<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c, branch v6.8</title>
<subtitle>Mirror of https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/
</subtitle>
<id>https://git.shady.money/linux/atom?h=v6.8</id>
<link rel='self' href='https://git.shady.money/linux/atom?h=v6.8'/>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/'/>
<updated>2024-01-31T20:33:52Z</updated>
<entry>
<title>drm/amdgpu: move the drm client creation behind drm device registration</title>
<updated>2024-01-31T20:33:52Z</updated>
<author>
<name>Le Ma</name>
<email>le.ma@amd.com</email>
</author>
<published>2024-01-25T04:00:34Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=db2aad036e77100e04a96c67f65ae7d49fb538fb'/>
<id>urn:sha1:db2aad036e77100e04a96c67f65ae7d49fb538fb</id>
<content type='text'>
This patch is to eliminate interrupt warning below:

  "[drm] Fence fallback timer expired on ring sdma0.0".

An early vm pt clearing job is sent to SDMA ahead of interrupt enabled.
And re-locating the drm client creation following after drm_dev_register
looks like a more proper flow.

v2: wrap the drm client creation

Fixes: 1819200166ce ("drm/amdkfd: Export DMABufs from KFD using GEM handles")
Signed-off-by: Le Ma &lt;le.ma@amd.com&gt;
Reviewed-by: Felix Kuehling &lt;felix.kuehling@amd.com&gt;
Reviewed-by: Lijo Lazar &lt;lijo.lazar@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdgpu: Enable GFXOFF for Compute on GFX11</title>
<updated>2024-01-18T21:45:19Z</updated>
<author>
<name>Ori Messinger</name>
<email>Ori.Messinger@amd.com</email>
</author>
<published>2023-11-22T05:12:13Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=aa0901a9008eeb2710292aff94e615adf7884d5f'/>
<id>urn:sha1:aa0901a9008eeb2710292aff94e615adf7884d5f</id>
<content type='text'>
On GFX version 11, GFXOFF was disabled due to a MES KIQ firmware
issue, which has since been fixed after version 64.
This patch only re-enables GFXOFF for GFX version 11 if the GPU's
MES KIQ firmware version is newer than version 64.

V2: Keep GFXOFF disabled on GFX11 if MES KIQ is below version 64.
V3: Add parentheses to avoid GCC warning for parentheses:
"suggest parentheses around comparison in operand of ‘&amp;’"
V4: Remove "V3" from commit title
V5: Change commit description and insert 'Acked-by'

Signed-off-by: Ori Messinger &lt;Ori.Messinger@amd.com&gt;
Acked-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Reviewed-by: Harish Kasiviswanathan &lt;Harish.Kasiviswanathan@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Cc: stable@vger.kernel.org
</content>
</entry>
<entry>
<title>drm/amdkfd: init drm_client with funcs hook</title>
<updated>2024-01-18T21:42:54Z</updated>
<author>
<name>Flora Cui</name>
<email>flora.cui@amd.com</email>
</author>
<published>2024-01-10T11:23:56Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=3c4e4eb5d872118fef1708abe933a410c5e07e3a'/>
<id>urn:sha1:3c4e4eb5d872118fef1708abe933a410c5e07e3a</id>
<content type='text'>
otherwise drm_client_dev_unregister() would try to
kfree(&amp;adev-&gt;kfd.client).

Fixes: 1819200166ce ("drm/amdkfd: Export DMABufs from KFD using GEM handles")
Signed-off-by: Flora Cui &lt;flora.cui@amd.com&gt;
Reviewed-by: Felix Kuehling &lt;felix.kuehling@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: Export DMABufs from KFD using GEM handles</title>
<updated>2023-12-13T20:09:55Z</updated>
<author>
<name>Felix Kuehling</name>
<email>Felix.Kuehling@amd.com</email>
</author>
<published>2023-08-24T18:41:30Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=1819200166ce511ac298dc96b9b17eb655a9edc4'/>
<id>urn:sha1:1819200166ce511ac298dc96b9b17eb655a9edc4</id>
<content type='text'>
Create GEM handles for exporting DMABufs using GEM-Prime APIs. The GEM
handles are created in a drm_client_dev context to avoid exposing them
in user mode contexts through a DMABuf import.

Signed-off-by: Felix Kuehling &lt;Felix.Kuehling@amd.com&gt;
Reviewed-by: Ramesh Errabolu &lt;Ramesh.Errabolu@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdgpu: update xgmi num links info post gc9.4.2</title>
<updated>2023-11-29T21:48:45Z</updated>
<author>
<name>Jonathan Kim</name>
<email>jonathan.kim@amd.com</email>
</author>
<published>2023-11-16T18:57:07Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=35c425f5cc251417ad681475dc9901ab6d3244ea'/>
<id>urn:sha1:35c425f5cc251417ad681475dc9901ab6d3244ea</id>
<content type='text'>
GC IP 9.4.2 and up support TA reporting of the number
of xGMI links between peers.

Tested-by: Vignesh Chander &lt;vignesh.chander@amd.com&gt;
Signed-off-by: Jonathan Kim &lt;jonathan.kim@amd.com&gt;
Reviewed-by: Mukul Joshi &lt;mukul.joshi@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: Move TLB flushing logic into amdgpu</title>
<updated>2023-11-17T14:29:53Z</updated>
<author>
<name>Felix Kuehling</name>
<email>Felix.Kuehling@amd.com</email>
</author>
<published>2023-02-24T23:22:32Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=94e2dae0a8bfd456abfd866f1eee8342f0858012'/>
<id>urn:sha1:94e2dae0a8bfd456abfd866f1eee8342f0858012</id>
<content type='text'>
This will make it possible for amdgpu GEM ioctls to flush TLBs on compute
VMs.

This removes VMID-based TLB flushing and always uses PASID-based
flushing. This still works because it scans the VMID-PASID mapping
registers to find the right VMID. It's only slightly less efficient. This
is not a production use case.

Signed-off-by: Felix Kuehling &lt;Felix.Kuehling@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: disable GFXOFF and PG during compute for GFX9</title>
<updated>2023-10-13T15:33:56Z</updated>
<author>
<name>Jesse Zhang</name>
<email>jesse.zhang@amd.com</email>
</author>
<published>2023-10-13T01:28:44Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=e341631f4a3129538cc398305649d54e6c0937d4'/>
<id>urn:sha1:e341631f4a3129538cc398305649d54e6c0937d4</id>
<content type='text'>
Temporary workaround to fix issues observed in some compute
applications when GFXOFF is enabled on GFX9.

Signed-off-by: Jesse Zhang &lt;Jesse.Zhang@amd.com&gt;
Acked-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdgpu: Use ttm_pages_limit to override vram reporting</title>
<updated>2023-10-04T22:36:42Z</updated>
<author>
<name>Rajneesh Bhardwaj</name>
<email>rajneesh.bhardwaj@amd.com</email>
</author>
<published>2023-10-02T19:27:01Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=f4bff6e0b9e1dc29bc54d7c7e1c01f2c701eb36f'/>
<id>urn:sha1:f4bff6e0b9e1dc29bc54d7c7e1c01f2c701eb36f</id>
<content type='text'>
On GFXIP9.4.3 APU, allow the memory reporting as per the ttm pages
limit in NPS1 mode.

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>drm/amdkfd: drop struct kfd_cu_info</title>
<updated>2023-10-03T19:41:13Z</updated>
<author>
<name>Alex Deucher</name>
<email>alexander.deucher@amd.com</email>
</author>
<published>2023-09-26T16:00:23Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=0021d70a0654e668d457758110abec33dfbd3ba5'/>
<id>urn:sha1:0021d70a0654e668d457758110abec33dfbd3ba5</id>
<content type='text'>
I think this was an abstraction back from when
kfd supported both radeon and amdgpu.  Since we just
support amdgpu now, there is no more need for this and
we can use the amdgpu structures directly.

This also avoids having the kfd_cu_info structures on
the stack when inlining which can blow up the stack.

Cc: Arnd Bergmann &lt;arnd@kernel.org&gt;
Acked-by: Arnd Bergmann &lt;arnd@arndb.de&gt;
Reviewed-by: Felix Kuehling &lt;Felix.Kuehling@amd.com&gt;
Acked-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: Use function for IP version check</title>
<updated>2023-09-20T16:23:28Z</updated>
<author>
<name>Lijo Lazar</name>
<email>lijo.lazar@amd.com</email>
</author>
<published>2023-09-11T08:18:11Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=4e8303cf2c4dd27374a16a8881ec1a1cd5baf86f'/>
<id>urn:sha1:4e8303cf2c4dd27374a16a8881ec1a1cd5baf86f</id>
<content type='text'>
Use an inline function for version check. Gives more flexibility to
handle any format changes.

Signed-off-by: Lijo Lazar &lt;lijo.lazar@amd.com&gt;
Reviewed-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
</feed>
