- 06 Jun, 2015 3 commits
-
-
Bob Liu authored
Extend xen/block to support multi-page ring, so that more requests can be issued by using more than one pages as the request ring between blkfront and backend. As a result, the performance can get improved significantly. We got some impressive improvements on our highend iscsi storage cluster backend. If using 64 pages as the ring, the IOPS increased about 15 times for the throughput testing and above doubled for the latency testing. The reason was the limit on outstanding requests is 32 if use only one-page ring, but in our case the iscsi lun was spread across about 100 physical drives, 32 was really not enough to keep them busy. Changes in v2: - Rebased to 4.0-rc6. - Document on how multi-page ring feature working to linux io/blkif.h. Changes in v3: - Remove changes to linux io/blkif.h and follow the protocol defined in io/blkif.h of XEN tree. - Rebased to 4.1-rc3 Changes in v4: - Turn to use 'ring-page-order' and 'max-ring-page-order'. - A few comments from Roger. Changes in v5: - Clarify with 4k granularity to comment - Address more comments from Roger Signed-off-by:
Bob Liu <bob.liu@oracle.com> Signed-off-by:
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
-
Bob Liu authored
The major responsibility of talk_to_blkback() is allocate and initialize the request ring and write the ring info to xenstore. But this work should be done after backend entered 'XenbusStateInitWait' as defined in the protocol file. See xen/include/public/io/blkif.h in XEN git tree: Front Back ================================= ===================================== XenbusStateInitialising XenbusStateInitialising o Query virtual device o Query backend device identification properties. data. o Setup OS device instance. o Open and validate backend device. o Publish backend features and transport parameters. | | V XenbusStateInitWait o Query backend features and transport parameters. o Allocate and initialize the request ring. There is no problem with this yet, but it is an violation of the design and furthermore it would not allow frontend/backend to negotiate 'multi-page' and 'multi-queue' features. Changes in v2: - Re-write the commit message to be more clear. Signed-off-by:Roger Pau Monné <roger.pau@citrix.com> Signed-off-by:
-
Bob Liu authored
This is a pre-patch for multi-page ring feature. In connect_ring, we can know exactly how many pages are used for the shared ring, delay pending_req allocation here so that we won't waste too much memory. Signed-off-by:
-
- 05 Jun, 2015 5 commits
-
-
Keith Busch authored
Namespaces may be dynamically allocated and deleted or attached and detached. This has the driver rescan the device for namespace changes after each device reset or namespace change asynchronous event. There could potentially be many detached namespaces that we don't want polluting /dev/ with unusable block handles, so this will delete disks if the namespace is not active as indicated by the response from identify namespace. This also skips adding the disk if no capacity is provisioned to the namespace in the first place. Signed-off-by:
Keith Busch <keith.busch@intel.com> Signed-off-by:
Jens Axboe <axboe@fb.com>
-
Jens Axboe authored
-
Jens Axboe authored
We export this function and NVMe wants to use it, but for some reason it was never added to the block header. Do that. Signed-off-by: -
Jon Derrick authored
Protects against reordering and/or preempting which would allow the kthread to access the queue descriptor before it is set up Signed-off-by:
Jon Derrick <jonathan.derrick@intel.com> Acked-by:
-
Keith Busch authored
We need the ability to perform an nvme controller reset as discussed on the mailing list thread: http://lists.infradead.org/pipermail/linux-nvme/2015-March/001585.html This adds a sysfs entry that when written to will reset perform an NVMe controller reset if the controller was successfully initialized in the first place. This also adds locking around resetting the device in the async probe method so the driver can't schedule two resets. Signed-off-by:
-
- 02 Jun, 2015 2 commits
-
-
Akinobu Mita authored
When irqmode=2 (IRQ completion handler is timer) and queue_mode=1 (Block interface to use is rq), the completion handler should restart request handling for any pending requests on a queue because request processing stops when the number of commands are queued more than hw_queue_depth (null_rq_prep_fn returns BLKPREP_DEFER). Without this change, the following command cannot finish. # modprobe null_blk irqmode=2 queue_mode=1 hw_queue_depth=1 # fio --name=t --rw=read --size=1g --direct=1 \ --ioengine=libaio --iodepth=64 --filename=/dev/nullb0 Signed-off-by:
Akinobu Mita <akinobu.mita@gmail.com> Cc: Jens Axboe <axboe@fb.com> Signed-off-by:
-
Akinobu Mita authored
When irqmode=2 (IRQ completion handler is timer), timer handler should be called on the same CPU where the timer has been started. Since completion_queues are per-cpu and the completion handler only touches completion_queue for local CPU, we need to prevent the handler from running on a different CPU where the timer has been started. Otherwise, the IO cannot be completed until another completion handler is executed on that CPU. Signed-off-by:
-
- 01 Jun, 2015 3 commits
-
-
Keith Busch authored
The driver needs to track shared tags to support multiple namespaces that may be dynamically allocated or deleted. Relying on the first request_queue's hctx's is not appropriate as we cannot clear outstanding tags for all namespaces using this handle, nor can the driver easily track all request_queue's hctx as namespaces are attached/detached. Instead, this patch uses the nvme_dev's tagset to get the shared tag resources instead of through a request_queue hctx. Signed-off-by:
-
Jens Axboe authored
-
Keith Busch authored
Storage controllers may expose multiple block devices that share hardware resources managed by blk-mq. This patch enhances the shared tags so a low-level driver can access the shared resources not tied to the unshared h/w contexts. This way the LLD can dynamically add and delete disks and request queues without having to track all the request_queue hctx's to iterate outstanding tags. Signed-off-by:
-
- 29 May, 2015 4 commits
-
-
Jens Axboe authored
We don't need to honor chunk sizes for IO that doesn't carry any data. Signed-off-by: -
Jens Axboe authored
We can safely merge anything that wont generate an SG list entry, so if the bio is data-less (discard), don't look at potential SG gaps. Signed-off-by: -
Keith Busch authored
Do not retry failed sync commands so the original status may be seen without issuing unnecessary retries. Signed-off-by:
-
Keith Busch authored
Signed-off-by:
-
- 26 May, 2015 1 commit
-
-
Julia Lawall authored
Remove unneeded variable used to store return value. Generated by: scripts/coccinelle/misc/returnvar.cocci Signed-off-by:
Fengguang Wu <fengguang.wu@intel.com> Signed-off-by:
Julia Lawall <julia.lawall@lip6.fr> Signed-off-by:
-
- 22 May, 2015 12 commits
-
-
Keith Busch authored
Replaces req->sense_len usage, which is not owned by the LLD, to req->special to contain the command result for driver created commands, and sets the result unconditionally on completion. Signed-off-by:
-
Christoph Hellwig authored
Currently dm-multipath has to clone the bios for every request sent to the lower devices, which wastes cpu cycles and ties down memory. This patch instead adds a new REQ_CLONE flag that instructs req_bio_endio to not complete bios attached to a request, which we set on clone requests similar to bios in a flush sequence. With this change I/O errors on a path failure only get propagated to dm-multipath, which can then either resubmit the I/O or complete the bios on the original request. I've done some basic testing of this on a Linux target with ALUA support, and it survives path failures during I/O nicely. Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
Mike Snitzer <snitzer@redhat.com> Signed-off-by:
-
Mike Snitzer authored
Commit c4cf5261 ("bio: skip atomic inc/dec of ->bi_remaining for non-chains") regressed all existing callers that followed this pattern: 1) saving a bio's original bi_end_io 2) wiring up an intermediate bi_end_io 3) restoring the original bi_end_io from intermediate bi_end_io 4) calling bio_endio() to execute the restored original bi_end_io The regression was due to BIO_CHAIN only ever getting set if bio_inc_remaining() is called. For the above pattern it isn't set until step 3 above (step 2 would've needed to establish BIO_CHAIN). As such the first bio_endio(), in step 2 above, never decremented __bi_remaining before calling the intermediate bi_end_io -- leaving __bi_remaining with the value 1 instead of 0. When bio_inc_remaining() occurred during step 3 it brought it to a value of 2. When the second bio_endio() was called, in step 4 above, it should've called the original bi_end_io but it didn't because there was an extra reference that wasn't dropped (due to atomic operations being optimized away since BIO_CHAIN wasn't set upfront). Fix this issue by removing the __bi_remaining management complexity for all callers that use the above pattern -- bio_chain() is the only interface that _needs_ to be concerned with __bi_remaining. For the above pattern callers just expect the bi_end_io they set to get called! Remove bio_endio_nodec() and also remove all bio_inc_remaining() calls that aren't associated with the bio_chain() interface. Also, the bio_inc_remaining() interface has been moved local to bio.c. Fixes: c4cf5261 ("bio: skip atomic inc/dec of ->bi_remaining for non-chains") Reviewed-by:
Jan Kara <jack@suse.cz> Signed-off-by:
-
Christoph Hellwig authored
Use block layer queues with an internal cmd_type to submit internally generated NVMe commands. This both simplifies the code a lot and allow for a better structure. For example now the LighNVM code can construct commands without knowing the details of the underlying I/O descriptors. Or a future NVMe over network target could inject commands, as well as could the SCSI translation and ioctl code be reused for such a beast. Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Christoph Hellwig authored
NVMe device always support the FUA bit, and the SCSI translations accepts the DPO bit, which doesn't have much of a meaning for us. Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Christoph Hellwig authored
Erorr handling for the scsi translation was completely broken, as there were two different positive error number spaces overlapping. Fix this up by removing one of them, and centralizing the generation of the other positive values in a single place. Also fix up a few places that didn't handle the NVMe error codes properly. Signed-off-by:
-
Christoph Hellwig authored
This function handles two totally different opcodes, so split it. Signed-off-by:
-
Christoph Hellwig authored
Most users want the generic device, so store that in struct nvme_dev instead of the pci_dev. This also happens to be a nice step towards making some code reusable for non-PCI transports. Signed-off-by:
-
Christoph Hellwig authored
Note that we keep the unused timeout argument, but allow callers to pass 0 instead of a timeout if they want the default. This will allow adding a timeout to the pass through path later on. Signed-off-by:
-
- 20 May, 2015 9 commits
-
-
Jens Axboe authored
gcc, righfully, complains: drivers/block/loop.c:1369:1: warning: label 'out' defined but not used [-Wunused-label] Kill it. Signed-off-by: -
Jarod Wilson authored
With the mutex_trylock bit gone from blkdev_reread_part(), the retry logic in dasd_scan_partitions() shouldn't be necessary. CC: Christoph Hellwig <hch@infradead.org> CC: Jens Axboe <axboe@kernel.dk> CC: Tejun Heo <tj@kernel.org> CC: Alexander Viro <viro@zeniv.linux.org.uk> CC: Markus Pargmann <mpa@pengutronix.de> CC: Stefan Weinhuber <wein@de.ibm.com> CC: Stefan Haberland <stefan.haberland@de.ibm.com> CC: Sebastian Ott <sebott@linux.vnet.ibm.com> CC: Fabian Frederick <fabf@skynet.be> CC: Ming Lei <ming.lei@canonical.com> CC: David Herrmann <dh.herrmann@gmail.com> CC: Andrew Morton <akpm@linux-foundation.org> CC: Peter Zijlstra <peterz@infradead.org> CC: nbd-general@lists.sourceforge.net CC: linux-s390@vger.kernel.org Reviewed-by:
Sebastian Ott <sebott@linux.vnet.ibm.com> Signed-off-by:
Ming Lei <ming.lei@canonical.com> Signed-off-by:
Jarod Wilson <jarod@redhat.com> Signed-off-by:
-
Ming Lei authored
Also remove the obsolete comment. Reviewed-by:
-
Ming Lei authored
Reviewed-by:
-
Ming Lei authored
loop_clr_fd() can be run piggyback with lo_release(), and under this situation, reread partition may always fail because bd_mutex has been held already. This patch detects the situation by the reference count, and call __blkdev_reread_part() to avoid acquiring the lock again. In the meantime, this patch switches to new kernel APIs of blkdev_reread_part() and __blkdev_reread_part(). Reviewed-by:
-
Ming Lei authored
The lo_ctl_mutex is held for running all ioctl handlers, and in some ioctl handlers, ioctl_by_bdev(BLKRRPART) is called for rereading partitions, which requires bd_mutex. So it is easy to cause failure because trylock(bd_mutex) may fail inside blkdev_reread_part(), and follows the lock context: blkid or other application: ->open() ->mutex_lock(bd_mutex) ->lo_open() ->mutex_lock(lo_ctl_mutex) losetup(set fd ioctl): ->mutex_lock(lo_ctl_mutex) ->ioctl_by_bdev(BLKRRPART) ->trylock(bd_mutex) This patch trys to eliminate the ABBA lock dependency by removing lo_ctl_mutext in lo_open() with the following approach: 1) make lo_refcnt as atomic_t and avoid acquiring lo_ctl_mutex in lo_open(): - for open vs. add/del loop, no any problem because of loop_index_mutex - freeze request queue during clr_fd, so I/O can't come until clearing fd is completed, like the effect of holding lo_ctl_mutex in lo_open - both open() and release() have been serialized by bd_mutex already 2) don't hold lo_ctl_mutex for decreasing/checking lo_refcnt in lo_release(), then lo_ctl_mutex is only required for the last release. Reviewed-by:
-
Jens Axboe authored
We need the blkdev_reread_part() changes for drivers to adapt.
-
Ming Lei authored
The only possible problem of using mutex_lock() instead of trylock is about deadlock. If there aren't any locks held before calling blkdev_reread_part(), deadlock can't be caused by this conversion. If there are locks held before calling blkdev_reread_part(), and if these locks arn't required in open, close handler and I/O path, deadlock shouldn't be caused too. Both user space's ioctl(BLKRRPART) and md_setup_drive() from init/do_mounts_md.c belongs to the 1st case, so the conversion is safe for the two cases. For loop, the previous patches in this pathset has fixed the ABBA lock dependency, so the conversion is OK. For nbd, tx_lock is held when calling the function: - both open and release won't hold the lock - when blkdev_reread_part() is run, I/O thread has been stopped already, so tx_lock won't be acquired in I/O path at that time. - so the conversion won't cause deadlock for nbd For dasd, both dasd_open(), dasd_release() and request function don't acquire any mutex/semphone, so the conversion should be safe. Reviewed-by:
-
Jarod Wilson authored
This patch exports blkdev_reread_part() for block drivers, also introduce __blkdev_reread_part(). For some drivers, such as loop, reread of partitions can be run from the release path, and bd_mutex may already be held prior to calling ioctl_by_bdev(bdev, BLKRRPART, 0), so introduce __blkdev_reread_part for use in such cases. CC: Christoph Hellwig <hch@lst.de> CC: Jens Axboe <axboe@kernel.dk> CC: Tejun Heo <tj@kernel.org> CC: Alexander Viro <viro@zeniv.linux.org.uk> CC: Markus Pargmann <mpa@pengutronix.de> CC: Stefan Weinhuber <wein@de.ibm.com> CC: Stefan Haberland <stefan.haberland@de.ibm.com> CC: Sebastian Ott <sebott@linux.vnet.ibm.com> CC: Fabian Frederick <fabf@skynet.be> CC: Ming Lei <ming.lei@canonical.com> CC: David Herrmann <dh.herrmann@gmail.com> CC: Andrew Morton <akpm@linux-foundation.org> CC: Peter Zijlstra <peterz@infradead.org> CC: nbd-general@lists.sourceforge.net CC: linux-s390@vger.kernel.org Reviewed-by:
-
- 19 May, 2015 1 commit
-
-
Christoph Hellwig authored
Stop abusing struct page functionality and the swap end_io handler, and instead add a modified version of the blk-lib.c bio_batch helpers. Also move the block I/O code into swap.c as they are directly tied into each other. Signed-off-by:
Pavel Machek <pavel@ucw.cz> Tested-by:
Ming Lin <mlin@kernel.org> Acked-by:
Rafael J. Wysocki <rjw@rjwysocki.net> Signed-off-by:
-
