[git pull] habanalabs pull request for kernel 5.9-rc1 (resend)

[git pull] habanalabs pull request for kernel 5.9-rc1 (resend)

[Oded Gabbay]

Hello Greg,

(Re-sending this pull request)

This is habanalabs pull request for the merge window of kernel 5.9. It
contains many small improvements to common and GAUDI code. Details are in
the tag.

Thanks,
Oded

The following changes since commit 7a4462a96777b64b22412f782de226c90290bf75:

  misc: rtsx: Use standard PCI definitions (2020-07-22 13:39:31 +0200)

are available in the Git repository at:

  git://people.freedesktop.org/~gabbayo/linux tags/misc-habanalabs-next-2020-07-23

for you to fetch changes up to 961a7325e6701eeb35ae46142aabc5732004bc77:

  habanalabs: Fix memory leak in error flow of context init (2020-07-23 22:07:02 +0300)

----------------------------------------------------------------
This tag contains the following changes for kernel 5.9-rc1:

- Remove rate limiters from GAUDI configuration (no longer needed).
- Set maximum amount of in-flight CS per ASIC type and increase
  the maximum amount for GAUDI.
- Refactor signal/wait command submissions code
- Calculate trace frequency from PLLs to show accurate profiling data
- Rephrase error messages to make them more clear to the common user
- Add statistics of dropped CS (counter per possible reason for drop)
- Get ECC information from firmware
- Remove support for partial SoC reset in Gaudi
- Halt device CPU only when reset is certain to happen. Sometimes we abort
  the reset procedure and in that case we can't leave device CPU in halt
  mode.
- set each CQ to its own work queue to prevent a race between completions
  on different CQs.
- Use queue pi/ci in order to determine queue occupancy. This is done to
  make the code reusable between current and future ASICs.
- Add more validations for user inputs.
- Refactor PCIe controller configuration to make the code reusable between
  current and future ASICs.
- Update firmware interface headers to latest version
- Move all common code to a dedicated common sub-folder

----------------------------------------------------------------
Adam Aharon (1):
      habanalabs: calculate trace frequency from PLL

Christine Gharzuzi (1):
      habanalabs: extract cpu boot status lookup

Moti Haimovski (1):
      habanalabs: check for DMA errors when clearing memory

Oded Gabbay (10):
      habanalabs: remove rate limiters from GAUDI
      uapi/habanalabs: fix some comments
      habanalabs: align armcp_packet structure to 8 bytes
      habanalabs: rephrase error messages
      habanalabs: Extract ECC information from FW
      habanalabs: remove soft-reset support from GAUDI
      habanalabs: halt device CPU only upon certain reset
      habanalabs: create common folder
      habanalabs: update hl_boot_if.h from firmware
      habanalabs: enable device before hw_init()

Ofir Bitton (11):
      habanalabs: Use pending CS amount per ASIC
      habanalabs: sync stream generic functionality
      habanalabs: Use mask instead of shift in sync stream registers
      habanalabs: Increase queues depth
      habanalabs: Add dropped cs statistics info struct
      habanalabs: PCIe iATU refactoring
      habanalabs: configure maximum queues per asic
      habanalabs: use queue pi/ci in order to determine queue occupancy
      habanalabs: Assign each CQ with its own work queue
      habanalabs: verify queue can contain all cs jobs
      habanalabs: create internal CB pool

Omer Shpigelman (2):
      habanalabs: rephrase error message
      habanalabs: remove unused hash

Tomer Tayar (2):
      habanalabs: use no flags on MMU cache invalidation
      habanalabs: Fix memory leak in error flow of context init

 drivers/misc/habanalabs/Makefile                   |  11 +-
 drivers/misc/habanalabs/common/Makefile            |   9 +
 drivers/misc/habanalabs/{ => common}/asid.c        |   0
 .../misc/habanalabs/{ => common}/command_buffer.c  |  82 +-
 .../habanalabs/{ => common}/command_submission.c   |  97 ++-
 drivers/misc/habanalabs/{ => common}/context.c     |  39 +-
 drivers/misc/habanalabs/{ => common}/debugfs.c     |   0
 drivers/misc/habanalabs/{ => common}/device.c      |  88 +-
 drivers/misc/habanalabs/{ => common}/firmware_if.c | 101 +--
 drivers/misc/habanalabs/{ => common}/habanalabs.h  | 172 ++--
 .../misc/habanalabs/{ => common}/habanalabs_drv.c  |   1 -
 .../habanalabs/{ => common}/habanalabs_ioctl.c     |  24 +
 drivers/misc/habanalabs/{ => common}/hw_queue.c    | 165 ++--
 drivers/misc/habanalabs/{ => common}/hwmon.c       |   0
 drivers/misc/habanalabs/{ => common}/irq.c         |  13 +-
 drivers/misc/habanalabs/{ => common}/memory.c      |   3 +-
 drivers/misc/habanalabs/{ => common}/mmu.c         |   1 -
 drivers/misc/habanalabs/{ => common}/pci.c         | 136 ++--
 drivers/misc/habanalabs/{ => common}/sysfs.c       |   3 +
 drivers/misc/habanalabs/gaudi/Makefile             |   2 +-
 drivers/misc/habanalabs/gaudi/gaudi.c              | 894 ++++++++-------------
 drivers/misc/habanalabs/gaudi/gaudiP.h             |  16 +-
 drivers/misc/habanalabs/gaudi/gaudi_coresight.c    |   6 +-
 drivers/misc/habanalabs/goya/goya.c                | 172 ++--
 drivers/misc/habanalabs/goya/goyaP.h               |  14 +-
 drivers/misc/habanalabs/goya/goya_coresight.c      |   6 +-
 .../habanalabs/include/{ => common}/armcp_if.h     |  14 +-
 .../habanalabs/include/{ => common}/hl_boot_if.h   |  14 +
 .../misc/habanalabs/include/{ => common}/qman_if.h |   0
 .../habanalabs/include/gaudi/asic_reg/gaudi_regs.h |  21 +-
 .../include/gaudi/asic_reg/psoc_cpu_pll_regs.h     | 114 +++
 .../misc/habanalabs/include/gaudi/gaudi_masks.h    |   3 +
 .../misc/habanalabs/include/gaudi/gaudi_packets.h  |   4 +-
 include/uapi/misc/habanalabs.h                     |  27 +-
 34 files changed, 1260 insertions(+), 992 deletions(-)
 create mode 100644 drivers/misc/habanalabs/common/Makefile
 rename drivers/misc/habanalabs/{ => common}/asid.c (100%)
 rename drivers/misc/habanalabs/{ => common}/command_buffer.c (85%)
 rename drivers/misc/habanalabs/{ => common}/command_submission.c (92%)
 rename drivers/misc/habanalabs/{ => common}/context.c (84%)
 rename drivers/misc/habanalabs/{ => common}/debugfs.c (100%)
 rename drivers/misc/habanalabs/{ => common}/device.c (94%)
 rename drivers/misc/habanalabs/{ => common}/firmware_if.c (89%)
 rename drivers/misc/habanalabs/{ => common}/habanalabs.h (94%)
 rename drivers/misc/habanalabs/{ => common}/habanalabs_drv.c (99%)
 rename drivers/misc/habanalabs/{ => common}/habanalabs_ioctl.c (95%)
 rename drivers/misc/habanalabs/{ => common}/hw_queue.c (86%)
 rename drivers/misc/habanalabs/{ => common}/hwmon.c (100%)
 rename drivers/misc/habanalabs/{ => common}/irq.c (95%)
 rename drivers/misc/habanalabs/{ => common}/memory.c (99%)
 rename drivers/misc/habanalabs/{ => common}/mmu.c (99%)
 rename drivers/misc/habanalabs/{ => common}/pci.c (73%)
 rename drivers/misc/habanalabs/{ => common}/sysfs.c (99%)
 rename drivers/misc/habanalabs/include/{ => common}/armcp_if.h (98%)
 rename drivers/misc/habanalabs/include/{ => common}/hl_boot_if.h (84%)
 rename drivers/misc/habanalabs/include/{ => common}/qman_if.h (100%)
 create mode 100644 drivers/misc/habanalabs/include/gaudi/asic_reg/psoc_cpu_pll_regs.h

Re: [git pull] habanalabs pull request for kernel 5.9-rc1 (resend)

[Greg KH]

On Thu, Jul 23, 2020 at 10:15:25PM +0300, Oded Gabbay wrote:
> Hello Greg,
> 
> (Re-sending this pull request)
> 
> This is habanalabs pull request for the merge window of kernel 5.9. It
> contains many small improvements to common and GAUDI code. Details are in
> the tag.
> 
> Thanks,
> Oded
> 
> The following changes since commit 7a4462a96777b64b22412f782de226c90290bf75:
> 
>   misc: rtsx: Use standard PCI definitions (2020-07-22 13:39:31 +0200)
> 
> are available in the Git repository at:
> 
>   git://people.freedesktop.org/~gabbayo/linux tags/misc-habanalabs-next-2020-07-23

{sigh}

Commit: 961a7325e670 ("habanalabs: Fix memory leak in error flow of context init")
	Fixes tag: Fixes: 786c94810698 ("habanalabs: Use pending cs amount per asic")
	Has these problem(s):
	        - Subject does not match target commit subject
	          Just use
		                git log -1 --format='Fixes: %h ("%s")'

The scripts I use to check this are here:
	https://github.com/gregkh/gregkh-linux/blob/master/work/verify_fixes.sh
	https://github.com/gregkh/gregkh-linux/blob/master/work/verify_signedoff.sh

And note, if I ignore these errors, you will just get the same response
from Stephen when it hits linux-next :(

thanks,

greg k-h

Re: [git pull] habanalabs pull request for kernel 5.9-rc1 (resend)

[Oded Gabbay]

On Fri, Jul 24, 2020 at 12:32 PM Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Thu, Jul 23, 2020 at 10:15:25PM +0300, Oded Gabbay wrote:
> > Hello Greg,
> >
> > (Re-sending this pull request)
> >
> > This is habanalabs pull request for the merge window of kernel 5.9. It
> > contains many small improvements to common and GAUDI code. Details are in
> > the tag.
> >
> > Thanks,
> > Oded
> >
> > The following changes since commit 7a4462a96777b64b22412f782de226c90290bf75:
> >
> >   misc: rtsx: Use standard PCI definitions (2020-07-22 13:39:31 +0200)
> >
> > are available in the Git repository at:
> >
> >   git://people.freedesktop.org/~gabbayo/linux tags/misc-habanalabs-next-2020-07-23
>
> {sigh}
>
> Commit: 961a7325e670 ("habanalabs: Fix memory leak in error flow of context init")
>         Fixes tag: Fixes: 786c94810698 ("habanalabs: Use pending cs amount per asic")
>         Has these problem(s):
>                 - Subject does not match target commit subject
>                   Just use
>                                 git log -1 --format='Fixes: %h ("%s")'
>
> The scripts I use to check this are here:
>         https://github.com/gregkh/gregkh-linux/blob/master/work/verify_fixes.sh
>         https://github.com/gregkh/gregkh-linux/blob/master/work/verify_signedoff.sh
>
> And note, if I ignore these errors, you will just get the same response
> from Stephen when it hits linux-next :(
>
> thanks,
>
> greg k-h

I'm sorry for all the trouble Greg, but I wanted to ask before I send
it again, won't the fixed sha-id be wrong once you merge this pull
request to your tree ?
Because this patch fixes a previous patch in this pull-request.
Maybe I should just squash them ?

Thanks,
Oded

Re: [git pull] habanalabs pull request for kernel 5.9-rc1 (resend)

[Greg KH]

On Fri, Jul 24, 2020 at 02:35:17PM +0300, Oded Gabbay wrote:
> On Fri, Jul 24, 2020 at 12:32 PM Greg KH <gregkh@linuxfoundation.org> wrote:
> >
> > On Thu, Jul 23, 2020 at 10:15:25PM +0300, Oded Gabbay wrote:
> > > Hello Greg,
> > >
> > > (Re-sending this pull request)
> > >
> > > This is habanalabs pull request for the merge window of kernel 5.9. It
> > > contains many small improvements to common and GAUDI code. Details are in
> > > the tag.
> > >
> > > Thanks,
> > > Oded
> > >
> > > The following changes since commit 7a4462a96777b64b22412f782de226c90290bf75:
> > >
> > >   misc: rtsx: Use standard PCI definitions (2020-07-22 13:39:31 +0200)
> > >
> > > are available in the Git repository at:
> > >
> > >   git://people.freedesktop.org/~gabbayo/linux tags/misc-habanalabs-next-2020-07-23
> >
> > {sigh}
> >
> > Commit: 961a7325e670 ("habanalabs: Fix memory leak in error flow of context init")
> >         Fixes tag: Fixes: 786c94810698 ("habanalabs: Use pending cs amount per asic")
> >         Has these problem(s):
> >                 - Subject does not match target commit subject
> >                   Just use
> >                                 git log -1 --format='Fixes: %h ("%s")'
> >
> > The scripts I use to check this are here:
> >         https://github.com/gregkh/gregkh-linux/blob/master/work/verify_fixes.sh
> >         https://github.com/gregkh/gregkh-linux/blob/master/work/verify_signedoff.sh
> >
> > And note, if I ignore these errors, you will just get the same response
> > from Stephen when it hits linux-next :(
> >
> > thanks,
> >
> > greg k-h
> 
> I'm sorry for all the trouble Greg, but I wanted to ask before I send
> it again, won't the fixed sha-id be wrong once you merge this pull
> request to your tree ?

Nope, that's because I am using git to merge it.

> Because this patch fixes a previous patch in this pull-request.
> Maybe I should just squash them ?

That's up to you.  If it was a different author, perhaps keep them
separate, it's your call.

thanks,

greg k-h