Tejun Heo [Mon, 2 Apr 2018 22:04:58 +0000 (15:04 -0700)]
blk-mq: Directly schedule q->timeout_work when aborting a request
Request abortion is performed by overriding deadline to now and
scheduling timeout handling immediately. For the latter part, the
code was using mod_timer(timeout, 0) which can't guarantee that the
timer runs afterwards. Let's schedule the underlying work item
directly instead.
This fixes the hangs during probing reported by Sitsofe but it isn't
yet clear to me how the failure can happen reliably if it's just the
above described race condition.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Sitsofe Wheeler <sitsofe@gmail.com>
Reported-by: Meelis Roos <mroos@linux.ee>
Fixes: 358f70da49d7 ("blk-mq: make blk_abort_request() trigger timeout path")
Cc: stable@vger.kernel.org # v4.16
Link: http://lkml.kernel.org/r/CALjAwxh-PVYFnYFCJpGOja+m5SzZ8Sa4J7ohxdK=r8NyOF-EMA@mail.gmail.com
Link: http://lkml.kernel.org/r/alpine.LRH.2.21.1802261049140.4893@math.ut.ee
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Souvik Banerjee [Fri, 30 Mar 2018 19:32:42 +0000 (14:32 -0500)]
blktrace: fix comment in blktrace_api.h
The `__u64 time` field of the blk_io_trace struct refers to
the time in nanoseconds, not in microseconds. It is set in
__blk_add_trace, which does the following:
t->time = ktime_to_ns(ktime_get());
ktime_to_ns returns ktime_t in nanoseconds, not microseconds.
Signed-off-by: Souvik Banerjee <souvik1997@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Matias Bjørling [Thu, 29 Mar 2018 22:05:24 +0000 (00:05 +0200)]
lightnvm: remove function name in strings
For the sysfs functions, the function names are embedded into their
error strings. If the function name later changes, the string may
not be updated accordingly. Update the strings to use __func__
to avoid this.
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Dan Carpenter [Thu, 29 Mar 2018 22:05:23 +0000 (00:05 +0200)]
lightnvm: pblk: remove some unnecessary NULL checks
Smatch complains that flush_workqueue() dereferences the work queue
pointer but then we check if it's NULL on the next line when it's too
late. These NULL checks can be removed because the module won't load if
we can't allocate the work queues.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Hans Holmberg [Thu, 29 Mar 2018 22:05:22 +0000 (00:05 +0200)]
lightnvm: pblk: don't recover unwritten lines
If the line has not been written to, we should not
try to recover any data from it, so check the state of the
chunks in the line before attempting to read smeta.
Signed-off-by: Hans Holmberg <hans.holmberg@cnexlabs.com>
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:21 +0000 (00:05 +0200)]
lightnvm: pblk: implement 2.0 support
Implement 2.0 support in pblk. This includes the address formatting and
mapping paths, as well as the sysfs entries for them.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:20 +0000 (00:05 +0200)]
lightnvm: pblk: implement get log report chunk
In preparation of pblk supporting 2.0, implement the get log report
chunk in pblk. Also, define the chunk states as given in the 2.0 spec.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:19 +0000 (00:05 +0200)]
lightnvm: pblk: rename ppaf* to addrf*
In preparation for 2.0 support in pblk, rename variables referring to
the address format to addrf and reserve ppaf for the 1.2 path.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:18 +0000 (00:05 +0200)]
lightnvm: pblk: check for supported version
At this point, only 1.2 spec is supported, thus check for it. Also,
since device-side L2P is only supported in the 1.2 spec, make sure to
only check its value under 1.2.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:17 +0000 (00:05 +0200)]
lightnvm: implement get log report chunk helpers
The 2.0 spec provides a report chunk log page that can be retrieved
using the stangard nvme get log page. This replaces the dedicated
get/put bad block table in 1.2.
This patch implements the helper functions to allow targets retrieve the
chunk metadata using get log page. It makes nvme_get_log_ext available
outside of nvme core so that we can use it form lightnvm.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:16 +0000 (00:05 +0200)]
lightnvm: make address conversions depend on generic device
On address conversions, use the generic device, instead of the target
device. This allows to use conversions outside of the target's realm.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:15 +0000 (00:05 +0200)]
lightnvm: add support for 2.0 address format
Add support for 2.0 address format. Also, align address bits for 1.2 and
2.0 to be able to operate on channel and luns without requiring a format
conversion. Use a generic address format for this purpose.
Also, convert the generic operations to the generic format in pblk.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:14 +0000 (00:05 +0200)]
lightnvm: normalize geometry nomenclature
Normalize nomenclature for naming channels, luns, chunks, planes and
sectors as well as derivations in order to improve readability.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:13 +0000 (00:05 +0200)]
lightnvm: complete geo structure with maxoc*
Complete the generic geometry structure with the maxoc and maxocpu
felds, present in the 2.0 spec. Also, expose them through sysfs.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:12 +0000 (00:05 +0200)]
lightnvm: add shorten OCSSD version in geo
Create a shorten version to use in the generic geometry.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:11 +0000 (00:05 +0200)]
lightnvm: add minor version to generic geometry
Separate the version between major and minor on the generic geometry and
represent it through sysfs in the 2.0 path. The 1.2 path only shows the
major version to preserve the existing user space interface.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:10 +0000 (00:05 +0200)]
lightnvm: simplify geometry structure
Currently, the device geometry is stored redundantly in the nvm_id and
nvm_geo structures at a device level. Moreover, when instantiating
targets on a specific number of LUNs, these structures are replicated
and manually modified to fit the instance channel and LUN partitioning.
Instead, create a generic geometry around nvm_geo, which can be used by
(i) the underlying device to describe the geometry of the whole device,
and (ii) instances to describe their geometry independently.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:05:09 +0000 (00:05 +0200)]
lightnvm: pblk: refactor init/exit sequences
Refactor init and exit sequences to eliminate dependencies among init
modules and improve readability.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Heiner Litz [Thu, 29 Mar 2018 22:05:08 +0000 (00:05 +0200)]
lightnvm: Avoid validation of default op value
Fixes: 38401d231de65 ("lightnvm: set target over-provision on create ioctl")
Signed-off-by: Heiner Litz <hlitz@ucsc.edu>
Reviewed-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Johannes Thumshirn [Thu, 29 Mar 2018 22:05:07 +0000 (00:05 +0200)]
lightnvm: centralize permission check for lightnvm ioctl
Currently all functions for handling the lightnvm core ioctl commands
do a check for CAP_SYS_ADMIN.
Change this to fail early in nvm_ctl_ioctl(), so we don't have to
duplicate the permission checks all over.
Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Heiner Litz [Thu, 29 Mar 2018 22:05:06 +0000 (00:05 +0200)]
lightnvm: fix bad block initialization
fix reading bad block device information to correctly setup the per line
blk_bitmap during lightnvm initialization
Signed-off-by: Heiner Litz <hlitz@ucsc.edu>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Matias Bjørling [Thu, 29 Mar 2018 22:05:05 +0000 (00:05 +0200)]
nvme: lightnvm: add late setup of block size and metadata
The nvme driver sets up the size of the nvme namespace in two steps.
First it initializes the device with standard logical block and
metadata sizes, and then sets the correct logical block and metadata
size. Due to the OCSSD 2.0 specification relies on the namespace to
expose these sizes for correct initialization, let it be updated
appropriately on the LightNVM side as well.
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Acked-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Matias Bjørling [Thu, 29 Mar 2018 22:05:04 +0000 (00:05 +0200)]
lightnvm: remove nvm_dev_ops->max_phys_sect
The value of max_phys_sect is always static. Instead of
defining it in the nvm_dev_ops structure, declare it as a global
value.
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Matias Bjørling [Thu, 29 Mar 2018 22:05:03 +0000 (00:05 +0200)]
lightnvm: remove max_rq_size
The field is no longer used.
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Matias Bjørling [Thu, 29 Mar 2018 22:05:02 +0000 (00:05 +0200)]
lightnvm: add 2.0 geometry identification
Implement the geometry data structures for 2.0 and enable a drive
to be identified as one, including exposing the appropriate 2.0
sysfs entries.
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Matias Bjørling [Thu, 29 Mar 2018 22:05:01 +0000 (00:05 +0200)]
lightnvm: flatten nvm_id_group into nvm_id
There are no groups in the 2.0 specification, make sure that the
nvm_id structure is flattened before 2.0 data structures are added.
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Matias Bjørling [Thu, 29 Mar 2018 22:05:00 +0000 (00:05 +0200)]
lightnvm: make 1.2 data structures explicit
Make the 1.2 data structures explicit, so it will be easy to identify
the 2.0 data structures. Also fix the order of which the nvme_nvm_*
are declared, such that they follow the nvme_nvm_command order.
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Javier González [Thu, 29 Mar 2018 22:04:59 +0000 (00:04 +0200)]
lightnvm: pblk: refactor bad block identification
In preparation for the OCSSD 2.0 spec. bad block identification,
refactor the current code to generalize bad block get/set functions and
structures.
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Hans Holmberg [Thu, 29 Mar 2018 22:04:58 +0000 (00:04 +0200)]
lightnvm: pblk: prevent race in pblk_rb_flush_point_set
Make sure that we are not advancing the sync pointer while
we're adding bios to the write buffer entry completion list.
This race condition results in bios not completing and was identified
by a hang when running xfstest generic/113.
Signed-off-by: Hans Holmberg <hans.holmberg@cnexlabs.com>
Reviewed-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Hans Holmberg [Thu, 29 Mar 2018 22:04:57 +0000 (00:04 +0200)]
lightnvm: pblk: allow allocation of new lines during shutdown
When shutting down pblk the write buffer is flushed and if the
current line can't fit the data in the write buffer we need
to allocate a new line, so remove the check that prevents this.
Signed-off-by: Hans Holmberg <hans.holmberg@cnexlabs.com>
Reviewed-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Hans Holmberg [Thu, 29 Mar 2018 22:04:56 +0000 (00:04 +0200)]
lightnvm: pblk: delete writer kick timer before stopping thread
Unless we delete the timer that wakes up the write thread
before we stop the thread we risk re-starting the thread, so
delete the timer first.
Signed-off-by: Hans Holmberg <hans.holmberg@cnexlabs.com>
Reviewed-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Hans Holmberg [Thu, 29 Mar 2018 22:04:55 +0000 (00:04 +0200)]
lightnvm: pblk: add padding distribution sysfs attribute
When pblk receives a sync, all data up to that point in the write buffer
must be comitted to persistent storage, and as flash memory comes with a
minimal write size there is a significant cost involved both in terms
of time for completing the sync and in terms of write amplification
padded sectors for filling up to the minimal write size.
In order to get a better understanding of the costs involved for syncs,
Add a sysfs attribute to pblk: padded_dist, showing a normalized
distribution of sectors padded. In order to facilitate measurements of
specific workloads during the lifetime of the pblk instance, the
distribution can be reset by writing 0 to the attribute.
Do this by introducing counters for each possible padding:
{0..(minimal write size - 1)} and calculate the normalized distribution
when showing the attribute.
Signed-off-by: Hans Holmberg <hans.holmberg@cnexlabs.com>
Signed-off-by: Javier González <javier@cnexlabs.com>
Rearranged total_buckets statement in pblk_sysfs_get_padding_dist
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Matias Bjørling [Thu, 29 Mar 2018 22:04:54 +0000 (00:04 +0200)]
lightnvm: remove multiple groups in 1.2 data structure
Only one id group from the 1.2 specification is supported. Make
sure that only the first group is accessible.
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Matias Bjørling [Thu, 29 Mar 2018 22:04:53 +0000 (00:04 +0200)]
lightnvm: remove mlc pairs structure
The known implementations of the 1.2 specification, and upcoming 2.0
implementation all expose a sequential list of pages to write.
Remove the data structure, as it is no longer needed.
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Hans Holmberg [Thu, 29 Mar 2018 22:04:52 +0000 (00:04 +0200)]
lightnvm: pblk: export write amplification counters to sysfs
In a SSD, write amplification, WA, is defined as the average
number of page writes per user page write. Write amplification
negatively affects write performance and decreases the lifetime
of the disk, so it's a useful metric to add to sysfs.
In plkb's case, the number of writes per user sector is the sum of:
(1) number of user writes
(2) number of sectors written by the garbage collector
(3) number of sectors padded (i.e. due to syncs)
This patch adds persistent counters for 1-3 and two sysfs attributes
to export these along with WA calculated with five decimals:
write_amp_mileage: the accumulated write amplification stats
for the lifetime of the pblk instance
write_amp_trip: resetable stats to facilitate delta measurements,
values reset at creation and if 0 is written
to the attribute.
64-bit counters are used as a 32 bit counter would wrap around
already after about 17 TB worth of user data. It will take a
long long time before the 64 bit sector counters wrap around.
The counters are stored after the bad block bitmap in the first
emeta sector of each written line. There is plenty of space in the
first emeta sector, so we don't need to bump the major version of
the line data format.
Signed-off-by: Hans Holmberg <hans.holmberg@cnexlabs.com>
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Hans Holmberg [Thu, 29 Mar 2018 22:04:51 +0000 (00:04 +0200)]
lightnvm: pblk: check data lines version on recovery
As a preparation for future bumps of data line persistent storage
versions, we need to start checking the emeta line version during
recovery. Also slit up the current emeta/smeta version into two
bytes (major,minor).
Recovering lines with the same major number as the current pblk data
line version must succeed. This means that any changes in the
persistent format must be:
(1) Backward compatible: if we switch back to and older
kernel, recovery of lines stored with major == current_major
and minor > current_minor must succeed.
(2) Forward compatible: switching to a newer kernel,
recovery of lines stored with major=current_major and
minor < minor must handle the data format differences
gracefully(i.e. initialize new data structures to default values).
If we detect lines that have a different major number than
the current we must abort recovery. The user must manually
migrate the data in this case.
Previously the version stored in the emeta header was copied
from smeta, which has version 1, so we need to set the minor
version to 1.
Signed-off-by: Hans Holmberg <hans.holmberg@cnexlabs.com>
Signed-off-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Hans Holmberg [Thu, 29 Mar 2018 22:04:50 +0000 (00:04 +0200)]
lightnvm: pblk: handle bad sectors in the emeta area correctly
Unless we check if there are bad sectors in the entire emeta-area
we risk ending up with valid bitmap / available sector count inconsistency.
This results in lines with a bad chunk at the last LUN marked as bad,
so go through the whole emeta area and mark up the invalid sectors.
Signed-off-by: Hans Holmberg <hans.holmberg@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Matias Bjørling [Thu, 29 Mar 2018 22:04:49 +0000 (00:04 +0200)]
lightnvm: remove chnl_offset in nvme_nvm_identity
The identity structure is initialized to zero in the beginning of
the nvme_nvm_identity function. The chnl_offset is separately set to
zero. Since both the variable and assignment is never changed, remove
them.
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Markus Elfring [Thu, 29 Mar 2018 22:04:48 +0000 (00:04 +0200)]
lightnvm/pblk-gc: Delete an error message for a failed memory allocation in pblk_gc_line_prepare_ws()
Omit an extra message for a memory allocation failure in this function.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Reviewed-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Keith Busch [Tue, 27 Mar 2018 15:39:06 +0000 (09:39 -0600)]
blk-mq: Allow PCI vector offset for mapping queues
The PCI interrupt vectors intended to be associated with a queue may
not start at 0; a driver may allocate pre_vectors for special use. This
patch adds an offset parameter so blk-mq may find the intended affinity
mask and updates all drivers using this API accordingly.
Cc: Don Brace <don.brace@microsemi.com>
Cc: <qla2xxx-upstream@qlogic.com>
Cc: <linux-scsi@vger.kernel.org>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Omar Sandoval [Tue, 27 Mar 2018 04:39:12 +0000 (21:39 -0700)]
loop: use killable lock in ioctls
Even after the previous patch to drop lo_ctl_mutex while calling
vfs_getattr(), there are other cases where we can end up sleeping for a
long time while holding lo_ctl_mutex. Let's avoid the uninterruptible
sleep from the ioctls.
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Omar Sandoval [Tue, 27 Mar 2018 04:39:11 +0000 (21:39 -0700)]
loop: don't call into filesystem while holding lo_ctl_mutex
We hit an issue where a loop device on NFS was stuck in
loop_get_status() doing vfs_getattr() after the NFS server died, which
caused a pile-up of uninterruptible processes waiting on lo_ctl_mutex.
There's no reason to hold this lock while we wait on the filesystem;
let's drop it so that other processes can do their thing. We need to
grab a reference on lo_backing_file while we use it, and we can get rid
of the check on lo_device, which has been unnecessary since commit
a34c0ae9ebd6 ("[PATCH] loop: remove the bio remapping capability") in
the linux-history tree.
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Paolo Valente [Mon, 26 Mar 2018 14:06:24 +0000 (16:06 +0200)]
block, bfq: lower-bound the estimated peak rate to 1
If a storage device handled by BFQ happens to be slower than 7.5 KB/s
for a certain amount of time (in the order of a second), then the
estimated peak rate of the device, maintained in BFQ, becomes equal to
0. The reason is the limited precision with which the rate is
represented (details on the range of representable values in the
comments introduced by this commit). This leads to a division-by-zero
error where the estimated peak rate is used as divisor. Such a type of
failure has been reported in [1].
This commit addresses this issue by:
1. Lower-bounding the estimated peak rate to 1
2. Adding and improving comments on the range of rates representable
[1] https://www.spinics.net/lists/kernel/msg2739205.html
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Signed-off-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Matias Bjørling [Wed, 21 Mar 2018 19:27:07 +0000 (20:27 +0100)]
nvme: make nvme_get_log_ext non-static
Enable the lightnvm integration to use the nvme_get_log_ext()
function.
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Christoph Hellwig [Tue, 20 Mar 2018 19:41:35 +0000 (20:41 +0100)]
nvmet: constify struct nvmet_fabrics_ops
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Christoph Hellwig [Tue, 20 Mar 2018 19:41:34 +0000 (20:41 +0100)]
nvmet: refactor configfs transport type handling
Have a common table of mappings from numerical transport ids to names, and
zero the transport specific area in common code in nvmet_addr_trtype_store.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Max Gurtovoy [Tue, 20 Mar 2018 12:20:41 +0000 (14:20 +0200)]
nvmet: move device_uuid configfs attr definition to suitable place
Signed-off-by: Max Gurtovoy <maxg@mellanox.com>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Nitzan Carmi [Tue, 20 Mar 2018 11:07:30 +0000 (11:07 +0000)]
nvme: Add .stop_ctrl to nvme ctrl ops
For consistancy reasons, any fabric-specific works
(e.g error recovery/reconnect) should be canceled in
nvme_stop_ctrl, as for all other NVMe pending works
(e.g. scan, keep alive).
The patch aims to simplify the logic of the code, as
we now only rely on a vague demand from any fabric
to flush its private workqueues at the beginning of
.delete_ctrl op.
Signed-off-by: Nitzan Carmi <nitzanc@mellanox.com>
Reviewed-by: Max Gurtovoy <maxg@mellanox.com>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Nitzan Carmi [Tue, 20 Mar 2018 11:07:29 +0000 (11:07 +0000)]
nvme-rdma: Allow DELETING state change failure in error_recovery
While error recovery is ongoing, it is OK to move
ctrl to DELETING state (from concurrent delete_work).
Thus we don't need a warning for that case.
Signed-off-by: Nitzan Carmi <nitzanc@mellanox.com>
Reviewed-by: Max Gurtovoy <maxg@mellanox.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Keith Busch [Mon, 19 Mar 2018 16:53:50 +0000 (10:53 -0600)]
nvme: Skip checking heads without namespaces
If a task is holding a reference to a namespace on a removed controller,
the head will not be released. If the same controller is added again
later, its namespaces may not be successfully added. Instead, the user
will see kernel message "Duplicate IDs for nsid <X>".
This patch fixes that by skipping heads that don't have namespaces when
considering if a new namespace is safe to add.
Reported-by: Alex Gagniuc <Alex_Gagniuc@Dellteam.com>
Cc: stable@vger.kernel.org
Signed-off-by: Keith Busch <keith.busch@intel.com>
Reviewed-by: Max Gurtovoy <maxg@mellanox.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Max Gurtovoy [Wed, 28 Feb 2018 11:12:39 +0000 (13:12 +0200)]
nvme-rdma: Don't flush delete_wq by default during remove_one
The .remove_one function is called for any ib_device removal.
In case the removed device has no reference in our driver, there
is no need to flush the work queue.
Reviewed-by: Israel Rukshin <israelr@mellanox.com>
Signed-off-by: Max Gurtovoy <maxg@mellanox.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Max Gurtovoy [Wed, 28 Feb 2018 11:12:38 +0000 (13:12 +0200)]
nvmet-rdma: Don't flush system_wq by default during remove_one
The .remove_one function is called for any ib_device removal.
In case the removed device has no reference in our driver, there
is no need to flush the system work queue.
Reviewed-by: Israel Rukshin <israelr@mellanox.com>
Signed-off-by: Max Gurtovoy <maxg@mellanox.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Israel Rukshin [Wed, 14 Mar 2018 10:22:45 +0000 (10:22 +0000)]
nvmet-rdma: Fix use after free in nvmet_rdma_cm_handler()
We free nvmet rdma queues while handling rdma_cm events.
In order to avoid this we destroy the qp and the queue after destroying
the cm_id which guarantees that all rdma_cm events are done.
Signed-off-by: Israel Rukshin <israelr@mellanox.com>
Reviewed-by: Max Gurtovoy <maxg@mellanox.com>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Israel Rukshin [Wed, 14 Mar 2018 10:22:44 +0000 (10:22 +0000)]
nvmet-rdma: Remove unused queue state
Signed-off-by: Israel Rukshin <israelr@mellanox.com>
Reviewed-by: Max Gurtovoy <maxg@mellanox.com>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
James Smart [Wed, 28 Feb 2018 22:49:11 +0000 (14:49 -0800)]
nvmet_fc: prevent new io rqsts in possible isr completions
When a bio completion calls back into the transport for a
back-end io device, the request completion path can free
the transport io job structure allowing it to be reused for
other operations. The transport has a defer_rcv queue which
holds temporary cmd rcv ops while waitng for io job structures.
when the job frees, if there's a cmd waiting, it is picked up
and submitted for processing, which can call back out to the
bio path if it's a read. Unfortunately, what is unknown is the
context of the original bio done call, and it may be in a state
(softirq) that is not compatible with submitting the new bio in
the same calling sequence. This is especially true when using
scsi back-end devices as scsi is in softirq when it makes the
done call.
Correct by scheduling the io to be started via workq rather
than calling the start new io path inline to the original bio
done path.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
James Smart [Tue, 6 Mar 2018 04:55:49 +0000 (20:55 -0800)]
nvme_fc: on remoteport reuse, set new nport_id and role.
When reattaching to a removed remoteport that has not yet been
fully deleted as it's waiting for reconnect timeouts, be sure to
re-set the ports nport id and role.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
James Smart [Wed, 28 Feb 2018 22:49:10 +0000 (14:49 -0800)]
nvme_fc: fix abort race on teardown with lld reject
Another abort race: An io request is started, becomes active,
and is attempted to be started with the lldd. At the same time
the controller is stopped/torndown and an itterator is run to
abort the ios. As the io is active, it is added to the outstanding
aborted io count. However on the original io request thread, the
driver ends up rejecting the io due to the condition that induced
the controller teardown. The driver reject path didn't check whether
it was in the outstanding io count. This left the count outstanding
stopping controller teardown.
Correct by, in the driver reject case, setting the state to
inactive and checking whether it was in the outstanding io count.
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
James Smart [Mon, 12 Mar 2018 16:32:22 +0000 (09:32 -0700)]
nvme_fc: io timeout should defer abort to ctrl reset
The current nvme_fc code, when an io times out, will abort the io
on the fc link, then call the error recovery routine to reset the
controller. It is during the reset of the controller that the
transport will wait for all ios to be aborted before sending a
Disconnect LS to the target.
However, the reset routine only waits for the io which it generates
the abort for to complete. Any io that was aborted just prior to the
reset isn't in it's list to wait for. Thus the Disconnect is getting
sent before the aborts have completed.
Correct by removing the abort in the timeout handler. The reset will
generate the abort. At that point the timeout handler can be simplified
to request the reset (via the error handler) and restart the timeout
timer.
Also fixes a small typo in a comment in the reset handler.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
James Smart [Tue, 13 Mar 2018 16:48:07 +0000 (09:48 -0700)]
nvme_fc: fix ctrl create failures racing with workq items
If there are errors during initial controller create, the transport
will teardown the partially initialized controller struct and free
the ctlr memory. Trouble is - most of those errors can occur due
to asynchronous events happening such io timeouts and subsystem
connectivity failures. Those failures invoke async workq items to
reset the controller and attempt reconnect. Those may be in progress
as the main thread frees the ctrl memory, resulting in NULL ptr oops.
Prevent this from happening by having the main ctrl failure thread
changing state to DELETING followed by synchronously cancelling any
pending queued work item. The change of state will prevent the
scheduling of resets or reconnect events.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jarosław Janik [Sun, 11 Mar 2018 18:51:56 +0000 (19:51 +0100)]
nvme-pci: disable APST for Samsung NVMe SSD 960 EVO + ASUS PRIME Z370-A
Yet another "incompatible" Samsung NVMe SSD 960 EVO and Asus motherboard
combination. 960 EVO device disappears from PCIe bus within few minutes
after boot-up when APST is in use and never gets back. Forcing
NVME_QUIRK_NO_APST is the only way to make this drive work with this
particular motherboard. NVME_QUIRK_NO_DEEPEST_PS doesn't work, upgrading
motherboard's BIOS didn't help either.
Since this is a desktop motherboard, the only drawback of not using APST
is increased device temperature.
Signed-off-by: Jarosław Janik <jaroslaw.janik@gmail.com>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Max Gurtovoy [Sun, 11 Mar 2018 15:46:06 +0000 (17:46 +0200)]
nvme: centralize ctrl removal prints
nvme_delete_ctrl can be called from various contexts in parallel,
and cause duplicated information prints, even though the specific
context doesn't perform the actual removal. Instead, print the
information when the actual removal occurs.
Signed-off-by: Max Gurtovoy <maxg@mellanox.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Keith Busch [Thu, 8 Mar 2018 21:50:32 +0000 (14:50 -0700)]
nvme-pci: Add .get_address ctrl callback
The nvme-fabrics exports the controller address to sysfs, and we'd
like to have parity with this feature for PCIe. This patch provides
the appropiate callback and returns the controller address as the pci
domain:bus:device.function.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Matias Bjørling [Mon, 26 Feb 2018 12:55:40 +0000 (13:55 +0100)]
nvme: implement log page low/high offset and dwords
NVMe 1.2.1 extends the get log page interface to include 64 bit
offset and increases the number of dwords to 32 bits. Implement
for future use.
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jianchao Wang [Mon, 12 Feb 2018 12:54:46 +0000 (20:54 +0800)]
nvme: change namespaces_mutext to namespaces_rwsem
namespaces_mutext is used to synchronize the operations on ctrl
namespaces list. Most of the time, it is a read operation.
On the other hand, there are many interfaces in nvme core that
need this lock, such as nvme_wait_freeze, and even more interfaces
will be added. If we use mutex here, circular dependency could be
introduced easily. For example:
context A context B
nvme_xxx nvme_xxx
hold namespaces_mutext require namespaces_mutext
sync context B
So it is better to change it from mutex to rwsem.
Reviewed-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jianchao Wang <jianchao.w.wang@oracle.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jianchao Wang [Mon, 12 Feb 2018 12:54:44 +0000 (20:54 +0800)]
nvme: fix the dangerous reference of namespaces list
nvme_remove_namespaces and nvme_remove_invalid_namespaces reference
the ctrl->namespaces list w/o holding namespaces_mutext. It is ok
to invoke nvme_ns_remove there, but what if there is others.
To be safer, reference the ctrl->namespaces list under
namespaces_mutext.
Reviewed-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jianchao Wang <jianchao.w.wang@oracle.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jianchao Wang [Mon, 12 Feb 2018 12:57:24 +0000 (20:57 +0800)]
nvme-pci: quiesce IO queues prior to disabling device HMB accesses
Quiesce IO queues prior to disabling device HMB accesses. A controller
using HMB may relay on it to efficiently complete IO commands.
Reviewed-by: Keith Busch <keith.busch@intel.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jianchao Wang <jianchao.w.wang@oracle.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Thomas Tai [Thu, 8 Feb 2018 18:38:30 +0000 (13:38 -0500)]
Documentation: nvme: Documentation for nvme fault injection
Add examples to show how to use nvme fault injection.
Signed-off-by: Thomas Tai <thomas.tai@oracle.com>
Reviewed-by: Eric Saint-Etienne <eric.saint.etienne@oracle.com>
Signed-off-by: Karl Volz <karl.volz@oracle.com>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Thomas Tai [Thu, 8 Feb 2018 18:38:29 +0000 (13:38 -0500)]
nvme: Add fault injection feature
Linux's fault injection framework provides a systematic way to support
error injection via debugfs in the /sys/kernel/debug directory. This
patch uses the framework to add error injection to NVMe driver. The
fault injection source code is stored in a separate file and only linked
if CONFIG_FAULT_INJECTION_DEBUG_FS kernel config is selected.
Once the error injection is enabled, NVME_SC_INVALID_OPCODE with no
retry will be injected into the nvme_end_request. Users can change
the default status code and no retry flag via debufs. Following example
shows how to enable and inject an error. For more examples, refer to
Documentation/fault-injection/nvme-fault-injection.txt
How to enable nvme fault injection:
First, enable CONFIG_FAULT_INJECTION_DEBUG_FS kernel config,
recompile the kernel. After booting up the kernel, do the
following.
How to inject an error:
mount /dev/nvme0n1 /mnt
echo 1 > /sys/kernel/debug/nvme0n1/fault_inject/times
echo 100 > /sys/kernel/debug/nvme0n1/fault_inject/probability
cp a.file /mnt
Expected Result:
cp: cannot stat ‘/mnt/a.file’: Input/output error
Message from dmesg:
FAULT_INJECTION: forcing a failure.
name fault_inject, interval 1, probability 100, space 0, times 1
CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.15.0-rc8+ #2
Hardware name: innotek GmbH VirtualBox/VirtualBox,
BIOS VirtualBox 12/01/2006
Call Trace:
<IRQ>
dump_stack+0x5c/0x7d
should_fail+0x148/0x170
nvme_should_fail+0x2f/0x50 [nvme_core]
nvme_process_cq+0xe7/0x1d0 [nvme]
nvme_irq+0x1e/0x40 [nvme]
__handle_irq_event_percpu+0x3a/0x190
handle_irq_event_percpu+0x30/0x70
handle_irq_event+0x36/0x60
handle_fasteoi_irq+0x78/0x120
handle_irq+0xa7/0x130
? tick_irq_enter+0xa8/0xc0
do_IRQ+0x43/0xc0
common_interrupt+0xa2/0xa2
</IRQ>
RIP: 0010:native_safe_halt+0x2/0x10
RSP: 0018:
ffffffff82003e90 EFLAGS:
00000246 ORIG_RAX:
ffffffffffffffdd
RAX:
ffffffff817a10c0 RBX:
ffffffff82012480 RCX:
0000000000000000
RDX:
0000000000000000 RSI:
0000000000000000 RDI:
0000000000000000
RBP:
0000000000000000 R08:
000000008e38ce64 R09:
0000000000000000
R10:
0000000000000000 R11:
0000000000000000 R12:
ffffffff82012480
R13:
ffffffff82012480 R14:
0000000000000000 R15:
0000000000000000
? __sched_text_end+0x4/0x4
default_idle+0x18/0xf0
do_idle+0x150/0x1d0
cpu_startup_entry+0x6f/0x80
start_kernel+0x4c4/0x4e4
? set_init_arg+0x55/0x55
secondary_startup_64+0xa5/0xb0
print_req_error: I/O error, dev nvme0n1, sector 9240
EXT4-fs error (device nvme0n1): ext4_find_entry:1436:
inode #2: comm cp: reading directory lblock 0
Signed-off-by: Thomas Tai <thomas.tai@oracle.com>
Reviewed-by: Eric Saint-Etienne <eric.saint.etienne@oracle.com>
Signed-off-by: Karl Volz <karl.volz@oracle.com>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Minwoo Im [Thu, 8 Feb 2018 13:56:31 +0000 (22:56 +0900)]
nvme: use define instead of magic value for identify size
NVME_IDENTIFY_DATA_SIZE was added to linux/nvme.h by following commit.
commit
0add5e8e588c ("nvmet: use NVME_IDENTIFY_DATA_SIZE")
Make it use NVME_IDENTIFY_DATA_SIZE define instead of magic value
0x1000 in case of identify data size.
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Minwoo Im <minwoo.im.dev@gmail.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Sagi Grimberg [Thu, 25 Jan 2018 11:56:46 +0000 (13:56 +0200)]
iscsi-target: use common inet_addr_is_any
Instead of open-coding it.
Reviewed-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Cc: "Nicholas A. Bellinger" <nab@linux-iscsi.org>
Cc: target-devel@vger.kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Sagi Grimberg [Wed, 24 Jan 2018 18:27:10 +0000 (20:27 +0200)]
nvmet: don't return "any" ip address in discovery log page
Its perfectly valid to assign a nvmet port to listen on "any"
IP address (traddr 0.0.0.0 for ipv4 address family) for IP based
transport ports. However, we must not return this address in
discovery log entries. Instead we need to return the address
where the request was accepted on (req->port address).
Since this is nvme transport specific, introduce an optional
.disc_traddr interface that is designed to check that a
port in question is bound to "any" IP address and if so, set
the traddr from the port where the request came from.
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Sagi Grimberg [Wed, 24 Jan 2018 18:24:24 +0000 (20:24 +0200)]
net/utils: Introduce inet_addr_is_any
Can be useful to check INET_ANY address for both ipv4/ipv6 addresses.
Reviewed-by: Bart Van Assche <bart.vanassche@wdc.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Shawn Lin [Thu, 22 Mar 2018 10:56:16 +0000 (18:56 +0800)]
mmc: block: Delete gendisk before cleaning up the request queue
dd if=/dev/urandom of=/dev/mmcblk1 bs=4k count=10000
with a SD card hotplug during transfer reports a warning below
introduced by commit
a063057d7c73 ("block: Fix a race between
request queue removal and the block cgroup controller"). So we
should now remove the disk, partition and bdi sysfs attributes
before cleaning up the request queue associated with the disk.
[ 410.331226] mmc1: card 59b4 removed
[ 410.348583] WARNING: CPU: 0 PID: 5 at block/blk-core.c:785
blk_cleanup_queue+0x138/0x140
[ 410.349294] Modules linked in:
[ 410.349570] CPU: 0 PID: 5 Comm: kworker/0:0 Not tainted
4.16.0-rc6-next-20180321-00004-gc2ad6a7 #263
[ 410.350363] Hardware name: Excavator-RK3399 Board (DT)
[ 410.350819] Workqueue: events_freezable mmc_rescan
[ 410.351242] pstate:
60000005 (nZCv daif -PAN -UAO)
[ 410.351663] pc : blk_cleanup_queue+0x138/0x140
[ 410.352054] lr : blk_cleanup_queue+0xac/0x140
[ 410.352436] sp :
ffff0000092cbb90
[ 410.352727] x29:
ffff0000092cbb90 x28:
0000000000000000
[ 410.353195] x27:
ffff8000f6f23030 x26:
ffff00000904e610
[ 410.353662] x25:
ffff8000f17cc808 x24:
ffff8000f1038200
[ 410.354128] x23:
0000000000000060 x22:
0000000000000000
[ 410.354595] x21:
ffff8000f11748d8 x20:
ffff8000f1038200
[ 410.355061] x19:
ffff8000f1174200 x18:
0000ffff936347d8
[ 410.355528] x17:
0000ffff935b93c0 x16:
ffff0000081263f8
[ 410.355994] x15:
0000000000000000 x14:
0000000000000400
[ 410.356461] x13:
0000000000000001 x12:
0000000000000001
[ 410.356927] x11:
0000000000000040 x10:
ffff8000f2400028
[ 410.357393] x9 :
ffff8000f2400040 x8 :
0000000000000000
[ 410.357860] x7 :
ffff8000f6f3a340 x6 :
ffff8000f6f3a340
[ 410.358326] x5 :
ffff8000f2400000 x4 :
ffff8000f6f3a340
[ 410.358792] x3 :
0000000000000000 x2 :
39c1333e45670800
[ 410.359259] x1 :
0000000000000000 x0 :
0000000000000003
[ 410.359726] Call trace:
[ 410.359943] blk_cleanup_queue+0x138/0x140
[ 410.360305] mmc_cleanup_queue+0x2c/0x48
[ 410.360652] mmc_blk_remove_req+0x1c/0x98
[ 410.361005] mmc_blk_remove+0x180/0x1c0
[ 410.361343] mmc_bus_remove+0x1c/0x28
[ 410.361670] device_release_driver_internal+0x154/0x1f0
[ 410.362128] device_release_driver+0x14/0x20
[ 410.362504] bus_remove_device+0xc8/0x108
[ 410.362858] device_del+0x120/0x350
[ 410.363167] mmc_remove_card+0x5c/0xb8
[ 410.363498] mmc_sd_detect+0x40/0x78
[ 410.363813] mmc_rescan+0x19c/0x368
[ 410.364123] process_one_work+0x1ac/0x318
[ 410.364477] worker_thread+0x50/0x450
[ 410.364801] kthread+0xf8/0x128
[ 410.365081] ret_from_fork+0x10/0x18
[ 410.365395] ---[ end trace
268e87a46c28968c ]---
Reviewed-by: Bart Van Assche <bart.vanassche@wdc.com>
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Mikulas Patocka [Wed, 21 Mar 2018 16:49:29 +0000 (12:49 -0400)]
Fix slab name "biovec-(1<<(21-12))"
I'm getting a slab named "biovec-(1<<(21-12))". It is caused by unintended
expansion of the macro BIO_MAX_PAGES. This patch renames it to biovec-max.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Cc: stable@vger.kernel.org # v4.14+
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Mikulas Patocka [Wed, 21 Mar 2018 16:42:25 +0000 (12:42 -0400)]
block: use 32-bit blk_status_t on Alpha
Early alpha processors cannot write a single byte or word; they read 8
bytes, modify the value in registers and write back 8 bytes.
The type blk_status_t is defined as one byte, it is often written
asynchronously by I/O completion routines, this asynchronous modification
can corrupt content of nearby bytes if these nearby bytes can be written
simultaneously by another CPU.
- one example of such corruption is the structure dm_io where
"blk_status_t status" is written by an asynchronous completion routine
and "atomic_t io_count" is modified synchronously
- another example is the structure dm_buffer where "unsigned hold_count"
is modified synchronously from process context and "blk_status_t
write_error" is modified asynchronously from bio completion routine
This patch fixes the bug by changing the type blk_status_t to 32 bits if
we are on Alpha and if we are compiling for a processor that doesn't have
the byte-word-extension.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Cc: stable@vger.kernel.org # 4.13+
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bart Van Assche [Mon, 19 Mar 2018 18:46:13 +0000 (11:46 -0700)]
block: Change a rcu_read_{lock,unlock}_sched() pair into rcu_read_{lock,unlock}()
scsi_device_quiesce() uses synchronize_rcu() to guarantee that the
effect of blk_set_preempt_only() will be visible for percpu_ref_tryget()
calls that occur after the queue unfreeze by using the approach
explained in https://lwn.net/Articles/573497/. The rcu read lock and
unlock calls in blk_queue_enter() form a pair with the synchronize_rcu()
call in scsi_device_quiesce(). Both scsi_device_quiesce() and
blk_queue_enter() must either use regular RCU or RCU-sched.
Since neither the RCU-protected code in blk_queue_enter() nor
blk_queue_usage_counter_release() sleeps, regular RCU protection
is sufficient. Note: scsi_device_quiesce() does not have to be
modified since it already uses synchronize_rcu().
Reported-by: Tejun Heo <tj@kernel.org>
Fixes: 3a0a529971ec ("block, scsi: Make SCSI quiesce and resume work reliably")
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Hannes Reinecke <hare@suse.com>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Johannes Thumshirn <jthumshirn@suse.de>
Cc: Oleksandr Natalenko <oleksandr@natalenko.name>
Cc: Martin Steigerwald <martin@lichtvoll.de>
Cc: stable@vger.kernel.org # v4.15
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bart Van Assche [Mon, 19 Mar 2018 00:36:33 +0000 (17:36 -0700)]
bcache: Fix a compiler warning in bcache_device_init()
Avoid that building with W=1 triggers the following compiler warning:
drivers/md/bcache/super.c:776:20: warning: comparison is always false due to limited range of data type [-Wtype-limits]
d->nr_stripes > SIZE_MAX / sizeof(atomic_t)) {
^
Reviewed-by: Coly Li <colyli@suse.de>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bart Van Assche [Mon, 19 Mar 2018 00:36:32 +0000 (17:36 -0700)]
bcache: Reduce the number of sparse complaints about lock imbalances
Add more annotations for sparse to inform it about which functions do
not have the same number of spin_lock() and spin_unlock() calls.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bart Van Assche [Mon, 19 Mar 2018 00:36:31 +0000 (17:36 -0700)]
bcache: Suppress more warnings about set-but-not-used variables
This patch does not change any functionality.
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Reviewed-by: Coly Li <colyli@suse.de>
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bart Van Assche [Mon, 19 Mar 2018 00:36:30 +0000 (17:36 -0700)]
bcache: Remove an unused variable
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bart Van Assche [Mon, 19 Mar 2018 00:36:29 +0000 (17:36 -0700)]
bcache: Fix kernel-doc warnings
Avoid that building with W=1 triggers warnings about the kernel-doc
headers.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bart Van Assche [Mon, 19 Mar 2018 00:36:28 +0000 (17:36 -0700)]
bcache: Annotate switch fall-through
This patch avoids that building with W=1 triggers complaints about
switch fall-throughs.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bart Van Assche [Mon, 19 Mar 2018 00:36:27 +0000 (17:36 -0700)]
bcache: Add __printf annotation to __bch_check_keys()
Make it possible for the compiler to verify the consistency of the
format string passed to __bch_check_keys() and the arguments that
should be formatted according to that format string.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bart Van Assche [Mon, 19 Mar 2018 00:36:26 +0000 (17:36 -0700)]
bcache: Fix indentation
This patch avoids that smatch complains about inconsistent indentation.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Reviewed-by: Coly Li <colyli@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Coly Li [Mon, 19 Mar 2018 00:36:25 +0000 (17:36 -0700)]
bcache: add io_disable to struct cached_dev
If a bcache device is configured to writeback mode, current code does not
handle write I/O errors on backing devices properly.
In writeback mode, write request is written to cache device, and
latter being flushed to backing device. If I/O failed when writing from
cache device to the backing device, bcache code just ignores the error and
upper layer code is NOT noticed that the backing device is broken.
This patch tries to handle backing device failure like how the cache device
failure is handled,
- Add a error counter 'io_errors' and error limit 'error_limit' in struct
cached_dev. Add another io_disable to struct cached_dev to disable I/Os
on the problematic backing device.
- When I/O error happens on backing device, increase io_errors counter. And
if io_errors reaches error_limit, set cache_dev->io_disable to true, and
stop the bcache device.
The result is, if backing device is broken of disconnected, and I/O errors
reach its error limit, backing device will be disabled and the associated
bcache device will be removed from system.
Changelog:
v2: remove "bcache: " prefix in pr_error(), and use correct name string to
print out bcache device gendisk name.
v1: indeed this is new added in v2 patch set.
Signed-off-by: Coly Li <colyli@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Cc: Michael Lyle <mlyle@lyle.org>
Cc: Junhui Tang <tang.junhui@zte.com.cn>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Coly Li [Mon, 19 Mar 2018 00:36:24 +0000 (17:36 -0700)]
bcache: add backing_request_endio() for bi_end_io
In order to catch I/O error of backing device, a separate bi_end_io
call back is required. Then a per backing device counter can record I/O
errors number and retire the backing device if the counter reaches a
per backing device I/O error limit.
This patch adds backing_request_endio() to bcache backing device I/O code
path, this is a preparation for further complicated backing device failure
handling. So far there is no real code logic change, I make this change a
separate patch to make sure it is stable and reliable for further work.
Changelog:
v2: Fix code comments typo, remove a redundant bch_writeback_add() line
added in v4 patch set.
v1: indeed this is new added in this patch set.
[mlyle: truncated commit subject]
Signed-off-by: Coly Li <colyli@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Cc: Junhui Tang <tang.junhui@zte.com.cn>
Cc: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Chengguang Xu [Mon, 19 Mar 2018 00:36:23 +0000 (17:36 -0700)]
bcache: move closure debug file into debug directory
In current code closure debug file is outside of debug directory
and when unloading module there is lack of removing operation
for closure debug file, so it will cause creating error when trying
to reload module.
This patch move closure debug file into "bcache" debug direcory
so that the file can get deleted properly.
Signed-off-by: Chengguang Xu <cgxu519@gmx.com>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Reviewed-by: Tang Junhui <tang.junhui@zte.com.cn>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Tang Junhui [Mon, 19 Mar 2018 00:36:22 +0000 (17:36 -0700)]
bcache: fix using of loop variable in memory shrink
In bch_mca_scan(), There are some confusion and logical error in the use of
loop variables. In this patch, we clarify them as:
1) nr: the number of btree nodes needs to scan, which will decrease after
we scan a btree node, and should not be less than 0;
2) i: the number of btree nodes have scanned, includes both
btree_cache_freeable and btree_cache, which should not be bigger than
btree_cache_used;
3) freed: the number of btree nodes have freed.
Signed-off-by: Tang Junhui <tang.junhui@zte.com.cn>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Tang Junhui [Mon, 19 Mar 2018 00:36:21 +0000 (17:36 -0700)]
bcache: fix error return value in memory shrink
In bch_mca_scan(), the return value should not be the number of freed btree
nodes, but the number of pages of freed btree nodes.
Signed-off-by: Tang Junhui <tang.junhui@zte.com.cn>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Tang Junhui [Mon, 19 Mar 2018 00:36:20 +0000 (17:36 -0700)]
bcache: fix incorrect sysfs output value of strip size
Stripe size is shown as zero when no strip in back end device:
[root@ceph132 ~]# cat /sys/block/sdd/bcache/stripe_size
0.0k
Actually it should be 1T Bytes (1 << 31 sectors), but in sysfs
interface, stripe_size was changed from sectors to bytes, and move
9 bits left, so the 32 bits variable overflows.
This patch change the variable to a 64 bits type before moving bits.
Signed-off-by: Tang Junhui <tang.junhui@zte.com.cn>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Tang Junhui [Mon, 19 Mar 2018 00:36:19 +0000 (17:36 -0700)]
bcache: fix inaccurate io state for detached bcache devices
When we run IO in a detached device, and run iostat to shows IO status,
normally it will show like bellow (Omitted some fields):
Device: ... avgrq-sz avgqu-sz await r_await w_await svctm %util
sdd ... 15.89 0.53 1.82 0.20 2.23 1.81 52.30
bcache0 ... 15.89 115.42 0.00 0.00 0.00 2.40 69.60
but after IO stopped, there are still very big avgqu-sz and %util
values as bellow:
Device: ... avgrq-sz avgqu-sz await r_await w_await svctm %util
bcache0 ... 0 5326.32 0.00 0.00 0.00 0.00 100.10
The reason for this issue is that, only generic_start_io_acct() called
and no generic_end_io_acct() called for detached device in
cached_dev_make_request(). See the code:
//start generic_start_io_acct()
generic_start_io_acct(q, rw, bio_sectors(bio), &d->disk->part0);
if (cached_dev_get(dc)) {
//will callback generic_end_io_acct()
}
else {
//will not call generic_end_io_acct()
}
This patch calls generic_end_io_acct() in the end of IO for detached
devices, so we can show IO state correctly.
(Modified to use GFP_NOIO in kzalloc() by Coly Li)
Changelog:
v2: fix typo.
v1: the initial version.
Signed-off-by: Tang Junhui <tang.junhui@zte.com.cn>
Reviewed-by: Coly Li <colyli@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Coly Li [Mon, 19 Mar 2018 00:36:18 +0000 (17:36 -0700)]
bcache: add stop_when_cache_set_failed option to backing device
When there are too many I/O errors on cache device, current bcache code
will retire the whole cache set, and detach all bcache devices. But the
detached bcache devices are not stopped, which is problematic when bcache
is in writeback mode.
If the retired cache set has dirty data of backing devices, continue
writing to bcache device will write to backing device directly. If the
LBA of write request has a dirty version cached on cache device, next time
when the cache device is re-registered and backing device re-attached to
it again, the stale dirty data on cache device will be written to backing
device, and overwrite latest directly written data. This situation causes
a quite data corruption.
But we cannot simply stop all attached bcache devices when the cache set is
broken or disconnected. For example, use bcache to accelerate performance
of an email service. In such workload, if cache device is broken but no
dirty data lost, keep the bcache device alive and permit email service
continue to access user data might be a better solution for the cache
device failure.
Nix <nix@esperi.org.uk> points out the issue and provides the above example
to explain why it might be necessary to not stop bcache device for broken
cache device. Pavel Goran <via-bcache@pvgoran.name> provides a brilliant
suggestion to provide "always" and "auto" options to per-cached device
sysfs file stop_when_cache_set_failed. If cache set is retiring and the
backing device has no dirty data on cache, it should be safe to keep the
bcache device alive. In this case, if stop_when_cache_set_failed is set to
"auto", the device failure handling code will not stop this bcache device
and permit application to access the backing device with a unattached
bcache device.
Changelog:
[mlyle: edited to not break string constants across lines]
v3: fix typos pointed out by Nix.
v2: change option values of stop_when_cache_set_failed from 1/0 to
"auto"/"always".
v1: initial version, stop_when_cache_set_failed can be 0 (not stop) or 1
(always stop).
Signed-off-by: Coly Li <colyli@suse.de>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Michael Lyle <mlyle@lyle.org>
Cc: Nix <nix@esperi.org.uk>
Cc: Pavel Goran <via-bcache@pvgoran.name>
Cc: Junhui Tang <tang.junhui@zte.com.cn>
Cc: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Coly Li [Mon, 19 Mar 2018 00:36:17 +0000 (17:36 -0700)]
bcache: add CACHE_SET_IO_DISABLE to struct cache_set flags
When too many I/Os failed on cache device, bch_cache_set_error() is called
in the error handling code path to retire whole problematic cache set. If
new I/O requests continue to come and take refcount dc->count, the cache
set won't be retired immediately, this is a problem.
Further more, there are several kernel thread and self-armed kernel work
may still running after bch_cache_set_error() is called. It needs to wait
quite a while for them to stop, or they won't stop at all. They also
prevent the cache set from being retired.
The solution in this patch is, to add per cache set flag to disable I/O
request on this cache and all attached backing devices. Then new coming I/O
requests can be rejected in *_make_request() before taking refcount, kernel
threads and self-armed kernel worker can stop very fast when flags bit
CACHE_SET_IO_DISABLE is set.
Because bcache also do internal I/Os for writeback, garbage collection,
bucket allocation, journaling, this kind of I/O should be disabled after
bch_cache_set_error() is called. So closure_bio_submit() is modified to
check whether CACHE_SET_IO_DISABLE is set on cache_set->flags. If set,
closure_bio_submit() will set bio->bi_status to BLK_STS_IOERR and
return, generic_make_request() won't be called.
A sysfs interface is also added to set or clear CACHE_SET_IO_DISABLE bit
from cache_set->flags, to disable or enable cache set I/O for debugging. It
is helpful to trigger more corner case issues for failed cache device.
Changelog
v4, add wait_for_kthread_stop(), and call it before exits writeback and gc
kernel threads.
v3, change CACHE_SET_IO_DISABLE from 4 to 3, since it is bit index.
remove "bcache: " prefix when printing out kernel message.
v2, more changes by previous review,
- Use CACHE_SET_IO_DISABLE of cache_set->flags, suggested by Junhui.
- Check CACHE_SET_IO_DISABLE in bch_btree_gc() to stop a while-loop, this
is reported and inspired from origal patch of Pavel Vazharov.
v1, initial version.
Signed-off-by: Coly Li <colyli@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Cc: Junhui Tang <tang.junhui@zte.com.cn>
Cc: Michael Lyle <mlyle@lyle.org>
Cc: Pavel Vazharov <freakpv@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Coly Li [Mon, 19 Mar 2018 00:36:16 +0000 (17:36 -0700)]
bcache: stop dc->writeback_rate_update properly
struct delayed_work writeback_rate_update in struct cache_dev is a delayed
worker to call function update_writeback_rate() in period (the interval is
defined by dc->writeback_rate_update_seconds).
When a metadate I/O error happens on cache device, bcache error handling
routine bch_cache_set_error() will call bch_cache_set_unregister() to
retire whole cache set. On the unregister code path, this delayed work is
stopped by calling cancel_delayed_work_sync(&dc->writeback_rate_update).
dc->writeback_rate_update is a special delayed work from others in bcache.
In its routine update_writeback_rate(), this delayed work is re-armed
itself. That means when cancel_delayed_work_sync() returns, this delayed
work can still be executed after several seconds defined by
dc->writeback_rate_update_seconds.
The problem is, after cancel_delayed_work_sync() returns, the cache set
unregister code path will continue and release memory of struct cache set.
Then the delayed work is scheduled to run, __update_writeback_rate()
will reference the already released cache_set memory, and trigger a NULL
pointer deference fault.
This patch introduces two more bcache device flags,
- BCACHE_DEV_WB_RUNNING
bit set: bcache device is in writeback mode and running, it is OK for
dc->writeback_rate_update to re-arm itself.
bit clear:bcache device is trying to stop dc->writeback_rate_update,
this delayed work should not re-arm itself and quit.
- BCACHE_DEV_RATE_DW_RUNNING
bit set: routine update_writeback_rate() is executing.
bit clear: routine update_writeback_rate() quits.
This patch also adds a function cancel_writeback_rate_update_dwork() to
wait for dc->writeback_rate_update quits before cancel it by calling
cancel_delayed_work_sync(). In order to avoid a deadlock by unexpected
quit dc->writeback_rate_update, after time_out seconds this function will
give up and continue to call cancel_delayed_work_sync().
And here I explain how this patch stops self re-armed delayed work properly
with the above stuffs.
update_writeback_rate() sets BCACHE_DEV_RATE_DW_RUNNING at its beginning
and clears BCACHE_DEV_RATE_DW_RUNNING at its end. Before calling
cancel_writeback_rate_update_dwork() clear flag BCACHE_DEV_WB_RUNNING.
Before calling cancel_delayed_work_sync() wait utill flag
BCACHE_DEV_RATE_DW_RUNNING is clear. So when calling
cancel_delayed_work_sync(), dc->writeback_rate_update must be already re-
armed, or quite by seeing BCACHE_DEV_WB_RUNNING cleared. In both cases
delayed work routine update_writeback_rate() won't be executed after
cancel_delayed_work_sync() returns.
Inside update_writeback_rate() before calling schedule_delayed_work(), flag
BCACHE_DEV_WB_RUNNING is checked before. If this flag is cleared, it means
someone is about to stop the delayed work. Because flag
BCACHE_DEV_RATE_DW_RUNNING is set already and cancel_delayed_work_sync()
has to wait for this flag to be cleared, we don't need to worry about race
condition here.
If update_writeback_rate() is scheduled to run after checking
BCACHE_DEV_RATE_DW_RUNNING and before calling cancel_delayed_work_sync()
in cancel_writeback_rate_update_dwork(), it is also safe. Because at this
moment BCACHE_DEV_WB_RUNNING is cleared with memory barrier. As I mentioned
previously, update_writeback_rate() will see BCACHE_DEV_WB_RUNNING is clear
and quit immediately.
Because there are more dependences inside update_writeback_rate() to struct
cache_set memory, dc->writeback_rate_update is not a simple self re-arm
delayed work. After trying many different methods (e.g. hold dc->count, or
use locks), this is the only way I can find which works to properly stop
dc->writeback_rate_update delayed work.
Changelog:
v3: change values of BCACHE_DEV_WB_RUNNING and BCACHE_DEV_RATE_DW_RUNNING
to bit index, for test_bit().
v2: Try to fix the race issue which is pointed out by Junhui.
v1: The initial version for review
Signed-off-by: Coly Li <colyli@suse.de>
Reviewed-by: Junhui Tang <tang.junhui@zte.com.cn>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Cc: Michael Lyle <mlyle@lyle.org>
Cc: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Coly Li [Mon, 19 Mar 2018 00:36:15 +0000 (17:36 -0700)]
bcache: quit dc->writeback_thread when BCACHE_DEV_DETACHING is set
In patch "bcache: fix cached_dev->count usage for bch_cache_set_error()",
cached_dev_get() is called when creating dc->writeback_thread, and
cached_dev_put() is called when exiting dc->writeback_thread. This
modification works well unless people detach the bcache device manually by
'echo 1 > /sys/block/bcache<N>/bcache/detach'
Because this sysfs interface only calls bch_cached_dev_detach() which wakes
up dc->writeback_thread but does not stop it. The reason is, before patch
"bcache: fix cached_dev->count usage for bch_cache_set_error()", inside
bch_writeback_thread(), if cache is not dirty after writeback,
cached_dev_put() will be called here. And in cached_dev_make_request() when
a new write request makes cache from clean to dirty, cached_dev_get() will
be called there. Since we don't operate dc->count in these locations,
refcount d->count cannot be dropped after cache becomes clean, and
cached_dev_detach_finish() won't be called to detach bcache device.
This patch fixes the issue by checking whether BCACHE_DEV_DETACHING is
set inside bch_writeback_thread(). If this bit is set and cache is clean
(no existing writeback_keys), break the while-loop, call cached_dev_put()
and quit the writeback thread.
Please note if cache is still dirty, even BCACHE_DEV_DETACHING is set the
writeback thread should continue to perform writeback, this is the original
design of manually detach.
It is safe to do the following check without locking, let me explain why,
+ if (!test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags) &&
+ (!atomic_read(&dc->has_dirty) || !dc->writeback_running)) {
If the kenrel thread does not sleep and continue to run due to conditions
are not updated in time on the running CPU core, it just consumes more CPU
cycles and has no hurt. This should-sleep-but-run is safe here. We just
focus on the should-run-but-sleep condition, which means the writeback
thread goes to sleep in mistake while it should continue to run.
1, First of all, no matter the writeback thread is hung or not,
kthread_stop() from cached_dev_detach_finish() will wake up it and
terminate by making kthread_should_stop() return true. And in normal
run time, bit on index BCACHE_DEV_DETACHING is always cleared, the
condition
!test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags)
is always true and can be ignored as constant value.
2, If one of the following conditions is true, the writeback thread should
go to sleep,
"!atomic_read(&dc->has_dirty)" or "!dc->writeback_running)"
each of them independently controls the writeback thread should sleep or
not, let's analyse them one by one.
2.1 condition "!atomic_read(&dc->has_dirty)"
If dc->has_dirty is set from 0 to 1 on another CPU core, bcache will
call bch_writeback_queue() immediately or call bch_writeback_add() which
indirectly calls bch_writeback_queue() too. In bch_writeback_queue(),
wake_up_process(dc->writeback_thread) is called. It sets writeback
thread's task state to TASK_RUNNING and following an implicit memory
barrier, then tries to wake up the writeback thread.
In writeback thread, its task state is set to TASK_INTERRUPTIBLE before
doing the condition check. If other CPU core sets the TASK_RUNNING state
after writeback thread setting TASK_INTERRUPTIBLE, the writeback thread
will be scheduled to run very soon because its state is not
TASK_INTERRUPTIBLE. If other CPU core sets the TASK_RUNNING state before
writeback thread setting TASK_INTERRUPTIBLE, the implict memory barrier
of wake_up_process() will make sure modification of dc->has_dirty on
other CPU core is updated and observed on the CPU core of writeback
thread. Therefore the condition check will correctly be false, and
continue writeback code without sleeping.
2.2 condition "!dc->writeback_running)"
dc->writeback_running can be changed via sysfs file, every time it is
modified, a following bch_writeback_queue() is alwasy called. So the
change is always observed on the CPU core of writeback thread. If
dc->writeback_running is changed from 0 to 1 on other CPU core, this
condition check will observe the modification and allow writeback
thread to continue to run without sleeping.
Now we can see, even without a locking protection, multiple conditions
check is safe here, no deadlock or process hang up will happen.
I compose a separte patch because that patch "bcache: fix cached_dev->count
usage for bch_cache_set_error()" already gets a "Reviewed-by:" from Hannes
Reinecke. Also this fix is not trivial and good for a separate patch.
Signed-off-by: Coly Li <colyli@suse.de>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Cc: Hannes Reinecke <hare@suse.com>
Cc: Huijun Tang <tang.junhui@zte.com.cn>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Coly Li [Mon, 19 Mar 2018 00:36:14 +0000 (17:36 -0700)]
bcache: fix cached_dev->count usage for bch_cache_set_error()
When bcache metadata I/O fails, bcache will call bch_cache_set_error()
to retire the whole cache set. The expected behavior to retire a cache
set is to unregister the cache set, and unregister all backing device
attached to this cache set, then remove sysfs entries of the cache set
and all attached backing devices, finally release memory of structs
cache_set, cache, cached_dev and bcache_device.
In my testing when journal I/O failure triggered by disconnected cache
device, sometimes the cache set cannot be retired, and its sysfs
entry /sys/fs/bcache/<uuid> still exits and the backing device also
references it. This is not expected behavior.
When metadata I/O failes, the call senquence to retire whole cache set is,
bch_cache_set_error()
bch_cache_set_unregister()
bch_cache_set_stop()
__cache_set_unregister() <- called as callback by calling
clousre_queue(&c->caching)
cache_set_flush() <- called as a callback when refcount
of cache_set->caching is 0
cache_set_free() <- called as a callback when refcount
of catch_set->cl is 0
bch_cache_set_release() <- called as a callback when refcount
of catch_set->kobj is 0
I find if kernel thread bch_writeback_thread() quits while-loop when
kthread_should_stop() is true and searched_full_index is false, clousre
callback cache_set_flush() set by continue_at() will never be called. The
result is, bcache fails to retire whole cache set.
cache_set_flush() will be called when refcount of closure c->caching is 0,
and in function bcache_device_detach() refcount of closure c->caching is
released to 0 by clousre_put(). In metadata error code path, function
bcache_device_detach() is called by cached_dev_detach_finish(). This is a
callback routine being called when cached_dev->count is 0. This refcount
is decreased by cached_dev_put().
The above dependence indicates, cache_set_flush() will be called when
refcount of cache_set->cl is 0, and refcount of cache_set->cl to be 0
when refcount of cache_dev->count is 0.
The reason why sometimes cache_dev->count is not 0 (when metadata I/O fails
and bch_cache_set_error() called) is, in bch_writeback_thread(), refcount
of cache_dev is not decreased properly.
In bch_writeback_thread(), cached_dev_put() is called only when
searched_full_index is true and cached_dev->writeback_keys is empty, a.k.a
there is no dirty data on cache. In most of run time it is correct, but
when bch_writeback_thread() quits the while-loop while cache is still
dirty, current code forget to call cached_dev_put() before this kernel
thread exits. This is why sometimes cache_set_flush() is not executed and
cache set fails to be retired.
The reason to call cached_dev_put() in bch_writeback_rate() is, when the
cache device changes from clean to dirty, cached_dev_get() is called, to
make sure during writeback operatiions both backing and cache devices
won't be released.
Adding following code in bch_writeback_thread() does not work,
static int bch_writeback_thread(void *arg)
}
+ if (atomic_read(&dc->has_dirty))
+ cached_dev_put()
+
return 0;
}
because writeback kernel thread can be waken up and start via sysfs entry:
echo 1 > /sys/block/bcache<N>/bcache/writeback_running
It is difficult to check whether backing device is dirty without race and
extra lock. So the above modification will introduce potential refcount
underflow in some conditions.
The correct fix is, to take cached dev refcount when creating the kernel
thread, and put it before the kernel thread exits. Then bcache does not
need to take a cached dev refcount when cache turns from clean to dirty,
or to put a cached dev refcount when cache turns from ditry to clean. The
writeback kernel thread is alwasy safe to reference data structure from
cache set, cache and cached device (because a refcount of cache device is
taken for it already), and no matter the kernel thread is stopped by I/O
errors or system reboot, cached_dev->count can always be used correctly.
The patch is simple, but understanding how it works is quite complicated.
Changelog:
v2: set dc->writeback_thread to NULL in this patch, as suggested by Hannes.
v1: initial version for review.
Signed-off-by: Coly Li <colyli@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Michael Lyle <mlyle@lyle.org>
Cc: Michael Lyle <mlyle@lyle.org>
Cc: Junhui Tang <tang.junhui@zte.com.cn>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Christoph Hellwig [Wed, 14 Mar 2018 15:56:53 +0000 (16:56 +0100)]
block: bio_check_eod() needs to consider partitions
bio_check_eod() should check partition size not the whole disk if
bio->bi_partno is non-zero. Do this by moving the call
to bio_check_eod() into blk_partition_remap().
Based on an earlier patch from Jiufei Xue.
Fixes: 74d46992e0d9 ("block: replace bi_bdev with a gendisk pointer and partitions index")
Reported-by: Jiufei Xue <jiufei.xue@linux.alibaba.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bart Van Assche [Wed, 14 Mar 2018 22:48:06 +0000 (15:48 -0700)]
block: Move SECTOR_SIZE and SECTOR_SHIFT definitions into <linux/blkdev.h>
It happens often while I'm preparing a patch for a block driver that
I'm wondering: is a definition of SECTOR_SIZE and/or SECTOR_SHIFT
available for this driver? Do I have to introduce definitions of these
constants before I can use these constants? To avoid this confusion,
move the existing definitions of SECTOR_SIZE and SECTOR_SHIFT into the
<linux/blkdev.h> header file such that these become available for all
block drivers. Make the SECTOR_SIZE definition in the uapi msdos_fs.h
header file conditional to avoid that including that header file after
<linux/blkdev.h> causes the compiler to complain about a SECTOR_SIZE
redefinition.
Note: the SECTOR_SIZE / SECTOR_SHIFT / SECTOR_BITS definitions have
not been removed from uapi header files nor from NAND drivers in
which these constants are used for another purpose than converting
block layer offsets and sizes into a number of sectors.
Cc: David S. Miller <davem@davemloft.net>
Cc: Mike Snitzer <snitzer@redhat.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bart Van Assche [Fri, 16 Mar 2018 17:31:11 +0000 (10:31 -0700)]
blk-mq-debugfs: Show more request state information
Since commit
634f9e4631a8 ("blk-mq: remove REQ_ATOM_COMPLETE usages
from blk-mq") blk_rq_is_complete() only reports whether or not a
request has completed for legacy queues. Hence modify the
blk-mq-debugfs code such that it shows the blk-mq request state
again.
Fixes: 634f9e4631a8 ("blk-mq: remove REQ_ATOM_COMPLETE usages from blk-mq")
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Joseph Qi [Fri, 16 Mar 2018 06:51:27 +0000 (14:51 +0800)]
blk-throttle: fix race between blkcg_bio_issue_check() and cgroup_rmdir()
We've triggered a WARNING in blk_throtl_bio() when throttling writeback
io, which complains blkg->refcnt is already 0 when calling blkg_get(),
and then kernel crashes with invalid page request.
After investigating this issue, we've found it is caused by a race
between blkcg_bio_issue_check() and cgroup_rmdir(), which is described
below:
writeback kworker cgroup_rmdir
cgroup_destroy_locked
kill_css
css_killed_ref_fn
css_killed_work_fn
offline_css
blkcg_css_offline
blkcg_bio_issue_check
rcu_read_lock
blkg_lookup
spin_trylock(q->queue_lock)
blkg_destroy
spin_unlock(q->queue_lock)
blk_throtl_bio
spin_lock_irq(q->queue_lock)
...
spin_unlock_irq(q->queue_lock)
rcu_read_unlock
Since rcu can only prevent blkg from releasing when it is being used,
the blkg->refcnt can be decreased to 0 during blkg_destroy() and schedule
blkg release.
Then trying to blkg_get() in blk_throtl_bio() will complains the WARNING.
And then the corresponding blkg_put() will schedule blkg release again,
which result in double free.
This race is introduced by commit
ae1188963611 ("blkcg: consolidate blkg
creation in blkcg_bio_issue_check()"). Before this commit, it will
lookup first and then try to lookup/create again with queue_lock. Since
revive this logic is a bit drastic, so fix it by only offlining pd during
blkcg_css_offline(), and move the rest destruction (especially
blkg_put()) into blkcg_css_free(), which should be the right way as
discussed.
Fixes: ae1188963611 ("blkcg: consolidate blkg creation in blkcg_bio_issue_check()")
Reported-by: Jiufei Xue <jiufei.xue@linux.alibaba.com>
Signed-off-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>