have trouble with atomic stat operation, we could
change it with percpu_counter so that it could solve atomic overhead and
unnecessary memory space by introducing unsigned long instead of 64bit
atomic_t.
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Acked-by: Jerome Marchand <jmarchan@redhat.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Remove `good' and `bad' compressed sub-requests stats. RW request may
cause a number of RW sub-requests. zram used to account `good' compressed
sub-queries (with compressed size less than 50% of original size), `bad'
compressed sub-queries (with compressed size greater that 75% of original
size), leaving sub-requests with compression size between 50% and 75% of
original size not accounted and not reported. zram already accounts each
sub-request's compression size so we can calculate real device compression
ratio.
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Acked-by: Jerome Marchand <jmarchan@redhat.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Do not pass rw argument down the __zram_make_request() -> zram_bvec_rw()
chain, decode it in zram_bvec_rw() instead. Besides, this is the place
where we distinguish READ and WRITE bio data directions, so account zram
RW stats here, instead of __zram_make_request(). This also allows to
account a real number of zram READ/WRITE operations, not just requests
(single RW request may cause a number of zram RW ops with separate
locking, compression/decompression, etc).
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Acked-by: Jerome Marchand <jmarchan@redhat.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Introduce init_done() helper function which allows us to drop `init_done'
struct zram member. init_done() uses the fact that ->init_done == 1
equals to ->meta != NULL.
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Acked-by: Jerome Marchand <jmarchan@redhat.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull Ceph updates from Sage Weil:
"The biggest chunk is a series of patches from Ilya that add support
for new Ceph osd and crush map features, including some new tunables,
primary affinity, and the new encoding that is needed for erasure
coding support. This brings things into parity with the server side
and the looming firefly release. There is also support for allocation
hints in RBD that help limit fragmentation on the server side.
There is also a series of patches from Zheng fixing NFS reexport,
directory fragmentation support, flock vs fnctl behavior, and some
issues with clustered MDS.
Finally, there are some miscellaneous fixes from Yunchuan Wen for
fscache, Fabian Frederick for ACLs, and from me for fsync(dirfd)
behavior"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client: (79 commits)
ceph: skip invalid dentry during dcache readdir
libceph: dump pool {read,write}_tier to debugfs
libceph: output primary affinity values on osdmap updates
ceph: flush cap release queue when trimming session caps
ceph: don't grabs open file reference for aborted request
ceph: drop extra open file reference in ceph_atomic_open()
ceph: preallocate buffer for readdir reply
libceph: enable PRIMARY_AFFINITY feature bit
libceph: redo ceph_calc_pg_primary() in terms of ceph_calc_pg_acting()
libceph: add support for osd primary affinity
libceph: add support for primary_temp mappings
libceph: return primary from ceph_calc_pg_acting()
libceph: switch ceph_calc_pg_acting() to new helpers
libceph: introduce apply_temps() helper
libceph: introduce pg_to_raw_osds() and raw_to_up_osds() helpers
libceph: ceph_can_shift_osds(pool) and pool type defines
libceph: ceph_osd_{exists,is_up,is_down}(osd) definitions
libceph: enable OSDMAP_ENC feature bit
libceph: primary_affinity decode bits
libceph: primary_affinity infrastructure
...
In an effort to reduce fragmentation, prefix every rbd write with
a CEPH_OSD_OP_SETALLOCHINT osd op with an expected_write_size value set
to the object size (1 << order). Backwards compatibility is taken care
of on the libceph/osd side.
"The CEPH_OSD_OP_SETALLOCHINT hint is durable, in that it's enough to
do it once. The reason every rbd write is prefixed is that rbd doesn't
explicitly create objects and relies on writes creating them
implicitly, so there is no place to stick a single hint op into. To
get around that we decided to prefix every rbd write with a hint (just
like write and setattr ops, hint op will create an object implicitly if
it doesn't exist)."
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Reviewed-by: Alex Elder <elder@linaro.org>
In preparation for prefixing rbd writes with an allocation hint
introduce a num_ops parameter for rbd_osd_req_create(). The rationale
is that not every write request is a write op that needs to be prefixed
(e.g. watch op), so the num_ops logic needs to be in the callers.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Our longest osd request now contains 3 ops: copyup+hint+write.
Also, CEPH_OSD_MAX_OP value in a BUG_ON in rbd_osd_req_callback() was
hard-coded to 2. Fix it, and switch to rbd_assert while at it.
Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Reviewed-by: Alex Elder <elder@linaro.org>
