public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Oded Gabbay <oded.gabbay@gmail.com>
To: gregkh@linuxfoundation.org
Cc: linux-kernel@vger.kernel.org
Subject: [git pull] habanalabs pull request for kernel 5.5
Date: Thu, 21 Nov 2019 16:42:23 +0200	[thread overview]
Message-ID: <20191121144223.GA14312@ogabbay-VM> (raw)

Hello Greg,

This is the pull request for habanalabs driver for kernel 5.5.

It mainly contains improvements to MMU and reset code, and exposes more
information through our INFO IOCTL.

Please see the tag message for more details on what this pull request
contains.

Thanks,
Oded

The following changes since commit ab64ec1db25e0cceab0bad15b03fd57e2b461b15:

  misc: Fix Kconfig indentation (2019-11-20 15:09:49 +0100)

are available in the Git repository at:

  git://people.freedesktop.org/~gabbayo/linux tags/misc-habanalabs-next-2019-11-21

for you to fetch changes up to 5feccddcf9922ee3c25587d5e609bf58503ad93e:

  habanalabs: add more protection of device during reset (2019-11-21 11:35:47 +0200)

----------------------------------------------------------------
This tag contains the following changes for kernel 5.5:

- MMU code improvements that includes:
  - Distinguish between "normal" unmapping and unmapping that is done as
    part of the tear-down of a user process. This improves performance of
    unmapping during reset of the device.
  - Add future ASIC support in generic MMU code.

- Improve device reset code by adding more protection around accessing the
  device during the reset process.

- Add new H/W queue type for future ASIC support

- Add more information to be retrieved by users through INFO IOCTL:
  - clock rate
  - board name
  - reset counters

- Small bug fixes and minor improvements to code.

----------------------------------------------------------------
Moti Haimovski (1):
      habanalabs: expose reset counters via existing INFO IOCTL

Oded Gabbay (16):
      habanalabs: handle F/W failure for sensor initialization
      habanalabs: set TPC Icache to 16 cache lines
      habanalabs: add opcode to INFO IOCTL to return clock rate
      habanalabs: expose card name in INFO IOCTL
      habanalabs: read F/W versions before failure
      habanalabs: use registers name defines for ETR block
      habanalabs: set ETR as non-secured
      habanalabs: increase max jobs number to 512
      habanalabs: don't print error when queues are full
      habanalabs: export uapi defines to user-space
      habanalabs: remove prints on successful device initialization
      habanalabs: use defines for F/W files
      habanalabs: make code more concise
      habanalabs: make the reset code more consistent
      habanalabs: flush EQ workers in hard reset
      habanalabs: add more protection of device during reset

Omer Shpigelman (9):
      habanalabs: re-factor memory module code
      habanalabs: type specific MMU cache invalidation
      habanalabs: re-factor MMU masks and documentation
      habanalabs: split MMU properties to PCI/DRAM
      habanalabs: prevent read/write from/to the device during hard reset
      habanalabs: optimize MMU unmap
      habanalabs: skip VA block list update in reset flow
      habanalabs: invalidate MMU cache only once
      habanalabs: remove unnecessary checks

Tomer Tayar (3):
      habanalabs: Fix typos
      habanalabs: Mark queue as expecting CB handle or address
      habanalabs: Add a new H/W queue type

YueHaibing (2):
      habanalabs: remove set but not used variable 'ctx'
      habanalabs: remove set but not used variable 'qman_base_addr'

 drivers/misc/habanalabs/command_submission.c       | 127 ++++---
 drivers/misc/habanalabs/debugfs.c                  | 112 ++++--
 drivers/misc/habanalabs/device.c                   |  18 +-
 drivers/misc/habanalabs/firmware_if.c              |   5 +-
 drivers/misc/habanalabs/goya/goya.c                |  78 ++--
 drivers/misc/habanalabs/goya/goyaP.h               |   2 +
 drivers/misc/habanalabs/goya/goya_coresight.c      |  53 +--
 drivers/misc/habanalabs/goya/goya_hwmgr.c          |  31 ++
 drivers/misc/habanalabs/habanalabs.h               | 171 +++++----
 drivers/misc/habanalabs/habanalabs_ioctl.c         |  73 +++-
 drivers/misc/habanalabs/hw_queue.c                 | 249 ++++++++++---
 .../habanalabs/include/goya/asic_reg/goya_masks.h  |   2 +
 .../habanalabs/include/goya/asic_reg/goya_regs.h   |   1 +
 .../include/goya/asic_reg/psoc_etr_regs.h          | 114 ++++++
 drivers/misc/habanalabs/include/hl_boot_if.h       |   2 +
 .../habanalabs/include/hw_ip/mmu/mmu_general.h     |   7 +-
 drivers/misc/habanalabs/include/qman_if.h          |  12 +
 drivers/misc/habanalabs/memory.c                   | 392 +++++++++++----------
 drivers/misc/habanalabs/mmu.c                      | 204 ++++++-----
 include/uapi/misc/habanalabs.h                     |  48 ++-
 20 files changed, 1159 insertions(+), 542 deletions(-)
 create mode 100644 drivers/misc/habanalabs/include/goya/asic_reg/psoc_etr_regs.h

             reply	other threads:[~2019-11-21 14:42 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-21 14:42 Oded Gabbay [this message]
2019-11-21 16:47 ` [git pull] habanalabs pull request for kernel 5.5 Greg KH

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20191121144223.GA14312@ogabbay-VM \
    --to=oded.gabbay@gmail.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox