<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-toradex.git/include/drm/drm_pagemap.h, branch v7.0-rc2</title>
<subtitle>Linux kernel for Apalis and Colibri modules</subtitle>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/'/>
<entry>
<title>drm/pagemap: pass pagemap_addr by reference</title>
<updated>2026-02-18T00:39:44+00:00</updated>
<author>
<name>Arnd Bergmann</name>
<email>arnd@arndb.de</email>
</author>
<published>2026-02-16T13:46:01+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=b61d56516647aed3cee373237ce2913b9faf9285'/>
<id>b61d56516647aed3cee373237ce2913b9faf9285</id>
<content type='text'>
Passing a structure by value into a function is sometimes problematic,
for a number of reasons. Of of these is a warning from the 32-bit arm
compiler:

drivers/gpu/drm/drm_gpusvm.c: In function '__drm_gpusvm_unmap_pages':
drivers/gpu/drm/drm_gpusvm.c:1152:33: note: parameter passing for argument of type 'struct drm_pagemap_addr' changed in GCC 9.1
 1152 |                                 dpagemap-&gt;ops-&gt;device_unmap(dpagemap,
      |                                 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 1153 |                                                             dev, *addr);
      |                                                             ~~~~~~~~~~~

This particular problem is harmless since we are not mixing compiler versions
inside of the compiler. However, passing this by reference avoids the warning
along with providing slightly better calling conventions as it avoids an
extra copy on the stack.

Fixes: 75af93b3f5d0 ("drm/pagemap, drm/xe: Support destination migration over interconnect")
Fixes: 2df55d9e66a2 ("drm/xe: Support pcie p2p dma as a fast interconnect")
Signed-off-by: Arnd Bergmann &lt;arnd@arndb.de&gt;
Reviewed-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Link: https://patch.msgid.link/20260216134644.1025365-1-arnd@kernel.org
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt;
(cherry picked from commit 95162db0208aee122d10ac1342fe97a1721cd258)
Signed-off-by: Rodrigo Vivi &lt;rodrigo.vivi@intel.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Passing a structure by value into a function is sometimes problematic,
for a number of reasons. Of of these is a warning from the 32-bit arm
compiler:

drivers/gpu/drm/drm_gpusvm.c: In function '__drm_gpusvm_unmap_pages':
drivers/gpu/drm/drm_gpusvm.c:1152:33: note: parameter passing for argument of type 'struct drm_pagemap_addr' changed in GCC 9.1
 1152 |                                 dpagemap-&gt;ops-&gt;device_unmap(dpagemap,
      |                                 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 1153 |                                                             dev, *addr);
      |                                                             ~~~~~~~~~~~

This particular problem is harmless since we are not mixing compiler versions
inside of the compiler. However, passing this by reference avoids the warning
along with providing slightly better calling conventions as it avoids an
extra copy on the stack.

Fixes: 75af93b3f5d0 ("drm/pagemap, drm/xe: Support destination migration over interconnect")
Fixes: 2df55d9e66a2 ("drm/xe: Support pcie p2p dma as a fast interconnect")
Signed-off-by: Arnd Bergmann &lt;arnd@arndb.de&gt;
Reviewed-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Link: https://patch.msgid.link/20260216134644.1025365-1-arnd@kernel.org
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt;
(cherry picked from commit 95162db0208aee122d10ac1342fe97a1721cd258)
Signed-off-by: Rodrigo Vivi &lt;rodrigo.vivi@intel.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>BackMerge tag 'v6.19-rc7' into drm-next</title>
<updated>2026-01-28T02:44:28+00:00</updated>
<author>
<name>Dave Airlie</name>
<email>airlied@redhat.com</email>
</author>
<published>2026-01-28T02:44:28+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=6704d98a4f48b7424edc0f7ae2a06c0a8af02e2f'/>
<id>6704d98a4f48b7424edc0f7ae2a06c0a8af02e2f</id>
<content type='text'>
Linux 6.19-rc7

This is needed for msm and rust trees.

Signed-off-by: Dave Airlie &lt;airlied@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Linux 6.19-rc7

This is needed for msm and rust trees.

Signed-off-by: Dave Airlie &lt;airlied@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>drm, drm/xe: Fix xe userptr in the absence of CONFIG_DEVICE_PRIVATE</title>
<updated>2026-01-22T09:41:18+00:00</updated>
<author>
<name>Thomas Hellström</name>
<email>thomas.hellstrom@linux.intel.com</email>
</author>
<published>2026-01-21T09:10:47+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=bdcdf968be314b6fc8835b99fb4519e7619671e6'/>
<id>bdcdf968be314b6fc8835b99fb4519e7619671e6</id>
<content type='text'>
CONFIG_DEVICE_PRIVATE is not selected by default by some distros,
for example Fedora, and that leads to a regression in the xe driver
since userptr support gets compiled out.

It turns out that DRM_GPUSVM, which is needed for xe userptr support
compiles also without CONFIG_DEVICE_PRIVATE, but doesn't compile
without CONFIG_ZONE_DEVICE.
Exclude the drm_pagemap files from compilation with !CONFIG_ZONE_DEVICE,
and remove the CONFIG_DEVICE_PRIVATE dependency from CONFIG_DRM_GPUSVM and
the xe driver's selection of it, re-enabling xe userptr for those configs.

v2:
- Don't compile the drm_pagemap files unless CONFIG_ZONE_DEVICE is set.
- Adjust the drm_pagemap.h header accordingly.

Fixes: 9e9787414882 ("drm/xe/userptr: replace xe_hmm with gpusvm")
Cc: Matthew Auld &lt;matthew.auld@intel.com&gt;
Cc: Himal Prasad Ghimiray &lt;himal.prasad.ghimiray@intel.com&gt;
Cc: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Cc: Matthew Brost &lt;matthew.brost@intel.com&gt;
Cc: "Thomas Hellström" &lt;thomas.hellstrom@linux.intel.com&gt;
Cc: Rodrigo Vivi &lt;rodrigo.vivi@intel.com&gt;
Cc: dri-devel@lists.freedesktop.org
Cc: &lt;stable@vger.kernel.org&gt; # v6.18+
Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Auld &lt;matthew.auld@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt;
Link: https://patch.msgid.link/20260121091048.41371-2-thomas.hellstrom@linux.intel.com
(cherry picked from commit 1e372b246199ca7a35f930177fea91b557dac16e)
Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
CONFIG_DEVICE_PRIVATE is not selected by default by some distros,
for example Fedora, and that leads to a regression in the xe driver
since userptr support gets compiled out.

It turns out that DRM_GPUSVM, which is needed for xe userptr support
compiles also without CONFIG_DEVICE_PRIVATE, but doesn't compile
without CONFIG_ZONE_DEVICE.
Exclude the drm_pagemap files from compilation with !CONFIG_ZONE_DEVICE,
and remove the CONFIG_DEVICE_PRIVATE dependency from CONFIG_DRM_GPUSVM and
the xe driver's selection of it, re-enabling xe userptr for those configs.

v2:
- Don't compile the drm_pagemap files unless CONFIG_ZONE_DEVICE is set.
- Adjust the drm_pagemap.h header accordingly.

Fixes: 9e9787414882 ("drm/xe/userptr: replace xe_hmm with gpusvm")
Cc: Matthew Auld &lt;matthew.auld@intel.com&gt;
Cc: Himal Prasad Ghimiray &lt;himal.prasad.ghimiray@intel.com&gt;
Cc: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Cc: Matthew Brost &lt;matthew.brost@intel.com&gt;
Cc: "Thomas Hellström" &lt;thomas.hellstrom@linux.intel.com&gt;
Cc: Rodrigo Vivi &lt;rodrigo.vivi@intel.com&gt;
Cc: dri-devel@lists.freedesktop.org
Cc: &lt;stable@vger.kernel.org&gt; # v6.18+
Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Auld &lt;matthew.auld@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt;
Link: https://patch.msgid.link/20260121091048.41371-2-thomas.hellstrom@linux.intel.com
(cherry picked from commit 1e372b246199ca7a35f930177fea91b557dac16e)
Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/pagemap, drm/xe: Ensure that the devmem allocation is idle before use</title>
<updated>2025-12-23T09:11:59+00:00</updated>
<author>
<name>Thomas Hellström</name>
<email>thomas.hellstrom@linux.intel.com</email>
</author>
<published>2025-12-19T11:32:59+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=754c23238438600e9236719f7e67aff2c4d02093'/>
<id>754c23238438600e9236719f7e67aff2c4d02093</id>
<content type='text'>
In situations where no system memory is migrated to devmem, and in
upcoming patches where another GPU is performing the migration to
the newly allocated devmem buffer, there is nothing to ensure any
ongoing clear to the devmem allocation or async eviction from the
devmem allocation is complete.

Address that by passing a struct dma_fence down to the copy
functions, and ensure it is waited for before migration is marked
complete.

v3:
- New patch.
v4:
- Update the logic used for determining when to wait for the
  pre_migrate_fence.
- Update the logic used for determining when to warn for the
  pre_migrate_fence since the scheduler fences apparently
  can signal out-of-order.
v5:
- Fix a UAF (CI)
- Remove references to source P2P migration (Himal)
- Put the pre_migrate_fence after migration.
v6:
- Pipeline the pre_migrate_fence dependency (Matt Brost)

Fixes: c5b3eb5a906c ("drm/xe: Add GPUSVM device memory copy vfunc functions")
Cc: Matthew Brost &lt;matthew.brost@intel.com&gt;
Cc: &lt;stable@vger.kernel.org&gt; # v6.15+
Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Brost &lt;matthew.brost@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-4-thomas.hellstrom@linux.intel.com
(cherry picked from commit 16b5ad31952476fb925c401897fc171cd37f536b)
Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
In situations where no system memory is migrated to devmem, and in
upcoming patches where another GPU is performing the migration to
the newly allocated devmem buffer, there is nothing to ensure any
ongoing clear to the devmem allocation or async eviction from the
devmem allocation is complete.

Address that by passing a struct dma_fence down to the copy
functions, and ensure it is waited for before migration is marked
complete.

v3:
- New patch.
v4:
- Update the logic used for determining when to wait for the
  pre_migrate_fence.
- Update the logic used for determining when to warn for the
  pre_migrate_fence since the scheduler fences apparently
  can signal out-of-order.
v5:
- Fix a UAF (CI)
- Remove references to source P2P migration (Himal)
- Put the pre_migrate_fence after migration.
v6:
- Pipeline the pre_migrate_fence dependency (Matt Brost)

Fixes: c5b3eb5a906c ("drm/xe: Add GPUSVM device memory copy vfunc functions")
Cc: Matthew Brost &lt;matthew.brost@intel.com&gt;
Cc: &lt;stable@vger.kernel.org&gt; # v6.15+
Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Brost &lt;matthew.brost@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-4-thomas.hellstrom@linux.intel.com
(cherry picked from commit 16b5ad31952476fb925c401897fc171cd37f536b)
Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/pagemap, drm/xe: Support destination migration over interconnect</title>
<updated>2025-12-23T09:00:49+00:00</updated>
<author>
<name>Thomas Hellström</name>
<email>thomas.hellstrom@linux.intel.com</email>
</author>
<published>2025-12-19T11:33:18+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=75af93b3f5d0ab760356365064521d9aa7b7fc49'/>
<id>75af93b3f5d0ab760356365064521d9aa7b7fc49</id>
<content type='text'>
Support destination migration over interconnect when migrating from
device-private pages with the same dev_pagemap owner.

Since we now also collect device-private pages to migrate,
also abort migration if the range to migrate is already
fully populated with pages from the desired pagemap.

Finally return -EBUSY from drm_pagemap_populate_mm()
if the migration can't be completed without first migrating all
pages in the range to system. It is expected that the caller
will perform that before retrying the call to
drm_pagemap_populate_mm().

v3:
- Fix a bug where the p2p dma-address was never used.
- Postpone enabling destination interconnect migration,
  since xe devices require source interconnect migration to
  ensure the source L2 cache is flushed at migration time.
- Update the drm_pagemap_migrate_to_devmem() interface to
  pass migration details.
v4:
- Define XE_INTERCONNECT_P2P unconditionally (CI)
- Include a missing header (CI)
v5:
- Use page order increments where possible (Matt Brost).
- Fix a negated value of can_migrate_same_pagemap.
- Move removal of some dead code to a separate patch (Matt Brost).
- Remove an unnecessary zdd get() and put() (Matt Brost).

Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Brost &lt;matthew.brost@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-23-thomas.hellstrom@linux.intel.com
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Support destination migration over interconnect when migrating from
device-private pages with the same dev_pagemap owner.

Since we now also collect device-private pages to migrate,
also abort migration if the range to migrate is already
fully populated with pages from the desired pagemap.

Finally return -EBUSY from drm_pagemap_populate_mm()
if the migration can't be completed without first migrating all
pages in the range to system. It is expected that the caller
will perform that before retrying the call to
drm_pagemap_populate_mm().

v3:
- Fix a bug where the p2p dma-address was never used.
- Postpone enabling destination interconnect migration,
  since xe devices require source interconnect migration to
  ensure the source L2 cache is flushed at migration time.
- Update the drm_pagemap_migrate_to_devmem() interface to
  pass migration details.
v4:
- Define XE_INTERCONNECT_P2P unconditionally (CI)
- Include a missing header (CI)
v5:
- Use page order increments where possible (Matt Brost).
- Fix a negated value of can_migrate_same_pagemap.
- Move removal of some dead code to a separate patch (Matt Brost).
- Remove an unnecessary zdd get() and put() (Matt Brost).

Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Brost &lt;matthew.brost@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-23-thomas.hellstrom@linux.intel.com
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/pagemap, drm/xe: Clean up the use of the device-private page owner</title>
<updated>2025-12-23T09:00:48+00:00</updated>
<author>
<name>Thomas Hellström</name>
<email>thomas.hellstrom@linux.intel.com</email>
</author>
<published>2025-12-19T11:33:15+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=5b64b23f6f1b1d5b9836ac5a0963998951ba53d8'/>
<id>5b64b23f6f1b1d5b9836ac5a0963998951ba53d8</id>
<content type='text'>
Use the dev_pagemap-&gt;owner field wherever possible, simplifying
the code slightly.

v3: New patch

Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Himal Prasad Ghimiray &lt;himal.prasad.ghimiray@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-20-thomas.hellstrom@linux.intel.com
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Use the dev_pagemap-&gt;owner field wherever possible, simplifying
the code slightly.

v3: New patch

Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Himal Prasad Ghimiray &lt;himal.prasad.ghimiray@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-20-thomas.hellstrom@linux.intel.com
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/pagemap: Add a drm_pagemap cache and shrinker</title>
<updated>2025-12-23T08:37:33+00:00</updated>
<author>
<name>Thomas Hellström</name>
<email>thomas.hellstrom@linux.intel.com</email>
</author>
<published>2025-12-19T11:33:03+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=77f14f2f2d73f6955b856e7c91ca9fb7046da191'/>
<id>77f14f2f2d73f6955b856e7c91ca9fb7046da191</id>
<content type='text'>
Pagemaps are costly to set up and tear down, and they consume a lot
of system memory for the struct pages. Ideally they should be
created only when needed.

Add a caching mechanism to allow doing just that: Create the drm_pagemaps
when needed for migration. Keep them around to avoid destruction and
re-creation latencies and destroy inactive/unused drm_pagemaps on memory
pressure using a shrinker.

Only add the helper functions. They will be hooked up to the xe driver
in the upcoming patch.

v2:
- Add lockdep checking for drm_pagemap_put(). (Matt Brost)
- Add a copyright notice. (Matt Brost)

Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Brost &lt;matthew.brost@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-8-thomas.hellstrom@linux.intel.com
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Pagemaps are costly to set up and tear down, and they consume a lot
of system memory for the struct pages. Ideally they should be
created only when needed.

Add a caching mechanism to allow doing just that: Create the drm_pagemaps
when needed for migration. Keep them around to avoid destruction and
re-creation latencies and destroy inactive/unused drm_pagemaps on memory
pressure using a shrinker.

Only add the helper functions. They will be hooked up to the xe driver
in the upcoming patch.

v2:
- Add lockdep checking for drm_pagemap_put(). (Matt Brost)
- Add a copyright notice. (Matt Brost)

Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Brost &lt;matthew.brost@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-8-thomas.hellstrom@linux.intel.com
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/pagemap, drm/xe: Manage drm_pagemap provider lifetimes</title>
<updated>2025-12-23T08:36:55+00:00</updated>
<author>
<name>Thomas Hellström</name>
<email>thomas.hellstrom@linux.intel.com</email>
</author>
<published>2025-12-19T11:33:02+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=a26084328ac40c12096ef01482a7520346379453'/>
<id>a26084328ac40c12096ef01482a7520346379453</id>
<content type='text'>
If a device holds a reference on a foregin device's drm_pagemap,
and a device unbind is executed on the foreign device,
Typically that foreign device would evict its device-private
pages and then continue its device-managed cleanup eventually
releasing its drm device and possibly allow for module unload.
However, since we're still holding a reference on a drm_pagemap,
when that reference is released and the provider module is
unloaded we'd execute out of undefined memory.

Therefore keep a reference on the provider device and module until
the last drm_pagemap reference is gone.

Note that in theory, the drm_gpusvm_helper module may be unloaded
as soon as the final module_put() of the provider driver module is
executed, so we need to add a module_exit() function that waits
for the work item executing the module_put() has completed.

v2:
- Better commit message (Matt Brost)

Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Brost &lt;matthew.brost@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-7-thomas.hellstrom@linux.intel.com
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
If a device holds a reference on a foregin device's drm_pagemap,
and a device unbind is executed on the foreign device,
Typically that foreign device would evict its device-private
pages and then continue its device-managed cleanup eventually
releasing its drm device and possibly allow for module unload.
However, since we're still holding a reference on a drm_pagemap,
when that reference is released and the provider module is
unloaded we'd execute out of undefined memory.

Therefore keep a reference on the provider device and module until
the last drm_pagemap reference is gone.

Note that in theory, the drm_gpusvm_helper module may be unloaded
as soon as the final module_put() of the provider driver module is
executed, so we need to add a module_exit() function that waits
for the work item executing the module_put() has completed.

v2:
- Better commit message (Matt Brost)

Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Brost &lt;matthew.brost@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-7-thomas.hellstrom@linux.intel.com
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/pagemap, drm/xe: Add refcounting to struct drm_pagemap</title>
<updated>2025-12-23T08:35:53+00:00</updated>
<author>
<name>Thomas Hellström</name>
<email>thomas.hellstrom@linux.intel.com</email>
</author>
<published>2025-12-19T11:33:00+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=a599b98607decdc899630fc99a3a2847f6b72965'/>
<id>a599b98607decdc899630fc99a3a2847f6b72965</id>
<content type='text'>
With the end goal of being able to free unused pagemaps
and allocate them on demand, add a refcount to struct drm_pagemap,
remove the xe embedded drm_pagemap, allocating and freeing it
explicitly.

v2:
- Make the drm_pagemap pointer in drm_gpusvm_pages reference-counted.
v3:
- Call drm_pagemap_get() before drm_pagemap_put() in drm_gpusvm_pages
  (Himal Prasad Ghimiray)

Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Brost &lt;matthew.brost@intel.com&gt; #v1
Reviewed-by: Himal Prasad Ghimiray &lt;himal.prasad.ghimiray@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-5-thomas.hellstrom@linux.intel.com
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
With the end goal of being able to free unused pagemaps
and allocate them on demand, add a refcount to struct drm_pagemap,
remove the xe embedded drm_pagemap, allocating and freeing it
explicitly.

v2:
- Make the drm_pagemap pointer in drm_gpusvm_pages reference-counted.
v3:
- Call drm_pagemap_get() before drm_pagemap_put() in drm_gpusvm_pages
  (Himal Prasad Ghimiray)

Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Brost &lt;matthew.brost@intel.com&gt; #v1
Reviewed-by: Himal Prasad Ghimiray &lt;himal.prasad.ghimiray@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-5-thomas.hellstrom@linux.intel.com
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/pagemap, drm/xe: Ensure that the devmem allocation is idle before use</title>
<updated>2025-12-23T08:35:11+00:00</updated>
<author>
<name>Thomas Hellström</name>
<email>thomas.hellstrom@linux.intel.com</email>
</author>
<published>2025-12-19T11:32:59+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=16b5ad31952476fb925c401897fc171cd37f536b'/>
<id>16b5ad31952476fb925c401897fc171cd37f536b</id>
<content type='text'>
In situations where no system memory is migrated to devmem, and in
upcoming patches where another GPU is performing the migration to
the newly allocated devmem buffer, there is nothing to ensure any
ongoing clear to the devmem allocation or async eviction from the
devmem allocation is complete.

Address that by passing a struct dma_fence down to the copy
functions, and ensure it is waited for before migration is marked
complete.

v3:
- New patch.
v4:
- Update the logic used for determining when to wait for the
  pre_migrate_fence.
- Update the logic used for determining when to warn for the
  pre_migrate_fence since the scheduler fences apparently
  can signal out-of-order.
v5:
- Fix a UAF (CI)
- Remove references to source P2P migration (Himal)
- Put the pre_migrate_fence after migration.
v6:
- Pipeline the pre_migrate_fence dependency (Matt Brost)

Fixes: c5b3eb5a906c ("drm/xe: Add GPUSVM device memory copy vfunc functions")
Cc: Matthew Brost &lt;matthew.brost@intel.com&gt;
Cc: &lt;stable@vger.kernel.org&gt; # v6.15+
Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Brost &lt;matthew.brost@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-4-thomas.hellstrom@linux.intel.com
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
In situations where no system memory is migrated to devmem, and in
upcoming patches where another GPU is performing the migration to
the newly allocated devmem buffer, there is nothing to ensure any
ongoing clear to the devmem allocation or async eviction from the
devmem allocation is complete.

Address that by passing a struct dma_fence down to the copy
functions, and ensure it is waited for before migration is marked
complete.

v3:
- New patch.
v4:
- Update the logic used for determining when to wait for the
  pre_migrate_fence.
- Update the logic used for determining when to warn for the
  pre_migrate_fence since the scheduler fences apparently
  can signal out-of-order.
v5:
- Fix a UAF (CI)
- Remove references to source P2P migration (Himal)
- Put the pre_migrate_fence after migration.
v6:
- Pipeline the pre_migrate_fence dependency (Matt Brost)

Fixes: c5b3eb5a906c ("drm/xe: Add GPUSVM device memory copy vfunc functions")
Cc: Matthew Brost &lt;matthew.brost@intel.com&gt;
Cc: &lt;stable@vger.kernel.org&gt; # v6.15+
Signed-off-by: Thomas Hellström &lt;thomas.hellstrom@linux.intel.com&gt;
Reviewed-by: Matthew Brost &lt;matthew.brost@intel.com&gt;
Acked-by: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt; # For merging through drm-xe.
Link: https://patch.msgid.link/20251219113320.183860-4-thomas.hellstrom@linux.intel.com
</pre>
</div>
</content>
</entry>
</feed>
