<feed xmlns='http://www.w3.org/2005/Atom'>
<title>kernel/linux.git/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c, branch v6.1.44</title>
<subtitle>Linux kernel stable tree (mirror)</subtitle>
<id>https://git.radix-linux.su/kernel/linux.git/atom?h=v6.1.44</id>
<link rel='self' href='https://git.radix-linux.su/kernel/linux.git/atom?h=v6.1.44'/>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/'/>
<updated>2022-11-21T21:40:30+00:00</updated>
<entry>
<title>drm/amdgpu: fix userptr HMM range handling v2</title>
<updated>2022-11-21T21:40:30+00:00</updated>
<author>
<name>Christian König</name>
<email>christian.koenig@amd.com</email>
</author>
<published>2022-11-10T11:31:41+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=4458da0bb09d4435956b4377685e8836935e9b9d'/>
<id>urn:sha1:4458da0bb09d4435956b4377685e8836935e9b9d</id>
<content type='text'>
The basic problem here is that it's not allowed to page fault while
holding the reservation lock.

So it can happen that multiple processes try to validate an userptr
at the same time.

Work around that by putting the HMM range object into the mutex
protected bo list for now.

v2: make sure range is set to NULL in case of an error

Signed-off-by: Christian König &lt;christian.koenig@amd.com&gt;
Reviewed-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Reviewed-by: Felix Kuehling &lt;Felix.Kuehling@amd.com&gt;
CC: stable@vger.kernel.org
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdgpu: Fix for BO move issue</title>
<updated>2022-10-19T02:14:07+00:00</updated>
<author>
<name>Arunpravin Paneer Selvam</name>
<email>Arunpravin.PaneerSelvam@amd.com</email>
</author>
<published>2022-10-18T14:08:38+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=8273b4048664fff356fd10059033f0e2f5a422a1'/>
<id>urn:sha1:8273b4048664fff356fd10059033f0e2f5a422a1</id>
<content type='text'>
A user reported a bug on CAPE VERDE system where uvd_v3_1
IP component failed to initialize as there is an issue with
BO move code from one memory to other.

In function amdgpu_mem_visible() called by amdgpu_bo_move(),
when there are no blocks to compare or if we have a single
block then break the loop.

Fixes: 312b4dc11d4f ("drm/amdgpu: Fix VRAM BO swap issue")
Signed-off-by: Arunpravin Paneer Selvam &lt;Arunpravin.PaneerSelvam@amd.com&gt;
Reviewed-by: Christian König &lt;christian.koenig@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdgpu: Fix VRAM BO swap issue</title>
<updated>2022-10-06T16:07:37+00:00</updated>
<author>
<name>Arunpravin Paneer Selvam</name>
<email>Arunpravin.PaneerSelvam@amd.com</email>
</author>
<published>2022-10-04T14:33:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=312b4dc11d4f74bfe03ea25ffe04c1f2fdd13cb9'/>
<id>urn:sha1:312b4dc11d4f74bfe03ea25ffe04c1f2fdd13cb9</id>
<content type='text'>
DRM buddy manager allocates the contiguous memory requests in
a single block or multiple blocks. So for the ttm move operation
(incase of low vram memory) we should consider all the blocks to
compute the total memory size which compared with the struct
ttm_resource num_pages in order to verify that the blocks are
contiguous for the eviction process.

v2: Added a Fixes tag
v3: Rewrite the code to save a bit of calculations and
    variables (Christian)

Fixes: c9cad937c0c5 ("drm/amdgpu: add drm buddy support to amdgpu")
Signed-off-by: Arunpravin Paneer Selvam &lt;Arunpravin.PaneerSelvam@amd.com&gt;
Reviewed-by: Christian König &lt;christian.koenig@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>Merge tag 'drm-misc-next-2022-09-09' of git://anongit.freedesktop.org/drm/drm-misc into drm-next</title>
<updated>2022-09-11T12:03:07+00:00</updated>
<author>
<name>Dave Airlie</name>
<email>airlied@redhat.com</email>
</author>
<published>2022-09-11T11:46:57+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=fb34d8a04e5876552cd0d4f9e14400ee13f116fb'/>
<id>urn:sha1:fb34d8a04e5876552cd0d4f9e14400ee13f116fb</id>
<content type='text'>
drm-misc-next for v6.1-rc1:

[airlied - fix sun4i_tv build]

UAPI Changes:
- Hide unregistered connectors from GETCONNECTOR ioctl.
- drm/virtio no longer advertises LINEAR modifier, as it doesn't work.
-

Cross-subsystem Changes:
- Fix GPF in udmabuf failure path.

Core Changes:
- Rework TTM placement to use intersect/compatible functions.
- Drop legacy DP-MST support.
- More DP-MST related fixes, and move all state into atomic.
- Make DRM_MIPI_DBI select DRM_KMS_HELPER.
- Add audio_infoframe packing for DP.
- Add logging when some atomic check functions fail.
- Assorted documentation updates and fixes.

Driver Changes:
- Assorted cleanups and fixes in msm, lcdif, nouveau, virtio,
  panel/ilitek, bridge/icn6211, tve200, gma500, bridge/*, panfrost, via,
  bochs, qxl, sun4i.
- Add add AUO B133UAN02.1, IVO M133NW4J-R3, Innolux N120ACA-EA1 eDP panels.
- Improve DP-MST modeset state handling in amdgpu, nouveau, i915.
- Drop DP-MST from radeon driver, it was broken and only user of legacy
  DP-MST.
- Handle unplugging better in vc4.
- Simplify drm cmdparser tests.
- Add DP support to ti-sn65dsi86.
- Add MT8195 DP support to mediatek.
- Support RGB565, XRGB64, and ARGB64 formats in vkms.
- Convert sun4i tv support to atomic.
- Refactor vc4/vec TV Modesetting, and fix timings.
- Use atomic helpers instead of simple display helpers in ssd130x.

Maintainer changes:
- Add Douglas Anderson as reviewer for panel-edp.

Signed-off-by: Dave Airlie &lt;airlied@redhat.com&gt;
From: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt;
Link: https://patchwork.freedesktop.org/patch/msgid/a489485b-3ebc-c734-0f80-aed963d89efe@linux.intel.com
</content>
</entry>
<entry>
<title>Merge tag 'drm-misc-next-2022-08-20-1' of git://anongit.freedesktop.org/drm/drm-misc into drm-next</title>
<updated>2022-09-06T08:56:04+00:00</updated>
<author>
<name>Daniel Vetter</name>
<email>daniel.vetter@ffwll.ch</email>
</author>
<published>2022-09-06T08:56:03+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=8284bae723f025cb6a8431566757a3854a3c53eb'/>
<id>urn:sha1:8284bae723f025cb6a8431566757a3854a3c53eb</id>
<content type='text'>
drm-misc-next for v6.1:

UAPI Changes:

Cross-subsystem Changes:
- DMA-buf: documentation updates.
- Assorted small fixes to vga16fb
- Fix fbdev drivers to use the aperture helpers.
- Make removal of conflicting drivers work correctly without fbdev enabled.

Core Changes:
- bridge, scheduler, dp-mst: Assorted small fixes.
- Add more format helpers to fourcc, and use it to replace the cpp usage.
- Add DRM_FORMAT_Cxx, DRM_FORMAT_Rxx (single channel), and DRM_FORMAT_Dxx
  ("darkness", inverted single channel)
- Add packed AYUV8888 and XYUV8888 formats.
- Assorted documentation updates.
- Rename ttm_bo_init to ttm_bo_init_validate.
- Allow TTM bo's to exist without backing store.
- Convert drm selftests to kunit.
- Add managed init functions for (panel) bridge, crtc, encoder and connector.
- Fix endianness handling in various format conversion helpers.
- Make tests pass on big-endian platforms, and add test for rgb888 -&gt; rgb565
- Move DRM_PLANE_HELPER_NO_SCALING to atomic helpers and rename, so
  drm_plane_helper is no longer needed in most drivers.
- Use idr_init_base instead of idr_init.
- Rename FB and GEM CMA helpers to DMA helpers.
- Rework XRGB8888 related conversion helpers, and add drm_fb_blit() that
  takes a iosys_map. Make drm_fb_memcpy take an iosys_map too.
- Move edid luminance calculation to core, and use it in i915.

Driver Changes:
- bridge/{adv7511,ti-sn65dsi86,parade-ps8640}, panel/{simple,nt35510,tc358767},
  nouveau, sun4i, mipi-dsi, mgag200, bochs, arm, komeda, vmwgfx, pl111:
  Assorted small fixes and doc updates.
- vc4: Rework hdmi power up, and depend on PM.
- panel/simple: Add Samsung LTL101AL01.
- ingenic: Add JZ4760(B) support, avoid a modeset when sharpness property
  is unchanged, and use the new PM ops.
- Revert some amdgpu commits that cause garbaged graphics when starting
  X, and reapply them with the real problem fixed.
- Completely rework vc4 init to use managed helpers.
- Rename via_drv to via_dri1, and move all stuff there only used by the
  dri1 implementation in preperation for atomic modeset.
- Use regmap bulk write in ssd130x.
- Power sequence and clock updates to it6505.
- Split panel-sitrox-st7701  init sequence and rework mode programming code.
- virtio: Improve error and edge conditions handling, and convert to use managed
  helpers.
- Add Samsung LTL101AL01, B120XAN01.0, R140NWF5 RH, DMT028VGHMCMI-1A T, panels.
- Add generic fbdev support to komeda.
- Split mgag200 modeset handling to make it more model-specific.
- Convert simpledrm to use atomic helpers.
- Improve udl suspend/disconnect handling.

Signed-off-by: Daniel Vetter &lt;daniel.vetter@ffwll.ch&gt;
From: Maarten Lankhorst &lt;maarten.lankhorst@linux.intel.com&gt;
Link: https://patchwork.freedesktop.org/patch/msgid/f0c71766-61e8-19b7-763a-5fbcdefc633d@linux.intel.com
</content>
</entry>
<entry>
<title>drm/ttm: Switch to using the new res callback</title>
<updated>2022-08-22T13:36:29+00:00</updated>
<author>
<name>Arunpravin Paneer Selvam</name>
<email>Arunpravin.PaneerSelvam@amd.com</email>
</author>
<published>2022-08-20T07:33:04+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=6d3c900c12d72667341bcff338c252e22728b942'/>
<id>urn:sha1:6d3c900c12d72667341bcff338c252e22728b942</id>
<content type='text'>
Apply new intersect and compatible callback instead
of having a generic placement range verfications.

v2: Added a separate callback for compatiblilty
    checks (Christian)
v3: Cleanups and removal of workarounds

Signed-off-by: Christian König &lt;christian.koenig@amd.com&gt;
Signed-off-by: Arunpravin Paneer Selvam &lt;Arunpravin.PaneerSelvam@amd.com&gt;
Reviewed-by: Christian König &lt;christian.koenig@amd.com&gt;
Link: https://patchwork.freedesktop.org/patch/msgid/20220820073304.178444-6-Arunpravin.PaneerSelvam@amd.com
</content>
</entry>
<entry>
<title>drm/amdgpu: Avoid direct cast to amdgpu_ttm_tt</title>
<updated>2022-08-10T19:41:23+00:00</updated>
<author>
<name>Rajneesh Bhardwaj</name>
<email>rajneesh.bhardwaj@amd.com</email>
</author>
<published>2022-04-20T13:37:56+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=c4c10a68e82b8a03df234ee744acabcc648df270'/>
<id>urn:sha1:c4c10a68e82b8a03df234ee744acabcc648df270</id>
<content type='text'>
For typesafety, use container_of() instead of implicit cast from struct
ttm_tt to struct amdgpu_ttm_tt.

Reviewed-by: Christian König &lt;christian.koenig@amd.com&gt;
Reviewed-by: Felix Kuehling &lt;Felix.Kuehling@amd.com&gt;
Signed-off-by: Rajneesh Bhardwaj &lt;rajneesh.bhardwaj@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>Merge drm/drm-next into drm-misc-next</title>
<updated>2022-07-13T08:33:00+00:00</updated>
<author>
<name>Maxime Ripard</name>
<email>maxime@cerno.tech</email>
</author>
<published>2022-07-13T08:33:00+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=4de395f2c632c31c575f72d49d4f4389c99dab68'/>
<id>urn:sha1:4de395f2c632c31c575f72d49d4f4389c99dab68</id>
<content type='text'>
I need to have some vc4 patches merged in -rc4, but drm-misc-next is
only at -rc2 for now.

Signed-off-by: Maxime Ripard &lt;maxime@cerno.tech&gt;
</content>
</entry>
<entry>
<title>drm/amdgpu: audit bo-&gt;resource usage</title>
<updated>2022-07-11T08:59:20+00:00</updated>
<author>
<name>Christian König</name>
<email>christian.koenig@amd.com</email>
</author>
<published>2021-12-17T15:13:09+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=63af82cf5e36b6ba3f3ebcdd1edd9f91934bfa59'/>
<id>urn:sha1:63af82cf5e36b6ba3f3ebcdd1edd9f91934bfa59</id>
<content type='text'>
Make sure we can at least move and release BOs without backing store.

Signed-off-by: Christian König &lt;christian.koenig@amd.com&gt;
Reviewed-by: Michael J. Ruhl &lt;michael.j.ruhl@intel.com&gt;
Link: https://patchwork.freedesktop.org/patch/msgid/20220707102453.3633-3-christian.koenig@amd.com
</content>
</entry>
<entry>
<title>drm/amdgpu: Adjust logic around GTT size (v3)</title>
<updated>2022-06-21T21:55:01+00:00</updated>
<author>
<name>Alex Deucher</name>
<email>alexander.deucher@amd.com</email>
</author>
<published>2022-05-19T14:21:08+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=f7ba887f606b1895dabf718547cf7397924f0d3a'/>
<id>urn:sha1:f7ba887f606b1895dabf718547cf7397924f0d3a</id>
<content type='text'>
Certain GL unit tests for large textures can cause problems
with the OOM killer since there is no way to link this memory
to a process.  This was originally mitigated (but not necessarily
eliminated) by limiting the GTT size.  The problem is this limit
is often too low for many modern games so just make the limit 1/2
of system memory. The OOM accounting needs to be addressed, but
we shouldn't prevent common 3D applications from being usable
just to potentially mitigate that corner case.

Set default GTT size to max(3G, 1/2 of system ram) by default.

v2: drop previous logic and default to 3/4 of ram
v3: default to half of ram to align with ttm
v4: fix spelling in comment (Kent)

Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1942
Reviewed-by: Marek Olšák &lt;marek.olsak@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
</feed>
