linux-scsi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCHv4 0/5] scsi: use xarray for devices and targets
@ 2020-05-29 13:47 Hannes Reinecke
  2020-05-29 13:47 ` [PATCH 1/5] scsi: convert target lookup to xarray Hannes Reinecke
                   ` (5 more replies)
  0 siblings, 6 replies; 9+ messages in thread
From: Hannes Reinecke @ 2020-05-29 13:47 UTC (permalink / raw)
  To: Martin K. Petersen
  Cc: Christoph Hellwig, Doug Gilbert, Daniel Wagner,
	Johannes Thumshirn, James Bottomley, linux-scsi, Hannes Reinecke

Hi all,

based on the ideas from Doug Gilbert here's now my take on using
xarrays for devices and targets.
It revolves around two ideas:

 - The scsi target 'channel' and 'id' numbers are never ever used
   to the full 32 bit range; channels are well below 10, and no
   driver is using more than 16 bits for the id. So we can reduce
   the type of 'channel' and 'id' to 16 bits, and use the 32 bit
   value 'channel << 16 | id' as the index into the target xarray.
 - Most SCSI targets are only using the first 32 bits of the 64 bit
   LUN structure. So there we can use the LUN number as the index into
   the xarray; larger LUN numbers won't fit, so we'll allocate a
   separate index for those. For these device lookup won't be as
   efficient, but one can argue that we're running into scalability
   issues long before that.

With these changes we can implement an efficient lookup mechanism,
devolving into direct lookup for most cases.
And iteration over devices should be as efficient as the current,
list-based, approach.

It also removes the current duality between host-based and
target-based device lists; now all devices are listed per target,
and a full iteration would need to iterate first over all targets
and then over all devices per target.

As usual, comments and reviews are welcome

Changes to v1:
- Fixup __scsi_iterate_devices()
- Include reviews from Johannes
- Minor fixes
- Include comments from Doug

Changes to v2:
- Reshuffle hunks as per suggestion from Johannes

Changes to v3:
- Use GFP_ATOMIC instead of GFP_KERNEL
- Fixup target iteration as reported by Doug Gilbert
- Update scsi_error to make use of the new iterators

Hannes Reinecke (5):
  scsi: convert target lookup to xarray
  target_core_pscsi: use __scsi_device_lookup()
  scsi: move target device list to xarray
  scsi: remove direct device lookup per host
  scsi_error: use xarray lookup instead of wrappers

 drivers/scsi/hosts.c               |   4 +-
 drivers/scsi/scsi.c                | 151 +++++++++++++++++++++++++++++--------
 drivers/scsi/scsi_error.c          |  35 +++++----
 drivers/scsi/scsi_lib.c            |   9 +--
 drivers/scsi/scsi_priv.h           |   2 +
 drivers/scsi/scsi_scan.c           |  73 +++++++++---------
 drivers/scsi/scsi_sysfs.c          |  48 ++++++++----
 drivers/target/target_core_pscsi.c |   8 +-
 include/scsi/scsi_device.h         |  32 +++++---
 include/scsi/scsi_host.h           |   5 +-
 10 files changed, 243 insertions(+), 124 deletions(-)

-- 
2.16.4


^ permalink raw reply	[flat|nested] 9+ messages in thread

end of thread, other threads:[~2020-05-31  8:04 UTC | newest]

Thread overview: 9+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2020-05-29 13:47 [PATCHv4 0/5] scsi: use xarray for devices and targets Hannes Reinecke
2020-05-29 13:47 ` [PATCH 1/5] scsi: convert target lookup to xarray Hannes Reinecke
2020-05-29 13:47 ` [PATCH 2/5] target_core_pscsi: use __scsi_device_lookup() Hannes Reinecke
2020-05-29 22:06   ` Mike Christie
2020-05-29 13:47 ` [PATCH 3/5] scsi: move target device list to xarray Hannes Reinecke
2020-05-29 13:47 ` [PATCH 4/5] scsi: remove direct device lookup per host Hannes Reinecke
2020-05-29 13:47 ` [PATCH 5/5] scsi_error: use xarray lookup instead of wrappers Hannes Reinecke
2020-05-31  6:58   ` kbuild test robot
2020-05-29 20:53 ` [PATCHv4 0/5] scsi: use xarray for devices and targets Douglas Gilbert

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).