Commit 25520d55cdb6 ("block: Inline blk_integrity in struct gendisk")
introduced blk_integrity_revalidate(), which seems to assume ownership
of the stable pages flag and unilaterally clears it if no blk_integrity
profile is registered:
if (bi->profile)
disk->queue->backing_dev_info->capabilities |=
BDI_CAP_STABLE_WRITES;
else
disk->queue->backing_dev_info->capabilities &=
~BDI_CAP_STABLE_WRITES;
It's called from revalidate_disk() and rescan_partitions(), making it
impossible to enable stable pages for drivers that support partitions
and don't use blk_integrity: while the call in revalidate_disk() can be
trivially worked around (see zram, which doesn't support partitions and
hence gets away with zram_revalidate_disk()), rescan_partitions() can
be triggered from userspace at any time. This breaks rbd, where the
ceph messenger is responsible for generating/verifying CRCs.
Since blk_integrity_{un,}register() "must" be used for (un)registering
the integrity profile with the block layer, move BDI_CAP_STABLE_WRITES
setting there. This way drivers that call blk_integrity_register() and
use integrity infrastructure won't interfere with drivers that don't
but still want stable pages.
Fixes: 25520d55cdb6 ("block: Inline blk_integrity in struct gendisk") Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Mike Snitzer <snitzer@redhat.com> Cc: stable@vger.kernel.org # 4.4+, needs backporting Tested-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Signed-off-by: Jens Axboe <axboe@fb.com>
From userspace calling ioctl(NVM_DEV_CREATE) was returning ENOMEM for
invalid arguments even though pblk (pblk_init) was returning correctly
-EINVAL to nvm_create_tgt inside core. This patch propagates the
correct return value to userspace.
Because pblk was introduced recently this only needs to go in 4.12.
Avoid that the following kernel bug gets triggered:
BUG: sleeping function called from invalid context at ./include/linux/buffer_head.h:349
in_atomic(): 1, irqs_disabled(): 0, pid: 8019, name: find
CPU: 10 PID: 8019 Comm: find Tainted: G W I 4.11.0-rc4-dbg+ #2
Call Trace:
dump_stack+0x68/0x93
___might_sleep+0x16e/0x230
__might_sleep+0x4a/0x80
__ext4_get_inode_loc+0x1e0/0x4e0
ext4_iget+0x70/0xbc0
ext4_iget_normal+0x2f/0x40
ext4_lookup+0xb6/0x1f0
lookup_slow+0x104/0x1e0
walk_component+0x19a/0x330
path_lookupat+0x4b/0x100
filename_lookup+0x9a/0x110
user_path_at_empty+0x36/0x40
vfs_statx+0x67/0xc0
SYSC_newfstatat+0x20/0x40
SyS_newfstatat+0xe/0x10
entry_SYSCALL_64_fastpath+0x18/0xad
This happens since the big if/else in blk_mq_make_request() doesn't
have final else section that also drops the ctx. Add that.
Fixes: b00c53e8f411 ("blk-mq: fix schedule-while-atomic with scheduler attached") Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com> Cc: Omar Sandoval <osandov@fb.com>
Added a bit more to the commit log.
Merge branch 'nvme-4.12' of git://git.infradead.org/nvme into for-4.12/block
Christoph writes:
This is the current NVMe pile: virtualization extensions, lots of FC
updates and various misc bits. There are a few more FC bits that didn't
make the cut, but we'd like to get this request out before the merge
window for sure.
We need to get the command payload from the request before
we attempt to dereference it.
Fixes: 4dda4735c581 ("mtip32xx: add a status field to struct mtip_cmd") Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@fb.com>
Currently most IOs which return the nvme error codes are retried on
the other path if those IOs returns EIO from NVMe driver. This
patch let Multipath distinguish nvme media error codes and some
generic or cmd-specific nvme error codes so that multipath will
not retry those kinds of IO, to save bandwidth.
Signed-off-by: Junxiong Guan <guanjunxiong@huawei.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
Keith Busch [Fri, 24 Feb 2017 22:59:28 +0000 (17:59 -0500)]
nvme/pci: Poll CQ on timeout
If an IO timeout occurs, it's helpful to know if the controller did not
post a completion or the driver missed an interrupt. While we never expect
the latter, this patch will make it possible to tell the difference so
we don't have to guess.
Signed-off-by: Keith Busch <keith.busch@intel.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Tested-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
James Smart [Tue, 11 Apr 2017 18:35:08 +0000 (11:35 -0700)]
nvme_fc: Move LS's to rport
Link LS's on the remoteport rather than the controller. LS's are
between nport's. Makes more sense, especially on async teardown where
the controller is torn down regardless of the LS (LS is more of a notifier
to the target of the teardown), to have them on the remoteport.
While revising ls send/done routines, issues were seen relative to
refcounting and cleanup, especially in async path. Reworked these code
paths.
Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
James Smart [Tue, 11 Apr 2017 18:32:31 +0000 (11:32 -0700)]
nvmet_fc: Rework target side abort handling
target transport:
----------------------
There are cases when there is a need to abort in-progress target
operations (writedata) so that controller termination or errors can
clean up. That can't happen currently as the abort is another target
op type, so it can't be used till the running one finishes (and it may
not). Solve by removing the abort op type and creating a separate
downcall from the transport to the lldd to request an io to be aborted.
The transport will abort ios on queue teardown or io errors. In general
the transport tries to call the lldd abort only when the io state is
idle. Meaning: ops that transmit data (readdata or rsp) will always
finish their transmit (or the lldd will see a state on the
link or initiator port that fails the transmit) and the done call for
the operation will occur. The transport will wait for the op done
upcall before calling the abort function, and as the io is idle, the
io can be cleaned up immediately after the abort call; Similarly, ios
that are not waiting for data or transmitting data must be in the nvmet
layer being processed. The transport will wait for the nvmet layer
completion before calling the abort function, and as the io is idle,
the io can be cleaned up immediately after the abort call; As for ops
that are waiting for data (writedata), they may be outstanding
indefinitely if the lldd doesn't see a condition where the initiatior
port or link is bad. In those cases, the transport will call the abort
function and wait for the lldd's op done upcall for the operation, where
it will then clean up the io.
Additionally, if a lldd receives an ABTS and matches it to an outstanding
request in the transport, A new new transport upcall was created to abort
the outstanding request in the transport. The transport expects any
outstanding op call (readdata or writedata) will completed by the lldd and
the operation upcall made. The transport doesn't act on the reported
abort (e.g. clean up the io) until an op done upcall occurs, a new op is
attempted, or the nvmet layer completes the io processing.
fcloop:
----------------------
Updated to support the new target apis.
On fcp io aborts from the initiator, the loopback context is updated to
NULL out the half that has completed. The initiator side is immediately
called after the abort request with an io completion (abort status).
On fcp io aborts from the target, the io is stopped and the initiator side
sees it as an aborted io. Target side ops, perhaps in progress while the
initiator side is done, continue but noop the data movement as there's no
structure on the initiator side to reference.
patch also contains:
----------------------
Revised lpfc to support the new abort api
commonized rsp buffer syncing and nulling of private data based on
calling paths.
errors in op done calls don't take action on the fod. They're bad
operations which implies the fod may be bad.
Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
James Smart [Tue, 11 Apr 2017 18:32:30 +0000 (11:32 -0700)]
nvme_fcloop: split job struct from transport for req_release
Current design has the fcloop job struct, used for both initiator and
target processing, allocated as part of the initiator request structure.
On aborts, the initiator side (based on the request) may terminate, yet
the target side wants to continue processing. the target side can't do
that if the initiator side goes away.
Revise fcloop to allocate an independent target side structure when it
starts an io from the initiator.
Added a lock to the request struct as well to synchronize pointer updates
on abort calls.
Modified target downcalls to recognize conditions where initiator has
aborted the io (thus nulled the pointer between job structs), thus
avoid referencing sgl lists which are gone and no longer making upcalls
to the initiator.
In conditions where the targetport is no longer connected, have the
initiator return an access failure rather than simulating a command
completion.
Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
James Smart [Tue, 11 Apr 2017 18:32:29 +0000 (11:32 -0700)]
nvmet_fc: add req_release to lldd api
With the advent of the opdone calls changing context, the lldd can no
longer assume that once the op->done call returns for RSP operations
that the request struct is no longer being accessed.
As such, revise the lldd api for a req_release callback that the
transport will call when the job is complete. This will also be used
with abort cases.
Fixed text in api header for change in io complete semantics.
Revised lpfc to support the new req_release api.
Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
James Smart [Tue, 11 Apr 2017 18:32:28 +0000 (11:32 -0700)]
nvmet_fc: add target feature flags for upcall isr contexts
Two new feature flags were added to control whether upcalls to the
transport result in context switches or stay in the calling context.
NVMET_FCTGTFEAT_CMD_IN_ISR:
By default, if the flag is not set, the transport assumes the
lldd is in a non-isr context and in the cpu context it should be
for the io queue. As such, the cmd handler is called directly in the
calling context.
If the flag is set, indicating the upcall is an isr context, the
transport mandates a transition to a workqueue. The workqueue assigned
to the queue is used for the context.
NVMET_FCTGTFEAT_OPDONE_IN_ISR
By default, if the flag is not set, the transport assumes the
lldd is in a non-isr context and in the cpu context it should be
for the io queue. As such, the fcp operation done callback is called
directly in the calling context.
If the flag is set, indicating the upcall is an isr context, the
transport mandates a transition to a workqueue. The workqueue assigned
to the queue is used for the context.
Updated lpfc for flags
Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
This is safer as it doesn't rely on the data being stored in
a single page in an sgl.
It also aids our effort to start phasing out users of sg_page. See [1].
For this we kmalloc some memory, copy to it and free at the end. Note:
we can't allocate this memory on the stack as the kbuild test robot
reports some frame size overflows on i386.
[1] https://lwn.net/Articles/720053/
Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Max Gurtovoy <maxg@mellanox.com> Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Helen Koike [Mon, 10 Apr 2017 15:51:07 +0000 (12:51 -0300)]
nvme: improve performance for virtual NVMe devices
This change provides a mechanism to reduce the number of MMIO doorbell
writes for the NVMe driver. When running in a virtualized environment
like QEMU, the cost of an MMIO is quite hefy here. The main idea for
the patch is provide the device two memory location locations:
1) to store the doorbell values so they can be lookup without the doorbell
MMIO write
2) to store an event index.
I believe the doorbell value is obvious, the event index not so much.
Similar to the virtio specification, the virtual device can tell the
driver (guest OS) not to write MMIO unless you are writing past this
value.
FYI: doorbell values are written by the nvme driver (guest OS) and the
event index is written by the virtual device (host OS).
The patch implements a new admin command that will communicate where
these two memory locations reside. If the command fails, the nvme
driver will work as before without any optimizations.
Contributions:
Eric Northup <digitaleric@google.com>
Frank Swiderski <fes@google.com>
Ted Tso <tytso@mit.edu>
Keith Busch <keith.busch@intel.com>
Just to give an idea on the performance boost with the vendor
extension: Running fio [1], a stock NVMe driver I get about 200K read
IOPs with my vendor patch I get about 1000K read IOPs. This was
running with a null device i.e. the backing device simply returned
success on every read IO request.
Signed-off-by: Rob Nelson <rlnelson@google.com>
[mlin: port for upstream] Signed-off-by: Ming Lin <mlin@kernel.org>
[koike: updated for upstream] Signed-off-by: Helen Koike <helen.koike@collabora.co.uk> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Keith Busch <keith.busch@intel.com>
Josef Bacik [Thu, 20 Apr 2017 19:47:01 +0000 (15:47 -0400)]
nbd: set the max segments to USHRT_MAX
I lack the basic understanding of what segments mean, so we were being
limited to 512kib requests even with higher max_sectors sizes set.
Setting the maximum number of segments to unlimited allows us to
actually have arbitrarily large IO's go through NBD.
Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: Jens Axboe <axboe@fb.com>
commit c13660a08c8b ("blk-mq-sched: change ->dispatch_requests()
to ->dispatch_request()") removed the last user of this function.
Hence also remove the function itself.
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com> Cc: Omar Sandoval <osandov@fb.com> Cc: Hannes Reinecke <hare@suse.com> Signed-off-by: Jens Axboe <axboe@fb.com>
blk-mq: add might_sleep check to blk_mq_get_driver_tag()
If the caller passes in wait=true, it has to be able to block
for a driver tag. We just had a bug where flush insertion
would block on tag allocation, while we had preempt disabled.
Ensure that we catch cases like that earlier next time.
Reviewed-by: Bart Van Assche <Bart.VanAssche@sandisk.com> Signed-off-by: Jens Axboe <axboe@fb.com>
Stephen Bates [Thu, 20 Apr 2017 22:59:11 +0000 (16:59 -0600)]
blk-mq: Fix poll_stat for new size-based bucketing.
Fixes an issue where the size of the poll_stat array in request_queue
does not match the size expected by the new size based bucketing for
IO completion polling.
Fixes: 720b8ccc4500 ("blk-mq: Add a polling specific stats function") Signed-off-by: Stephen Bates <sbates@raithlin.com> Signed-off-by: Jens Axboe <axboe@fb.com>
blk-mq: fix schedule-while-atomic with scheduler attached
We must have dropped the ctx before we call
blk_mq_sched_insert_request() with can_block=true, otherwise we risk
that a flush request can block on insertion if we are currently out of
tags.
Stephen Bates [Fri, 7 Apr 2017 12:24:03 +0000 (06:24 -0600)]
blk-mq: Add a polling specific stats function
Rather than bucketing IO statisics based on direction only we also
bucket based on the IO size. This leads to improved polling
performance. Update the bucket callback function and use it in the
polling latency estimation.
Signed-off-by: Stephen Bates <sbates@raithlin.com> Signed-off-by: Jens Axboe <axboe@fb.com>
This is for the legacy floppy and ataflop drivers that currently abuse
->errors for this purpose. It's stashed away in a union to not grow
the struct size, the other fields are either used by modern drivers
for different purposes or the I/O scheduler before queing the I/O
to drivers.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@fb.com>
blk-mq: remove the error argument to blk_mq_complete_request
Now that all drivers that call blk_mq_complete_requests have a
->complete callback we can remove the direct call to blk_mq_end_request,
as well as the error argument to blk_mq_complete_request.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Bart Van Assche <Bart.VanAssche@sandisk.com> Signed-off-by: Jens Axboe <axboe@fb.com>
xen-blkfron is the last users using rq->errros for passing back error to
blk-mq, and I'd like to get rid of that. In the longer run the driver
should be moving more of the completion processing into .complete, but
this is the minimal change to move forward for now.
Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Roger Pau Monné <roger.pau@citrix.com> Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Jens Axboe <axboe@fb.com>
scsi: introduce a result field in struct scsi_request
This passes on the scsi_cmnd result field to users of passthrough
requests. Currently we abuse req->errors for this purpose, but that
field will go away in its current form.
Note that the old IDE code abuses the errors field in very creative
ways and stores all kinds of different values in it. I didn't dare
to touch this magic, so the abuses are brought forward 1:1.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Reviewed-by: Bart Van Assche <Bart.VanAssche@sandisk.com> Signed-off-by: Jens Axboe <axboe@fb.com>
Remove passing req->errors (which at that point is always 0) to
blk_mq_complete_request, and rely on the virtio status code for the
serial number passthrough request.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Jens Axboe <axboe@fb.com>
Currently it's used by the lighnvm passthrough ioctl, but we'd like to make
it private in preparation of block layer specific error code. Lighnvm already
returns the real NVMe status anyway, so I think we can just limit it to
returning -EIO for any status set.
This will need a careful audit from the lightnvm folks, though.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@fb.com>
We want our own clearly defined error field for NVMe passthrough commands,
and the request errors field is going away in its current form.
Just store the status and result field in the nvme_request field from
hardirq completion context (using a new helper) and then generate a
Linux errno for the block layer only when we actually need it.
Because we can't overload the status value with a negative error code
for cancelled command we now have a flags filed in struct nvme_request
that contains a bit for this condition.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Jens Axboe <axboe@fb.com>
nvme-fc: fix status code handling in nvme_fc_fcpio_done
nvme_complete_async_event expects the little endian status code
including the phase bit, and a new completion handler I plan to
introduce will do so as well.
Change the status variable into the little endian format with the
phase bit used in the NVMe CQE to fix / enable this.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:46 +0000 (12:24 +0200)]
ubifs: Convert to separately allocated bdi
Allocate struct backing_dev_info separately instead of embedding it
inside the superblock. This unifies handling of bdi among users.
CC: Richard Weinberger <richard@nod.at> CC: Artem Bityutskiy <dedekind1@gmail.com> CC: Adrian Hunter <adrian.hunter@intel.com> CC: linux-mtd@lists.infradead.org Acked-by: Richard Weinberger <richard@nod.at> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:44 +0000 (12:24 +0200)]
ncpfs: Convert to separately allocated bdi
Allocate struct backing_dev_info separately instead of embedding it
inside the superblock. This unifies handling of bdi among users.
CC: Petr Vandrovec <petr@vandrovec.name> Acked-by: Petr Vandrovec <petr@vandrovec.name> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:43 +0000 (12:24 +0200)]
nilfs2: Convert to properly refcounting bdi
Similarly to set_bdev_super() NILFS2 just used block device reference to
bdi. Convert it to properly getting bdi reference. The reference will
get automatically dropped on superblock destruction.
CC: linux-nilfs@vger.kernel.org Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:42 +0000 (12:24 +0200)]
gfs2: Convert to properly refcounting bdi
Similarly to set_bdev_super() GFS2 just used block device reference to
bdi. Convert it to properly getting bdi reference. The reference will
get automatically dropped on superblock destruction.
CC: Steven Whitehouse <swhiteho@redhat.com> CC: Bob Peterson <rpeterso@redhat.com> CC: cluster-devel@redhat.com Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:38 +0000 (12:24 +0200)]
coda: Convert to separately allocated bdi
Allocate struct backing_dev_info separately instead of embedding it
inside the superblock. This unifies handling of bdi among users.
CC: Jan Harkes <jaharkes@cs.cmu.edu> CC: coda@cs.cmu.edu CC: codalist@coda.cs.cmu.edu Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:37 +0000 (12:24 +0200)]
mtd: Convert to dynamically allocated bdi infrastructure
MTD already allocates backing_dev_info dynamically. Convert it to use
generic infrastructure for this including proper refcounting. We drop
mtd->backing_dev_info as its only use was to pass mtd_bdi pointer from
one file into another and if we wanted to keep that in a clean way, we'd
have to make mtd hold and drop bdi reference as needed which seems
pointless for passing one global pointer...
CC: David Woodhouse <dwmw2@infradead.org> CC: Brian Norris <computersforpeace@gmail.com> CC: linux-mtd@lists.infradead.org Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:36 +0000 (12:24 +0200)]
afs: Convert to separately allocated bdi
Allocate struct backing_dev_info separately instead of embedding it
inside the superblock. This unifies handling of bdi among users.
CC: David Howells <dhowells@redhat.com> CC: linux-afs@lists.infradead.org Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:34 +0000 (12:24 +0200)]
cifs: Convert to separately allocated bdi
Allocate struct backing_dev_info separately instead of embedding it
inside superblock. This unifies handling of bdi among users.
CC: Steve French <sfrench@samba.org> CC: linux-cifs@vger.kernel.org Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:32 +0000 (12:24 +0200)]
btrfs: Convert to separately allocated bdi
Allocate struct backing_dev_info separately instead of embedding it
inside superblock. This unifies handling of bdi among users.
CC: Chris Mason <clm@fb.com> CC: Josef Bacik <jbacik@fb.com> CC: David Sterba <dsterba@suse.com> CC: linux-btrfs@vger.kernel.org Reviewed-by: Liu Bo <bo.li.liu@oracle.com> Reviewed-by: David Sterba <dsterba@suse.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:31 +0000 (12:24 +0200)]
9p: Convert to separately allocated bdi
Allocate struct backing_dev_info separately instead of embedding it
inside session. This unifies handling of bdi among users.
CC: Eric Van Hensbergen <ericvh@gmail.com> CC: Ron Minnich <rminnich@sandia.gov> CC: Latchesar Ionkov <lucho@ionkov.net> CC: v9fs-developer@lists.sourceforge.net Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:30 +0000 (12:24 +0200)]
lustre: Convert to separately allocated bdi
Allocate struct backing_dev_info separately instead of embedding it
inside superblock. This unifies handling of bdi among users.
CC: Oleg Drokin <oleg.drokin@intel.com> CC: Andreas Dilger <andreas.dilger@intel.com> CC: James Simmons <jsimmons@infradead.org> CC: lustre-devel@lists.lustre.org Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:29 +0000 (12:24 +0200)]
fs: Get proper reference for s_bdi
So far we just relied on block device to hold a bdi reference for us
while the filesystem is mounted. While that works perfectly fine, it is
a bit awkward that we have a pointer to a refcounted structure in the
superblock without proper reference. So make s_bdi hold a proper
reference to block device's BDI. No filesystem using mount_bdev()
actually changes s_bdi so this is safe and will make bdev filesystems
work the same way as filesystems needing to set up their private bdi.
Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
Jan Kara [Wed, 12 Apr 2017 10:24:28 +0000 (12:24 +0200)]
fs: Provide infrastructure for dynamic BDIs in filesystems
Provide helper functions for setting up dynamically allocated
backing_dev_info structures for filesystems and cleaning them up on
superblock destruction.
Jan Kara [Wed, 12 Apr 2017 10:24:26 +0000 (12:24 +0200)]
block: Unregister bdi on last reference drop
Most users will want to unregister bdi when dropping last reference to a
bdi. Only a few users (like block devices) want to play more complex
tricks with bdi registration and unregistration. So unregister bdi when
the last reference to bdi is dropped and just make sure we don't
unregister the bdi the second time if it is already unregistered.
Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
blk-throttle: fix unused variable warning with BLK_DEV_THROTTLING_LOW=n
We trigger this warning:
block/blk-throttle.c: In function ‘blk_throtl_bio’:
block/blk-throttle.c:2042:6: warning: variable ‘ret’ set but not used [-Wunused-but-set-variable]
int ret;
^~~
since we only assign 'ret' if BLK_DEV_THROTTLING_LOW is off, we never
check it.
Reported-by: Bart Van Assche <bart.vanassche@sandisk.com> Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com> Signed-off-by: Jens Axboe <axboe@fb.com>
If we don't have CGROUPS enabled, the compile ends in the
following misery:
In file included from ../block/bfq-iosched.c:105:0:
../block/bfq-iosched.h:819:22: error: array type has incomplete element type
extern struct cftype bfq_blkcg_legacy_files[];
^
../block/bfq-iosched.h:820:22: error: array type has incomplete element type
extern struct cftype bfq_blkg_files[];
^
Move the declarations under the right ifdef.
Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Jens Axboe <axboe@fb.com>
Colin Ian King [Thu, 20 Apr 2017 14:07:18 +0000 (15:07 +0100)]
block, bfq: don't dereference bic before null checking it
The call to bfq_check_ioprio_change will dereference bic, however,
the null check for bic is after this call. Move the the null
check on bic to before the call to avoid any potential null
pointer dereference issues.
Detected by CoverityScan, CID#1430138 ("Dereference before null check")
Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Jens Axboe <axboe@fb.com>
On an error path in NVM_DEV_CREATE ioctl blk_put_queue is being called
twice: one via blk_cleanup_queue and another via put_disk. Straight fix
seems to remove queue pointer so that disk_release never ends up caling
blk_put_queue again.
Since ioprio_best() translates IOPRIO_CLASS_NONE into IOPRIO_CLASS_BE
and since lower numerical priority values represent a higher priority
a simple numerical comparison is sufficient.
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com> Reviewed-by: Adam Manzanares <adam.manzanares@wdc.com> Tested-by: Adam Manzanares <adam.manzanares@wdc.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Cc: Matias Bjørling <m@bjorling.me> Signed-off-by: Jens Axboe <axboe@fb.com>
Since only a single caller remains, inline blk_rq_set_prio(). Initialize
req->ioprio even if no I/O priority has been set in the bio nor in the
I/O context.
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com> Reviewed-by: Adam Manzanares <adam.manzanares@wdc.com> Tested-by: Adam Manzanares <adam.manzanares@wdc.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Cc: Matias Bjørling <m@bjorling.me> Signed-off-by: Jens Axboe <axboe@fb.com>
lightnvm: Use blk_init_request_from_bio() instead of open-coding it
This patch changes the behavior of the lightnvm driver as follows:
* REQ_FAILFAST_MASK is set for read-ahead requests.
* If no I/O priority has been set in the bio, the I/O priority is
copied from the I/O context.
* The rq_disk member is initialized if bio->bi_bdev != NULL.
* The bio sector offset is copied into req->__sector instead of
retaining the value -1 set by blk_mq_alloc_request().
* req->errors is initialized to zero.
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Cc: Matias Bjørling <m@bjorling.me> Cc: Adam Manzanares <adam.manzanares@wdc.com> Signed-off-by: Jens Axboe <axboe@fb.com>
null_blk: Use blk_init_request_from_bio() instead of open-coding it
This patch changes the behavior of the null_blk driver for the
LightNVM mode as follows:
* REQ_FAILFAST_MASK is set for read-ahead requests.
* If no I/O priority has been set in the bio, the I/O priority is
copied from the I/O context.
* The rq_disk member is initialized if bio->bi_bdev != NULL.
* req->errors is initialized to zero.
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Cc: Matias Bjørling <m@bjorling.me> Cc: Adam Manzanares <adam.manzanares@wdc.com> Signed-off-by: Jens Axboe <axboe@fb.com>
The driver uses both u64 and sector_t to refer to offsets, and assigns between the
two. This causes one harmless warning when sector_t is 32-bit:
drivers/lightnvm/pblk-rb.c: In function 'pblk_rb_write_entry_gc':
include/linux/lightnvm.h:215:20: error: large integer implicitly truncated to unsigned type [-Werror=overflow]
drivers/lightnvm/pblk-rb.c:324:22: note: in expansion of macro 'ADDR_EMPTY'
As the driver is already doing this inconsistently, changing the type
won't make it worse and is an easy way to avoid the warning.
Jan Kara [Wed, 19 Apr 2017 09:33:27 +0000 (11:33 +0200)]
block: Make writeback throttling defaults consistent for SQ devices
When CFQ is used as an elevator, it disables writeback throttling
because they don't play well together. Later when a different elevator
is chosen for the device, writeback throttling doesn't get enabled
again as it should. Make sure CFQ enables writeback throttling (if it
should be enabled by default) when we switch from it to another IO
scheduler.
Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
Paolo Valente [Wed, 19 Apr 2017 14:48:24 +0000 (08:48 -0600)]
block, bfq: split bfq-iosched.c into multiple source files
The BFQ I/O scheduler features an optimal fair-queuing
(proportional-share) scheduling algorithm, enriched with several
mechanisms to boost throughput and reduce latency for interactive and
real-time applications. This makes BFQ a large and complex piece of
code. This commit addresses this issue by splitting BFQ into three
main, independent components, and by moving each component into a
separate source file:
1. Main algorithm: handles the interaction with the kernel, and
decides which requests to dispatch; it uses the following two further
components to achieve its goals.
2. Scheduling engine (Hierarchical B-WF2Q+ scheduling algorithm):
computes the schedule, using weights and budgets provided by the above
component.
3. cgroups support: handles group operations (creation, destruction,
move, ...).
Signed-off-by: Paolo Valente <paolo.valente@linaro.org> Signed-off-by: Jens Axboe <axboe@fb.com>
Paolo Valente [Wed, 12 Apr 2017 16:23:21 +0000 (18:23 +0200)]
block, bfq: remove all get and put of I/O contexts
When a bfq queue is set in service and when it is merged, a reference
to the I/O context associated with the queue is taken. This reference
is then released when the queue is deselected from service or
split. More precisely, the release of the reference is postponed to
when the scheduler lock is released, to avoid nesting between the
scheduler and the I/O-context lock. In fact, such nesting would lead
to deadlocks, because of other code paths that take the same locks in
the opposite order. This postponing of I/O-context releases does
complicate code.
This commit addresses these issue by modifying involved operations in
such a way to not need to get the above I/O-context references any
more. Then it also removes any get and release of these references.
Signed-off-by: Paolo Valente <paolo.valente@linaro.org> Signed-off-by: Jens Axboe <axboe@fb.com>
Many popular I/O-intensive services or applications spawn or
reactivate many parallel threads/processes during short time
intervals. Examples are systemd during boot or git grep. These
services or applications benefit mostly from a high throughput: the
quicker the I/O generated by their processes is cumulatively served,
the sooner the target job of these services or applications gets
completed. As a consequence, it is almost always counterproductive to
weight-raise any of the queues associated to the processes of these
services or applications: in most cases it would just lower the
throughput, mainly because weight-raising also implies device idling.
To address this issue, an I/O scheduler needs, first, to detect which
queues are associated with these services or applications. In this
respect, we have that, from the I/O-scheduler standpoint, these
services or applications cause bursts of activations, i.e.,
activations of different queues occurring shortly after each
other. However, a shorter burst of activations may be caused also by
the start of an application that does not consist in a lot of parallel
I/O-bound threads (see the comments on the function bfq_handle_burst
for details).
In view of these facts, this commit introduces:
1) an heuristic to detect (only) bursts of queue activations caused by
services or applications consisting in many parallel I/O-bound
threads;
2) the prevention of device idling and weight-raising for the queues
belonging to these bursts.
Paolo Valente [Wed, 12 Apr 2017 16:23:19 +0000 (18:23 +0200)]
block, bfq: boost the throughput with random I/O on NCQ-capable HDDs
This patch is basically the counterpart, for NCQ-capable rotational
devices, of the previous patch. Exactly as the previous patch does on
flash-based devices and for any workload, this patch disables device
idling on rotational devices, but only for random I/O. In fact, only
with these queues disabling idling boosts the throughput on
NCQ-capable rotational devices. To not break service guarantees,
idling is disabled for NCQ-enabled rotational devices only when the
same symmetry conditions considered in the previous patches hold.
Paolo Valente [Wed, 12 Apr 2017 16:23:18 +0000 (18:23 +0200)]
block, bfq: boost the throughput on NCQ-capable flash-based devices
This patch boosts the throughput on NCQ-capable flash-based devices,
while still preserving latency guarantees for interactive and soft
real-time applications. The throughput is boosted by just not idling
the device when the in-service queue remains empty, even if the queue
is sync and has a non-null idle window. This helps to keep the drive's
internal queue full, which is necessary to achieve maximum
performance. This solution to boost the throughput is a port of
commits a68bbdd and f7d7b7a for CFQ.
As already highlighted in a previous patch, allowing the device to
prefetch and internally reorder requests trivially causes loss of
control on the request service order, and hence on service guarantees.
Fortunately, as discussed in detail in the comments on the function
bfq_bfqq_may_idle(), if every process has to receive the same
fraction of the throughput, then the service order enforced by the
internal scheduler of a flash-based device is relatively close to that
enforced by BFQ. In particular, it is close enough to let service
guarantees be substantially preserved.
Things change in an asymmetric scenario, i.e., if not every process
has to receive the same fraction of the throughput. In this case, to
guarantee the desired throughput distribution, the device must be
prevented from prefetching requests. This is exactly what this patch
does in asymmetric scenarios.
block, bfq: reduce idling only in symmetric scenarios
A seeky queue (i..e, a queue containing random requests) is assigned a
very small device-idling slice, for throughput issues. Unfortunately,
given the process associated with a seeky queue, this behavior causes
the following problem: if the process, say P, performs sync I/O and
has a higher weight than some other processes doing I/O and associated
with non-seeky queues, then BFQ may fail to guarantee to P its
reserved share of the throughput. The reason is that idling is key
for providing service guarantees to processes doing sync I/O [1].
This commit addresses this issue by allowing the device-idling slice
to be reduced for a seeky queue only if the scenario happens to be
symmetric, i.e., if all the queues are to receive the same share of
the throughput.
[1] P. Valente, A. Avanzini, "Evolution of the BFQ Storage I/O
Scheduler", Proceedings of the First Workshop on Mobile System
Technologies (MST-2015), May 2015.
http://algogroup.unimore.it/people/paolo/disk_sched/mst-2015.pdf
A set of processes may happen to perform interleaved reads, i.e.,
read requests whose union would give rise to a sequential read pattern.
There are two typical cases: first, processes reading fixed-size chunks
of data at a fixed distance from each other; second, processes reading
variable-size chunks at variable distances. The latter case occurs for
example with QEMU, which splits the I/O generated by a guest into
multiple chunks, and lets these chunks be served by a pool of I/O
threads, iteratively assigning the next chunk of I/O to the first
available thread. CFQ denotes as 'cooperating' a set of processes that
are doing interleaved I/O, and when it detects cooperating processes,
it merges their queues to obtain a sequential I/O pattern from the union
of their I/O requests, and hence boost the throughput.
Unfortunately, in the following frequent case, the mechanism
implemented in CFQ for detecting cooperating processes and merging
their queues is not responsive enough to handle also the fluctuating
I/O pattern of the second type of processes. Suppose that one process
of the second type issues a request close to the next request to serve
of another process of the same type. At that time the two processes
would be considered as cooperating. But, if the request issued by the
first process is to be merged with some other already-queued request,
then, from the moment at which this request arrives, to the moment
when CFQ controls whether the two processes are cooperating, the two
processes are likely to be already doing I/O in distant zones of the
disk surface or device memory.
CFQ uses however preemption to get a sequential read pattern out of
the read requests performed by the second type of processes too. As a
consequence, CFQ uses two different mechanisms to achieve the same
goal: boosting the throughput with interleaved I/O.
This patch introduces Early Queue Merge (EQM), a unified mechanism to
get a sequential read pattern with both types of processes. The main
idea is to immediately check whether a newly-arrived request lets some
pair of processes become cooperating, both in the case of actual
request insertion and, to be responsive with the second type of
processes, in the case of request merge. Both types of processes are
then handled by just merging their queues.
Paolo Valente [Wed, 12 Apr 2017 16:23:15 +0000 (18:23 +0200)]
block, bfq: reduce latency during request-pool saturation
This patch introduces an heuristic that reduces latency when the
I/O-request pool is saturated. This goal is achieved by disabling
device idling, for non-weight-raised queues, when there are weight-
raised queues with pending or in-flight requests. In fact, as
explained in more detail in the comment on the function
bfq_bfqq_may_idle(), this reduces the rate at which processes
associated with non-weight-raised queues grab requests from the pool,
thereby increasing the probability that processes associated with
weight-raised queues get a request immediately (or at least soon) when
they need one. Along the same line, if there are weight-raised queues,
then this patch halves the service rate of async (write) requests for
non-weight-raised queues.
Paolo Valente [Wed, 12 Apr 2017 16:23:14 +0000 (18:23 +0200)]
block, bfq: preserve a low latency also with NCQ-capable drives
I/O schedulers typically allow NCQ-capable drives to prefetch I/O
requests, as NCQ boosts the throughput exactly by prefetching and
internally reordering requests.
Unfortunately, as discussed in detail and shown experimentally in [1],
this may cause fairness and latency guarantees to be violated. The
main problem is that the internal scheduler of an NCQ-capable drive
may postpone the service of some unlucky (prefetched) requests as long
as it deems serving other requests more appropriate to boost the
throughput.
This patch addresses this issue by not disabling device idling for
weight-raised queues, even if the device supports NCQ. This allows BFQ
to start serving a new queue, and therefore allows the drive to
prefetch new requests, only after the idling timeout expires. At that
time, all the outstanding requests of the expired queue have been most
certainly served.
[1] P. Valente and M. Andreolini, "Improving Application
Responsiveness with the BFQ Disk I/O Scheduler", Proceedings of
the 5th Annual International Systems and Storage Conference
(SYSTOR '12), June 2012.
Slightly extended version:
http://algogroup.unimore.it/people/paolo/disk_sched/bfq-v1-suite-
results.pdf
Paolo Valente [Wed, 12 Apr 2017 16:23:13 +0000 (18:23 +0200)]
block, bfq: reduce I/O latency for soft real-time applications
To guarantee a low latency also to the I/O requests issued by soft
real-time applications, this patch introduces a further heuristic,
which weight-raises (in the sense explained in the previous patch)
also the queues associated to applications deemed as soft real-time.
To be deemed as soft real-time, an application must meet two
requirements. First, the application must not require an average
bandwidth higher than the approximate bandwidth required to playback
or record a compressed high-definition video. Second, the request
pattern of the application must be isochronous, i.e., after issuing a
request or a batch of requests, the application must stop issuing new
requests until all its pending requests have been completed. After
that, the application may issue a new batch, and so on.
As for the second requirement, it is critical to require also that,
after all the pending requests of the application have been completed,
an adequate minimum amount of time elapses before the application
starts issuing new requests. This prevents also greedy (i.e.,
I/O-bound) applications from being incorrectly deemed, occasionally,
as soft real-time. In fact, if *any amount of time* is fine, then even
a greedy application may, paradoxically, meet both the above
requirements, if: (1) the application performs random I/O and/or the
device is slow, and (2) the CPU load is high. The reason is the
following. First, if condition (1) is true, then, during the service
of the application, the throughput may be low enough to let the
application meet the bandwidth requirement. Second, if condition (2)
is true as well, then the application may occasionally behave in an
apparently isochronous way, because it may simply stop issuing
requests while the CPUs are busy serving other processes.
To address this issue, the heuristic leverages the simple fact that
greedy applications issue *all* their requests as quickly as they can,
whereas soft real-time applications spend some time processing data
after each batch of requests is completed. In particular, the
heuristic works as follows. First, according to the above isochrony
requirement, the heuristic checks whether an application may be soft
real-time, thereby giving to the application the opportunity to be
deemed as such, only when both the following two conditions happen to
hold: 1) the queue associated with the application has expired and is
empty, 2) there is no outstanding request of the application.
Suppose that both conditions hold at time, say, t_c and that the
application issues its next request at time, say, t_i. At time t_c the
heuristic computes the next time instant, called soft_rt_next_start in
the code, such that, only if t_i >= soft_rt_next_start, then both the
next conditions will hold when the application issues its next
request: 1) the application will meet the above bandwidth requirement,
2) a given minimum time interval, say Delta, will have elapsed from
time t_c (so as to filter out greedy application).
The current value of Delta is a little bit higher than the value that
we have found, experimentally, to be adequate on a real,
general-purpose machine. In particular we had to increase Delta to
make the filter quite precise also in slower, embedded systems, and in
KVM/QEMU virtual machines (details in the comments on the code).
If the application actually issues its next request after time
soft_rt_next_start, then its associated queue will be weight-raised
for a relatively short time interval. If, during this time interval,
the application proves again to meet the bandwidth and isochrony
requirements, then the end of the weight-raising period for the queue
is moved forward, and so on. Note that an application whose associated
queue never happens to be empty when it expires will never have the
opportunity to be deemed as soft real-time.
Paolo Valente [Wed, 12 Apr 2017 16:23:12 +0000 (18:23 +0200)]
block, bfq: improve responsiveness
This patch introduces a simple heuristic to load applications quickly,
and to perform the I/O requested by interactive applications just as
quickly. To this purpose, both a newly-created queue and a queue
associated with an interactive application (we explain in a moment how
BFQ decides whether the associated application is interactive),
receive the following two special treatments:
1) The weight of the queue is raised.
2) The queue unconditionally enjoys device idling when it empties; in
fact, if the requests of a queue are sync, then performing device
idling for the queue is a necessary condition to guarantee that the
queue receives a fraction of the throughput proportional to its weight
(see [1] for details).
For brevity, we call just weight-raising the combination of these
two preferential treatments. For a newly-created queue,
weight-raising starts immediately and lasts for a time interval that:
1) depends on the device speed and type (rotational or
non-rotational), and 2) is equal to the time needed to load (start up)
a large-size application on that device, with cold caches and with no
additional workload.
Finally, as for guaranteeing a fast execution to interactive,
I/O-related tasks (such as opening a file), consider that any
interactive application blocks and waits for user input both after
starting up and after executing some task. After a while, the user may
trigger new operations, after which the application stops again, and
so on. Accordingly, the low-latency heuristic weight-raises again a
queue in case it becomes backlogged after being idle for a
sufficiently long (configurable) time. The weight-raising then lasts
for the same time as for a just-created queue.
According to our experiments, the combination of this low-latency
heuristic and of the improvements described in the previous patch
allows BFQ to guarantee a high application responsiveness.
[1] P. Valente, A. Avanzini, "Evolution of the BFQ Storage I/O
Scheduler", Proceedings of the First Workshop on Mobile System
Technologies (MST-2015), May 2015.
http://algogroup.unimore.it/people/paolo/disk_sched/mst-2015.pdf