[PATCH] drm/i915/gem: Pin the L-shape quirked object as unshrinkable

[PATCH] drm/i915/gem: Pin the L-shape quirked object as unshrinkable

[Matthew Auld]

From: Chris Wilson <chris@chris-wilson.co.uk>

When instantiating a tiled object on an L-shaped memory machine, we mark
the object as unshrinkable to prevent the shrinker from trying to swap
out the pages. We have to do this as we do not know the swizzling on the
individual pages, and so the data will be scrambled across swap out/in.

Not only do we need to move the object off the shrinker list, we need to
mark the object with shrink_pin so that the counter is consistent across
calls to madvise.

Fixes: 0175969e489a ("drm/i915/gem: Use shrinkable status for unknown swizzle quirks")
References: https://gitlab.freedesktop.org/drm/intel/-/issues/3293
Reported-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
---
 drivers/gpu/drm/i915/gem/i915_gem_pages.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index aed8a37ccdc9..7361971c177d 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -63,6 +63,8 @@ void __i915_gem_object_set_pages(struct drm_i915_gem_object *obj,
 	    i915->quirks & QUIRK_PIN_SWIZZLED_PAGES) {
 		GEM_BUG_ON(i915_gem_object_has_tiling_quirk(obj));
 		i915_gem_object_set_tiling_quirk(obj);
+		GEM_BUG_ON(!list_empty(&obj->mm.link));
+		atomic_inc(&obj->mm.shrink_pin);
 		shrinkable = false;
 	}
 
-- 
2.26.3

[Intel-gfx] [PATCH] drm/i915/gem: Pin the L-shape quirked object as unshrinkable

[Matthew Auld]

From: Chris Wilson <chris@chris-wilson.co.uk>

When instantiating a tiled object on an L-shaped memory machine, we mark
the object as unshrinkable to prevent the shrinker from trying to swap
out the pages. We have to do this as we do not know the swizzling on the
individual pages, and so the data will be scrambled across swap out/in.

Not only do we need to move the object off the shrinker list, we need to
mark the object with shrink_pin so that the counter is consistent across
calls to madvise.

Fixes: 0175969e489a ("drm/i915/gem: Use shrinkable status for unknown swizzle quirks")
References: https://gitlab.freedesktop.org/drm/intel/-/issues/3293
Reported-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
---
 drivers/gpu/drm/i915/gem/i915_gem_pages.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index aed8a37ccdc9..7361971c177d 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -63,6 +63,8 @@ void __i915_gem_object_set_pages(struct drm_i915_gem_object *obj,
 	    i915->quirks & QUIRK_PIN_SWIZZLED_PAGES) {
 		GEM_BUG_ON(i915_gem_object_has_tiling_quirk(obj));
 		i915_gem_object_set_tiling_quirk(obj);
+		GEM_BUG_ON(!list_empty(&obj->mm.link));
+		atomic_inc(&obj->mm.shrink_pin);
 		shrinkable = false;
 	}
 
-- 
2.26.3

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915/gem: Pin the L-shape quirked object as unshrinkable

[Patchwork]

[-- Attachment #1.1: Type: text/plain, Size: 12499 bytes --]

== Series Details ==

Series: drm/i915/gem: Pin the L-shape quirked object as unshrinkable
URL   : https://patchwork.freedesktop.org/series/90065/
State : failure

== Summary ==

CI Bug Log - changes from CI_DRM_10073 -> Patchwork_20110
====================================================

Summary
-------

  **FAILURE**

  Serious unknown changes coming with Patchwork_20110 absolutely need to be
  verified manually.
  
  If you think the reported changes have nothing to do with the changes
  introduced in Patchwork_20110, please notify your bug team to allow them
  to document this new failure mode, which will reduce false positives in CI.

  External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/index.html

Possible new issues
-------------------

  Here are the unknown changes that may have been introduced in Patchwork_20110:

### IGT changes ###

#### Possible regressions ####

  * igt@kms_pipe_crc_basic@hang-read-crc-pipe-a:
    - fi-pnv-d510:        [PASS][1] -> [FAIL][2]
   [1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-pnv-d510/igt@kms_pipe_crc_basic@hang-read-crc-pipe-a.html
   [2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-pnv-d510/igt@kms_pipe_crc_basic@hang-read-crc-pipe-a.html

  * igt@kms_pipe_crc_basic@nonblocking-crc-pipe-a-frame-sequence:
    - fi-ilk-650:         [PASS][3] -> [FAIL][4] +6 similar issues
   [3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-ilk-650/igt@kms_pipe_crc_basic@nonblocking-crc-pipe-a-frame-sequence.html
   [4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-ilk-650/igt@kms_pipe_crc_basic@nonblocking-crc-pipe-a-frame-sequence.html

  * igt@kms_pipe_crc_basic@read-crc-pipe-b:
    - fi-elk-e7500:       [PASS][5] -> [FAIL][6] +1 similar issue
   [5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-elk-e7500/igt@kms_pipe_crc_basic@read-crc-pipe-b.html
   [6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-elk-e7500/igt@kms_pipe_crc_basic@read-crc-pipe-b.html

  * igt@kms_pipe_crc_basic@suspend-read-crc-pipe-a:
    - fi-bsw-kefka:       [PASS][7] -> [FAIL][8]
   [7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bsw-kefka/igt@kms_pipe_crc_basic@suspend-read-crc-pipe-a.html
   [8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bsw-kefka/igt@kms_pipe_crc_basic@suspend-read-crc-pipe-a.html

  
#### Suppressed ####

  The following results come from untrusted machines, tests, or statuses.
  They do not affect the overall result.

  * igt@runner@aborted:
    - {fi-rkl-11500t}:    NOTRUN -> [FAIL][9]
   [9]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-rkl-11500t/igt@runner@aborted.html

  
Known issues
------------

  Here are the changes found in Patchwork_20110 that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@gem_busy@busy@all:
    - fi-bxt-dsi:         [PASS][10] -> [FAIL][11] ([i915#3457])
   [10]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bxt-dsi/igt@gem_busy@busy@all.html
   [11]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bxt-dsi/igt@gem_busy@busy@all.html

  * igt@gem_exec_fence@nb-await@vcs0:
    - fi-bsw-kefka:       [PASS][12] -> [FAIL][13] ([i915#3457])
   [12]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bsw-kefka/igt@gem_exec_fence@nb-await@vcs0.html
   [13]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bsw-kefka/igt@gem_exec_fence@nb-await@vcs0.html
    - fi-bsw-n3050:       [PASS][14] -> [FAIL][15] ([i915#3457])
   [14]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bsw-n3050/igt@gem_exec_fence@nb-await@vcs0.html
   [15]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bsw-n3050/igt@gem_exec_fence@nb-await@vcs0.html

  * igt@gem_exec_fence@nb-await@vecs0:
    - fi-bsw-nick:        [PASS][16] -> [FAIL][17] ([i915#3457]) +2 similar issues
   [16]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bsw-nick/igt@gem_exec_fence@nb-await@vecs0.html
   [17]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bsw-nick/igt@gem_exec_fence@nb-await@vecs0.html

  * igt@gem_wait@busy@all:
    - fi-bsw-kefka:       [PASS][18] -> [FAIL][19] ([i915#3177] / [i915#3457])
   [18]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bsw-kefka/igt@gem_wait@busy@all.html
   [19]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bsw-kefka/igt@gem_wait@busy@all.html

  * igt@kms_frontbuffer_tracking@basic:
    - fi-icl-u2:          [PASS][20] -> [FAIL][21] ([i915#49])
   [20]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-icl-u2/igt@kms_frontbuffer_tracking@basic.html
   [21]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-icl-u2/igt@kms_frontbuffer_tracking@basic.html

  
#### Possible fixes ####

  * igt@gem_busy@busy@all:
    - fi-bsw-kefka:       [FAIL][22] ([i915#3457]) -> [PASS][23] +1 similar issue
   [22]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bsw-kefka/igt@gem_busy@busy@all.html
   [23]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bsw-kefka/igt@gem_busy@busy@all.html

  * igt@gem_exec_fence@nb-await@rcs0:
    - fi-elk-e7500:       [FAIL][24] ([i915#3457]) -> [PASS][25]
   [24]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-elk-e7500/igt@gem_exec_fence@nb-await@rcs0.html
   [25]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-elk-e7500/igt@gem_exec_fence@nb-await@rcs0.html

  * igt@gem_exec_fence@nb-await@vecs0:
    - fi-bsw-n3050:       [FAIL][26] ([i915#3457]) -> [PASS][27]
   [26]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bsw-n3050/igt@gem_exec_fence@nb-await@vecs0.html
   [27]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bsw-n3050/igt@gem_exec_fence@nb-await@vecs0.html

  * igt@gem_wait@busy@all:
    - fi-apl-guc:         [FAIL][28] ([i915#3457]) -> [PASS][29]
   [28]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-apl-guc/igt@gem_wait@busy@all.html
   [29]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-apl-guc/igt@gem_wait@busy@all.html

  * igt@gem_wait@wait@all:
    - fi-bsw-nick:        [FAIL][30] ([i915#3457]) -> [PASS][31] +2 similar issues
   [30]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bsw-nick/igt@gem_wait@wait@all.html
   [31]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bsw-nick/igt@gem_wait@wait@all.html

  * igt@kms_busy@basic@modeset:
    - fi-ilk-650:         [INCOMPLETE][32] ([i915#3457]) -> [PASS][33]
   [32]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-ilk-650/igt@kms_busy@basic@modeset.html
   [33]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-ilk-650/igt@kms_busy@basic@modeset.html

  * igt@kms_frontbuffer_tracking@basic:
    - fi-tgl-u2:          [FAIL][34] ([i915#2416]) -> [PASS][35]
   [34]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-tgl-u2/igt@kms_frontbuffer_tracking@basic.html
   [35]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-tgl-u2/igt@kms_frontbuffer_tracking@basic.html

  * igt@kms_pipe_crc_basic@suspend-read-crc-pipe-a:
    - fi-bwr-2160:        [FAIL][36] -> [PASS][37]
   [36]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bwr-2160/igt@kms_pipe_crc_basic@suspend-read-crc-pipe-a.html
   [37]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bwr-2160/igt@kms_pipe_crc_basic@suspend-read-crc-pipe-a.html

  
#### Warnings ####

  * igt@gem_wait@busy@all:
    - fi-bsw-nick:        [FAIL][38] ([i915#3457]) -> [FAIL][39] ([i915#3177] / [i915#3457])
   [38]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bsw-nick/igt@gem_wait@busy@all.html
   [39]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bsw-nick/igt@gem_wait@busy@all.html

  * igt@i915_module_load@reload:
    - fi-bsw-kefka:       [DMESG-WARN][40] ([i915#1982] / [i915#3457]) -> [DMESG-FAIL][41] ([i915#1982] / [i915#3457])
   [40]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bsw-kefka/igt@i915_module_load@reload.html
   [41]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bsw-kefka/igt@i915_module_load@reload.html

  * igt@i915_selftest@live@mman:
    - fi-bwr-2160:        [DMESG-FAIL][42] ([i915#3457]) -> [DMESG-WARN][43] ([i915#3457])
   [42]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bwr-2160/igt@i915_selftest@live@mman.html
   [43]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bwr-2160/igt@i915_selftest@live@mman.html

  * igt@runner@aborted:
    - fi-cfl-8700k:       [FAIL][44] ([i915#2426] / [i915#3363]) -> [FAIL][45] ([i915#3363])
   [44]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-cfl-8700k/igt@runner@aborted.html
   [45]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-cfl-8700k/igt@runner@aborted.html
    - fi-bdw-5557u:       [FAIL][46] -> [FAIL][47] ([i915#1602] / [i915#2029])
   [46]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-bdw-5557u/igt@runner@aborted.html
   [47]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-bdw-5557u/igt@runner@aborted.html
    - fi-kbl-soraka:      [FAIL][48] ([i915#1436] / [i915#2426] / [i915#3363]) -> [FAIL][49] ([i915#1436] / [i915#3363])
   [48]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-kbl-soraka/igt@runner@aborted.html
   [49]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-kbl-soraka/igt@runner@aborted.html
    - fi-cml-u2:          [FAIL][50] ([i915#3363]) -> [FAIL][51] ([i915#2082] / [i915#2426] / [i915#3363])
   [50]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-cml-u2/igt@runner@aborted.html
   [51]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-cml-u2/igt@runner@aborted.html
    - fi-skl-6700k2:      [FAIL][52] ([i915#1436] / [i915#2426] / [i915#3363]) -> [FAIL][53] ([i915#1436] / [i915#3363])
   [52]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10073/fi-skl-6700k2/igt@runner@aborted.html
   [53]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/fi-skl-6700k2/igt@runner@aborted.html

  
  {name}: This element is suppressed. This means it is ignored when computing
          the status of the difference (SUCCESS, WARNING, or FAILURE).

  [fdo#109285]: https://bugs.freedesktop.org/show_bug.cgi?id=109285
  [fdo#111827]: https://bugs.freedesktop.org/show_bug.cgi?id=111827
  [i915#1436]: https://gitlab.freedesktop.org/drm/intel/issues/1436
  [i915#1602]: https://gitlab.freedesktop.org/drm/intel/issues/1602
  [i915#1888]: https://gitlab.freedesktop.org/drm/intel/issues/1888
  [i915#1982]: https://gitlab.freedesktop.org/drm/intel/issues/1982
  [i915#2029]: https://gitlab.freedesktop.org/drm/intel/issues/2029
  [i915#2082]: https://gitlab.freedesktop.org/drm/intel/issues/2082
  [i915#2190]: https://gitlab.freedesktop.org/drm/intel/issues/2190
  [i915#2416]: https://gitlab.freedesktop.org/drm/intel/issues/2416
  [i915#2426]: https://gitlab.freedesktop.org/drm/intel/issues/2426
  [i915#2932]: https://gitlab.freedesktop.org/drm/intel/issues/2932
  [i915#2966]: https://gitlab.freedesktop.org/drm/intel/issues/2966
  [i915#3177]: https://gitlab.freedesktop.org/drm/intel/issues/3177
  [i915#3276]: https://gitlab.freedesktop.org/drm/intel/issues/3276
  [i915#3277]: https://gitlab.freedesktop.org/drm/intel/issues/3277
  [i915#3282]: https://gitlab.freedesktop.org/drm/intel/issues/3282
  [i915#3283]: https://gitlab.freedesktop.org/drm/intel/issues/3283
  [i915#3363]: https://gitlab.freedesktop.org/drm/intel/issues/3363
  [i915#3457]: https://gitlab.freedesktop.org/drm/intel/issues/3457
  [i915#3468]: https://gitlab.freedesktop.org/drm/intel/issues/3468
  [i915#49]: https://gitlab.freedesktop.org/drm/intel/issues/49


Participating hosts (45 -> 41)
------------------------------

  Additional (1): fi-rkl-11500t 
  Missing    (5): fi-ilk-m540 fi-hsw-4200u fi-bsw-cyan fi-icl-y fi-bdw-samus 


Build changes
-------------

  * Linux: CI_DRM_10073 -> Patchwork_20110

  CI-20190529: 20190529
  CI_DRM_10073: 66e056be09be2e0345f434775f75a3aa6ac2120f @ git://anongit.freedesktop.org/gfx-ci/linux
  IGT_6083: d28aee5c5f528aa6c352c3339f20aaed4d698ffa @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
  Patchwork_20110: 4bfaedd3ec02f2b39dbcff4e62cbe0c54ef8d4bb @ git://anongit.freedesktop.org/gfx-ci/linux


== Linux commits ==

4bfaedd3ec02 drm/i915/gem: Pin the L-shape quirked object as unshrinkable

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20110/index.html

[-- Attachment #1.2: Type: text/html, Size: 15078 bytes --]

[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH] drm/i915/gem: Pin the L-shape quirked object as unshrinkable

[Matthew Auld]

From: Chris Wilson <chris@chris-wilson.co.uk>

When instantiating a tiled object on an L-shaped memory machine, we mark
the object as unshrinkable to prevent the shrinker from trying to swap
out the pages. We have to do this as we do not know the swizzling on the
individual pages, and so the data will be scrambled across swap out/in.

Not only do we need to move the object off the shrinker list, we need to
mark the object with shrink_pin so that the counter is consistent across
calls to madvise.

v2: in the madvise ioctl we need to check if the object is currently
shrinkable/purgeable, not if the object type supports shrinking

Fixes: 0175969e489a ("drm/i915/gem: Use shrinkable status for unknown swizzle quirks")
References: https://gitlab.freedesktop.org/drm/intel/-/issues/3293
References: https://gitlab.freedesktop.org/drm/intel/-/issues/3450
Reported-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Tested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: <stable@vger.kernel.org> # v5.12+
---
 drivers/gpu/drm/i915/gem/i915_gem_pages.c |  2 ++
 drivers/gpu/drm/i915/i915_gem.c           | 11 +++++------
 2 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index aed8a37ccdc9..7361971c177d 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -63,6 +63,8 @@ void __i915_gem_object_set_pages(struct drm_i915_gem_object *obj,
 	    i915->quirks & QUIRK_PIN_SWIZZLED_PAGES) {
 		GEM_BUG_ON(i915_gem_object_has_tiling_quirk(obj));
 		i915_gem_object_set_tiling_quirk(obj);
+		GEM_BUG_ON(!list_empty(&obj->mm.link));
+		atomic_inc(&obj->mm.shrink_pin);
 		shrinkable = false;
 	}
 
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index d0018c5f88bd..cffd7f4f87dc 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1009,12 +1009,11 @@ i915_gem_madvise_ioctl(struct drm_device *dev, void *data,
 		obj->mm.madv = args->madv;
 
 	if (i915_gem_object_has_pages(obj)) {
-		struct list_head *list;
+		unsigned long flags;
 
-		if (i915_gem_object_is_shrinkable(obj)) {
-			unsigned long flags;
-
-			spin_lock_irqsave(&i915->mm.obj_lock, flags);
+		spin_lock_irqsave(&i915->mm.obj_lock, flags);
+		if (!list_empty(&obj->mm.link)) {
+			struct list_head *list;
 
 			if (obj->mm.madv != I915_MADV_WILLNEED)
 				list = &i915->mm.purge_list;
@@ -1022,8 +1021,8 @@ i915_gem_madvise_ioctl(struct drm_device *dev, void *data,
 				list = &i915->mm.shrink_list;
 			list_move_tail(&obj->mm.link, list);
 
-			spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
 		}
+		spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
 	}
 
 	/* if the object is no longer attached, discard its backing storage */
-- 
2.26.3

[PATCH] drm/i915/gem: Pin the L-shape quirked object as unshrinkable

[Matthew Auld]

From: Chris Wilson <chris@chris-wilson.co.uk>

When instantiating a tiled object on an L-shaped memory machine, we mark
the object as unshrinkable to prevent the shrinker from trying to swap
out the pages. We have to do this as we do not know the swizzling on the
individual pages, and so the data will be scrambled across swap out/in.

Not only do we need to move the object off the shrinker list, we need to
mark the object with shrink_pin so that the counter is consistent across
calls to madvise.

v2: in the madvise ioctl we need to check if the object is currently
shrinkable/purgeable, not if the object type supports shrinking

Fixes: 0175969e489a ("drm/i915/gem: Use shrinkable status for unknown swizzle quirks")
References: https://gitlab.freedesktop.org/drm/intel/-/issues/3293
References: https://gitlab.freedesktop.org/drm/intel/-/issues/3450
Reported-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Tested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: <stable@vger.kernel.org> # v5.12+
---
 drivers/gpu/drm/i915/gem/i915_gem_pages.c |  2 ++
 drivers/gpu/drm/i915/i915_gem.c           | 11 +++++------
 2 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index aed8a37ccdc9..7361971c177d 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -63,6 +63,8 @@ void __i915_gem_object_set_pages(struct drm_i915_gem_object *obj,
 	    i915->quirks & QUIRK_PIN_SWIZZLED_PAGES) {
 		GEM_BUG_ON(i915_gem_object_has_tiling_quirk(obj));
 		i915_gem_object_set_tiling_quirk(obj);
+		GEM_BUG_ON(!list_empty(&obj->mm.link));
+		atomic_inc(&obj->mm.shrink_pin);
 		shrinkable = false;
 	}
 
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index d0018c5f88bd..cffd7f4f87dc 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1009,12 +1009,11 @@ i915_gem_madvise_ioctl(struct drm_device *dev, void *data,
 		obj->mm.madv = args->madv;
 
 	if (i915_gem_object_has_pages(obj)) {
-		struct list_head *list;
+		unsigned long flags;
 
-		if (i915_gem_object_is_shrinkable(obj)) {
-			unsigned long flags;
-
-			spin_lock_irqsave(&i915->mm.obj_lock, flags);
+		spin_lock_irqsave(&i915->mm.obj_lock, flags);
+		if (!list_empty(&obj->mm.link)) {
+			struct list_head *list;
 
 			if (obj->mm.madv != I915_MADV_WILLNEED)
 				list = &i915->mm.purge_list;
@@ -1022,8 +1021,8 @@ i915_gem_madvise_ioctl(struct drm_device *dev, void *data,
 				list = &i915->mm.shrink_list;
 			list_move_tail(&obj->mm.link, list);
 
-			spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
 		}
+		spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
 	}
 
 	/* if the object is no longer attached, discard its backing storage */
-- 
2.26.3