21 Mar, 2014

1 commit

• 676141e48   blk-mq: don't dump CPU -> hw queue map on driver load ... Browse Code »

  Now that we are out of initial debug/bringup mode, remove
  the verbose dump of the mapping table.

  Provide the mapping table in sysfs, under the hardware queue
  directory, in the cpu_list file.

  Signed-off-by: Jens Axboe

  Jens Axboe

  2014-03-21 03:31:44 +0800  

07 Dec, 2013

1 commit

• 851573660   block: fix memory leaks on unplugging block device ... Browse Code »

  All objects, which are allocated in blk_mq_register_disk, must be
  released in blk_mq_unregister_disk.

  I use a KVM virtual machine and virtio disk to reproduce this issue.

  kmemleak: 18 new suspected memory leaks (see /sys/kernel/debug/kmemleak)
  $ cat /sys/kernel/debug/kmemleak | head -n 30
  unreferenced object 0xffff8800b6636150 (size 8):
  comm "kworker/0:2", pid 65, jiffies 4294809903 (age 86.358s)
  hex dump (first 8 bytes):
  76 69 72 74 69 6f 34 00 virtio4.
  backtrace:
  [] kmemleak_alloc+0x4e/0xb0
  [] __kmalloc_track_caller+0xf5/0x260
  [] kstrdup+0x31/0x60
  [] sysfs_new_dirent+0x2e/0x140
  [] create_dir+0x38/0xe0
  [] sysfs_create_dir_ns+0x73/0xc0
  [] kobject_add_internal+0xc9/0x340
  [] kobject_add+0x65/0xb0
  [] device_add+0x128/0x660
  [] device_register+0x1a/0x20
  [] register_virtio_device+0x98/0xe0
  [] virtio_pci_probe+0x12e/0x1c0
  [] local_pci_probe+0x45/0xa0
  [] pci_device_probe+0x121/0x130
  [] driver_probe_device+0x87/0x390
  [] __device_attach+0x3b/0x40
  unreferenced object 0xffff8800b65aa1d8 (size 144):

  Fixes: 320ae51feed5 (blk-mq: new multi-queue block IO queueing mechanism)
  Cc: Jens Axboe
  Signed-off-by: Andrey Vagin
  Signed-off-by: Jens Axboe

  Andrey Vagin

  2013-12-07 00:18:02 +0800  

25 Oct, 2013

1 commit

• 320ae51fe   blk-mq: new multi-queue block IO queueing mechanism ... Browse Code »

  2

  Linux currently has two models for block devices:

  - The classic request_fn based approach, where drivers use struct
  request units for IO. The block layer provides various helper
  functionalities to let drivers share code, things like tag
  management, timeout handling, queueing, etc.

  - The "stacked" approach, where a driver squeezes in between the
  block layer and IO submitter. Since this bypasses the IO stack,
  driver generally have to manage everything themselves.

  With drivers being written for new high IOPS devices, the classic
  request_fn based driver doesn't work well enough. The design dates
  back to when both SMP and high IOPS was rare. It has problems with
  scaling to bigger machines, and runs into scaling issues even on
  smaller machines when you have IOPS in the hundreds of thousands
  per device.

  The stacked approach is then most often selected as the model
  for the driver. But this means that everybody has to re-invent
  everything, and along with that we get all the problems again
  that the shared approach solved.

  This commit introduces blk-mq, block multi queue support. The
  design is centered around per-cpu queues for queueing IO, which
  then funnel down into x number of hardware submission queues.
  We might have a 1:1 mapping between the two, or it might be
  an N:M mapping. That all depends on what the hardware supports.

  blk-mq provides various helper functions, which include:

  - Scalable support for request tagging. Most devices need to
  be able to uniquely identify a request both in the driver and
  to the hardware. The tagging uses per-cpu caches for freed
  tags, to enable cache hot reuse.

  - Timeout handling without tracking request on a per-device
  basis. Basically the driver should be able to get a notification,
  if a request happens to fail.

  - Optional support for non 1:1 mappings between issue and
  submission queues. blk-mq can redirect IO completions to the
  desired location.

  - Support for per-request payloads. Drivers almost always need
  to associate a request structure with some driver private
  command structure. Drivers can tell blk-mq this at init time,
  and then any request handed to the driver will have the
  required size of memory associated with it.

  - Support for merging of IO, and plugging. The stacked model
  gets neither of these. Even for high IOPS devices, merging
  sequential IO reduces per-command overhead and thus
  increases bandwidth.

  For now, this is provided as a potential 3rd queueing model, with
  the hope being that, as it matures, it can replace both the classic
  and stacked model. That would get us back to having just 1 real
  model for block devices, leaving the stacked approach to dm/md
  devices (as it was originally intended).

  Contributions in this patch from the following people:

  Shaohua Li
  Alexander Gordeev
  Christoph Hellwig
  Mike Christie
  Matias Bjorling
  Jeff Moyer

  Acked-by: Christoph Hellwig
  Signed-off-by: Jens Axboe

  Jens Axboe

  2013-10-25 18:56:00 +0800