Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Christian König" <christian.koenig@amd.com>
To: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>,
	intel-xe@lists.freedesktop.org
Cc: Somalapuram Amaranath <Amaranath.Somalapuram@amd.com>,
	dri-devel@lists.freedesktop.org,
	Matthew Brost <matthew.brost@intel.com>,
	"Kuehling, Felix" <Felix.Kuehling@amd.com>
Subject: Re: [PATCH v2 0/9] TTM shrinker helpers and xe buffer object shrinker
Date: Tue, 16 Apr 2024 13:55:24 +0200	[thread overview]
Message-ID: <bb679fc6-7890-4b4a-a8cd-8ae7ce087a9f@amd.com> (raw)
In-Reply-To: <20240416100730.6666-1-thomas.hellstrom@linux.intel.com>

While patches 1-4 look good from a high level I still think it needs 
some prerequisite and re-ordering.

First of all make all the cleanups separate patches. In other words that 
ttm_resource_manager_next() takes only the cursor as argument, adding 
ttm_resource_cursor_fini()/ttm_resource_cursor_fini_locked() as one 
patch and then ttm_lru_bulk_move_init()/ttm_lru_bulk_move_fini() as second.

With that done I think we should first switch over TTM and all drivers 
using it to drm_exec as part of it's context object.

Then I would switch over to using LRU hitches for both swapping and 
eviction.

And when that's finally done we can take a look into the partial shmem 
swapping :)

And Felix is really (and mean *really*) looking forward to the partial 
shmem swapping as well.

Regards,
Christian.

Am 16.04.24 um 12:07 schrieb Thomas Hellström:
> This series implements TTM shrinker / eviction helpers and an xe bo
> shrinker. It builds on two previous series, *and obsoletes these*. First
>
> https://www.mail-archive.com/dri-devel@lists.freedesktop.org/msg484425.html
>
> for patch 1-4, which IMO still could be reviewed and pushed as a
> separate series.
>
> Second the previous TTM shrinker series
>
> https://lore.kernel.org/linux-mm/b7491378-defd-4f1c-31e2-29e4c77e2d67@amd.com/T/
>
> Where the comment about layering
> https://lore.kernel.org/linux-mm/b7491378-defd-4f1c-31e2-29e4c77e2d67@amd.com/T/#ma918844aa8a6efe8768fdcda0c6590d5c93850c9
>
> now addressed, and this version also implements shmem objects for backup
> rather than direct swap-cache insertions, which was used in the previuos
> series. It turns out that with per-page backup / shrinking, shmem objects
> appears to work just as well as direct swap-cache insertions with the
> added benefit that was introduced in the previous TTM shrinker series to
> avoid running out of swap entries isn't really needed.
>
> In any case, patch 1-4 are better described in their separate series.
> (RFC is removed for those).
>
> Patch 5 could in theory be skipped but introduces a possibility to easily
> add or test multiple backup backends, like the direct swap-cache
> insertion or even files into fast dedicated nvme storage for for example.
>
> Patch 6 introduces helpers in the ttm_pool code for page-by-page shrinking
> and recovery. It avoids having to temporarily allocate a huge amount of
> memory to be able to shrink a buffer object. It also introduces the
> possibility to immediately write-back pages if needed, since that tends
> to be a bit delayed when left to kswapd.
>
> Patch 7 Adds a simple error injection to the above code to help increase
> test coverage.
>
> Patch 8 introduces a LRU walk helper for eviction and shrinking. It's
> currently xe-only but not xe-specific and can easily be moved to TTM when
> used by more than one driver or when eviction is implemented using it.
>
> Patch 9 introduces a helper callback for shrinking (Also ready to be
> moved to TTM) and an xe-specific shrinker implementation. It also
> adds a kunit test to test the shrinker functionality by trying to
> allocate twice the available amount of RAM as buffer objects. If there
> is no swap-space available, the buffer objects are marked
> purgeable.
>
> v2:
> - Squash obsolete revision history in the patch commit messages.
> - Fix a couple of review comments by Christian
> - Don't store the mem_type in the TTM managers but in the
>    resource cursor.
> - Rename introduced TTM *back_up* function names to *backup*
> - Add ttm pool recovery fault injection.
> - Shrinker xe kunit test
> - Various bugfixes
>
> Cc: Somalapuram Amaranath <Amaranath.Somalapuram@amd.com>
> Cc: Christian König <christian.koenig@amd.com>
> Cc: <dri-devel@lists.freedesktop.org>
>
> Thomas Hellström (8):
>    drm/ttm: Allow TTM LRU list nodes of different types
>    drm/ttm: Use LRU hitches
>    drm/ttm, drm/amdgpu, drm/xe: Consider hitch moves within bulk sublist
>      moves
>    drm/ttm: Allow continued swapout after -ENOSPC falure
>    drm/ttm: Add a virtual base class for graphics memory backup
>    drm/ttm/pool: Provide a helper to shrink pages.
>    drm/xe, drm/ttm: Provide a generic LRU walker helper
>    drm/xe: Add a shrinker for xe bos
>
>   drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c |   4 +
>   drivers/gpu/drm/ttm/Makefile           |   2 +-
>   drivers/gpu/drm/ttm/ttm_backup_shmem.c | 137 +++++++++
>   drivers/gpu/drm/ttm/ttm_bo.c           |   1 +
>   drivers/gpu/drm/ttm/ttm_device.c       |  33 ++-
>   drivers/gpu/drm/ttm/ttm_pool.c         | 391 ++++++++++++++++++++++++-
>   drivers/gpu/drm/ttm/ttm_resource.c     | 231 ++++++++++++---
>   drivers/gpu/drm/ttm/ttm_tt.c           |  34 +++
>   drivers/gpu/drm/xe/Makefile            |   2 +
>   drivers/gpu/drm/xe/xe_bo.c             | 123 ++++++--
>   drivers/gpu/drm/xe/xe_bo.h             |   3 +
>   drivers/gpu/drm/xe/xe_device.c         |   8 +
>   drivers/gpu/drm/xe/xe_device_types.h   |   2 +
>   drivers/gpu/drm/xe/xe_shrinker.c       | 237 +++++++++++++++
>   drivers/gpu/drm/xe/xe_shrinker.h       |  18 ++
>   drivers/gpu/drm/xe/xe_ttm_helpers.c    | 224 ++++++++++++++
>   drivers/gpu/drm/xe/xe_ttm_helpers.h    |  63 ++++
>   drivers/gpu/drm/xe/xe_vm.c             |   4 +
>   include/drm/ttm/ttm_backup.h           | 136 +++++++++
>   include/drm/ttm/ttm_device.h           |   2 +
>   include/drm/ttm/ttm_pool.h             |   4 +
>   include/drm/ttm/ttm_resource.h         |  96 +++++-
>   include/drm/ttm/ttm_tt.h               |  19 ++
>   23 files changed, 1683 insertions(+), 91 deletions(-)
>   create mode 100644 drivers/gpu/drm/ttm/ttm_backup_shmem.c
>   create mode 100644 drivers/gpu/drm/xe/xe_shrinker.c
>   create mode 100644 drivers/gpu/drm/xe/xe_shrinker.h
>   create mode 100644 drivers/gpu/drm/xe/xe_ttm_helpers.c
>   create mode 100644 drivers/gpu/drm/xe/xe_ttm_helpers.h
>   create mode 100644 include/drm/ttm/ttm_backup.h
>


  parent reply	other threads:[~2024-04-16 11:55 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-16 10:07 [PATCH v2 0/9] TTM shrinker helpers and xe buffer object shrinker Thomas Hellström
2024-04-16 10:07 ` [PATCH v2 1/9] drm/ttm: Allow TTM LRU list nodes of different types Thomas Hellström
2024-04-17  1:15   ` Matthew Brost
2024-04-17  6:09     ` Christian König
2024-05-02 11:41       ` Thomas Hellström
2024-04-16 10:07 ` [PATCH v2 2/9] drm/ttm: Use LRU hitches Thomas Hellström
2024-04-26  1:23   ` Matthew Brost
2024-04-16 10:07 ` [PATCH v2 3/9] drm/ttm, drm/amdgpu, drm/xe: Consider hitch moves within bulk sublist moves Thomas Hellström
2024-04-26  1:20   ` Matthew Brost
2024-04-16 10:07 ` [PATCH v2 4/9] drm/ttm: Allow continued swapout after -ENOSPC falure Thomas Hellström
2024-04-17  1:44   ` Matthew Brost
2024-04-16 10:07 ` [PATCH v2 5/9] drm/ttm: Add a virtual base class for graphics memory backup Thomas Hellström
2024-04-16 10:07 ` [PATCH v2 6/9] drm/ttm/pool: Provide a helper to shrink pages Thomas Hellström
2024-04-16 10:07 ` [PATCH v2 7/9] drm/ttm: Use fault-injection to test error paths Thomas Hellström
2024-04-16 10:07 ` [PATCH v2 8/9] drm/xe, drm/ttm: Provide a generic LRU walker helper Thomas Hellström
2024-04-16 10:07 ` [PATCH v2 9/9] drm/xe: Add a shrinker for xe bos Thomas Hellström
2024-04-16 21:32   ` kernel test robot
2024-04-17 13:45   ` kernel test robot
2024-04-16 10:33 ` ✓ CI.Patch_applied: success for TTM shrinker helpers and xe buffer object shrinker (rev2) Patchwork
2024-04-16 10:33 ` ✗ CI.checkpatch: warning " Patchwork
2024-04-16 10:35 ` ✓ CI.KUnit: success " Patchwork
2024-04-16 10:46 ` ✓ CI.Build: " Patchwork
2024-04-16 10:49 ` ✓ CI.Hooks: " Patchwork
2024-04-16 10:50 ` ✗ CI.checksparse: warning " Patchwork
2024-04-16 11:37 ` ✓ CI.BAT: success " Patchwork
2024-04-16 11:55 ` Christian König [this message]
2024-04-16 13:08   ` [PATCH v2 0/9] TTM shrinker helpers and xe buffer object shrinker Thomas Hellström
2024-04-16 13:24     ` Christian König
2024-04-17  8:37 ` ✗ CI.FULL: failure for TTM shrinker helpers and xe buffer object shrinker (rev2) Patchwork

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=bb679fc6-7890-4b4a-a8cd-8ae7ce087a9f@amd.com \
    --to=christian.koenig@amd.com \
    --cc=Amaranath.Somalapuram@amd.com \
    --cc=Felix.Kuehling@amd.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=matthew.brost@intel.com \
    --cc=thomas.hellstrom@linux.intel.com \
    /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