<feed xmlns='http://www.w3.org/2005/Atom'>
<title>kernel/linux.git/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c, branch v6.12.80</title>
<subtitle>Linux kernel stable tree (mirror)</subtitle>
<id>https://git.radix-linux.su/kernel/linux.git/atom?h=v6.12.80</id>
<link rel='self' href='https://git.radix-linux.su/kernel/linux.git/atom?h=v6.12.80'/>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/'/>
<updated>2024-01-05T21:10:43+00:00</updated>
<entry>
<title>drm/amdgpu: Drop 'fence' check in 'to_amdgpu_amdkfd_fence()'</title>
<updated>2024-01-05T21:10:43+00:00</updated>
<author>
<name>Srinivasan Shanmugam</name>
<email>srinivasan.shanmugam@amd.com</email>
</author>
<published>2023-12-27T07:24:44+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=bf2ad4fb8adca89374b54b225d494e0b1956dbea'/>
<id>urn:sha1:bf2ad4fb8adca89374b54b225d494e0b1956dbea</id>
<content type='text'>
Return value of container_of(...) can't be null, so null check is not
required for 'fence'. Hence drop its NULL check.

Fixes the below:
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c:93 to_amdgpu_amdkfd_fence() warn: can 'fence' even be NULL?

Cc: Felix Kuehling &lt;Felix.Kuehling@amd.com&gt;
Cc: Christian König &lt;christian.koenig@amd.com&gt;
Cc: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Signed-off-by: Srinivasan Shanmugam &lt;srinivasan.shanmugam@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/amdgpu: Allow TTM to evict svm bo from same process</title>
<updated>2022-07-28T20:05:14+00:00</updated>
<author>
<name>Philip Yang</name>
<email>Philip.Yang@amd.com</email>
</author>
<published>2022-07-22T19:07:00+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=79b2c54f19110235041d8f1250e3c3694047a902'/>
<id>urn:sha1:79b2c54f19110235041d8f1250e3c3694047a902</id>
<content type='text'>
To support SVM range VRAM overcommitment, TTM should be able to evict
svm bo of same process to system memory, to get space to alloc new svm
bo.

Signed-off-by: Philip Yang &lt;Philip.Yang@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/amd/amdgpu: Fix errors in documentation of function parameters</title>
<updated>2021-05-10T22:06:42+00:00</updated>
<author>
<name>Fabio M. De Francesco</name>
<email>fmdefrancesco@gmail.com</email>
</author>
<published>2021-04-23T11:38:46+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=1fdbbc123f55dec0c175e742283950a6ae486ce6'/>
<id>urn:sha1:1fdbbc123f55dec0c175e742283950a6ae486ce6</id>
<content type='text'>
In the function documentation, I removed the excess parameters,
described the undocumented ones, and fixed the syntax errors.

Signed-off-by: Fabio M. De Francesco &lt;fmdefrancesco@gmail.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdgpu: add svm_bo eviction to enable_signal cb</title>
<updated>2021-04-21T01:49:56+00:00</updated>
<author>
<name>Alex Sierra</name>
<email>alex.sierra@amd.com</email>
</author>
<published>2020-07-28T21:03:05+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=485bea1f90b3347bb5c1f8adad533f14d8d8ff1c'/>
<id>urn:sha1:485bea1f90b3347bb5c1f8adad533f14d8d8ff1c</id>
<content type='text'>
Add to amdgpu_amdkfd_fence.enable_signal callback, support
for svm_bo fence eviction.

Signed-off-by: Alex Sierra &lt;alex.sierra@amd.com&gt;
Reviewed-by: Felix Kuehling &lt;Felix.Kuehling@amd.com&gt;
Signed-off-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: add svm_bo reference for eviction fence</title>
<updated>2021-04-21T01:49:22+00:00</updated>
<author>
<name>Alex Sierra</name>
<email>alex.sierra@amd.com</email>
</author>
<published>2020-07-28T18:35:32+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=eb2cec5537bbab46a319b4b0bcd71c320c382d2b'/>
<id>urn:sha1:eb2cec5537bbab46a319b4b0bcd71c320c382d2b</id>
<content type='text'>
[why]
As part of the SVM functionality, the eviction mechanism used for
SVM_BOs is different. This mechanism uses one eviction fence per prange,
instead of one fence per kfd_process.

[how]
A svm_bo reference to amdgpu_amdkfd_fence to allow differentiate between
SVM_BO or regular BO evictions. This also include modifications to set the
reference at the fence creation call.

Signed-off-by: Alex Sierra &lt;alex.sierra@amd.com&gt;
Reviewed-by: Felix Kuehling &lt;Felix.Kuehling@amd.com&gt;
Signed-off-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/ttm: device naming cleanup</title>
<updated>2021-01-21T13:51:45+00:00</updated>
<author>
<name>Christian König</name>
<email>christian.koenig@amd.com</email>
</author>
<published>2020-10-01T12:51:40+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=8af8a109b34fa88b8b91f25d11485b37d37549c3'/>
<id>urn:sha1:8af8a109b34fa88b8b91f25d11485b37d37549c3</id>
<content type='text'>
Rename ttm_bo_device to ttm_device.
Rename ttm_bo_driver to ttm_device_funcs.
Rename ttm_bo_global to ttm_global.

Move global and device related functions to ttm_device.[ch].

No functional change.

Signed-off-by: Christian König &lt;christian.koenig@amd.com&gt;
Acked-by: Daniel Vetter &lt;daniel.vetter@ffwll.ch&gt;
Link: https://patchwork.freedesktop.org/patch/415222/
</content>
</entry>
<entry>
<title>drm/amdgpu: Simplify kgd2kfd interface</title>
<updated>2019-01-14T20:04:29+00:00</updated>
<author>
<name>Amber Lin</name>
<email>Amber.Lin@amd.com</email>
</author>
<published>2018-12-13T20:18:01+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=8e07e2676a42e7d3e5fe8eebac6262ec975664a1'/>
<id>urn:sha1:8e07e2676a42e7d3e5fe8eebac6262ec975664a1</id>
<content type='text'>
After amdkfd is merged into amdgpu module, amdgpu can call amdkfd
functions directly.

Signed-off-by: Amber Lin &lt;Amber.Lin@amd.com&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: Relocate kgd2kfd function declaration</title>
<updated>2019-01-14T20:04:28+00:00</updated>
<author>
<name>Amber Lin</name>
<email>Amber.Lin@amd.com</email>
</author>
<published>2018-12-13T16:57:35+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=2d3d25b616a09c16c2506f23289532a31638620f'/>
<id>urn:sha1:2d3d25b616a09c16c2506f23289532a31638620f</id>
<content type='text'>
Since amdkfd is merged into amdgpu module and amdgpu can access amdkfd
directly, move declaration of kgd2kfd functions from kfd_priv.h to
amdgpu_amdkfd.h

Signed-off-by: Amber Lin &lt;Amber.Lin@amd.com&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: Remove unecessary dma_fence_ops</title>
<updated>2018-07-03T11:13:08+00:00</updated>
<author>
<name>Daniel Vetter</name>
<email>daniel.vetter@ffwll.ch</email>
</author>
<published>2018-05-03T14:25:53+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=95f9e2636a60424c27dc16c7674a03feef9e1303'/>
<id>urn:sha1:95f9e2636a60424c27dc16c7674a03feef9e1303</id>
<content type='text'>
dma_fence_default_wait is the default now.

Reviewed-by: Christian König &lt;christian.koenig@amd.com&gt;
Signed-off-by: Daniel Vetter &lt;daniel.vetter@ffwll.ch&gt;
Cc: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Cc: "Christian König" &lt;christian.koenig@amd.com&gt;
Cc: Monk Liu &lt;Monk.Liu@amd.com&gt;
Cc: pding &lt;Pixel.Ding@amd.com&gt;
Cc: Andrey Grodzovsky &lt;andrey.grodzovsky@amd.com&gt;
Cc: Evan Quan &lt;evan.quan@amd.com&gt;
Cc: Daniel Vetter &lt;daniel.vetter@ffwll.ch&gt;
Cc: Kees Cook &lt;keescook@chromium.org&gt;
Link: https://patchwork.freedesktop.org/patch/msgid/20180503142603.28513-6-daniel.vetter@ffwll.ch
</content>
</entry>
<entry>
<title>drm/amdgpu: Add KFD eviction fence</title>
<updated>2018-02-07T01:32:35+00:00</updated>
<author>
<name>Felix Kuehling</name>
<email>Felix.Kuehling@amd.com</email>
</author>
<published>2018-02-07T01:32:35+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=d8d019ccffb838bb0dd98e583b5c25ccc0bc6ece'/>
<id>urn:sha1:d8d019ccffb838bb0dd98e583b5c25ccc0bc6ece</id>
<content type='text'>
This fence is used by KFD to keep memory resident while user mode
queues are enabled. Trying to evict memory will trigger the
enable_signaling callback, which starts a KFD eviction, which
involves preempting user mode queues before signaling the fence.
There is one such fence per process.

v2:
* Grab a reference to mm_struct
* Dereference fence after NULL check
* Simplify fence release, no need to signal without anyone waiting
* Added signed-off-by Harish, who is the original author of this code

v3:
* update MAINTAINERS file
* change amd_kfd_ prefix to amdkfd_
* remove useless initialization of variable to NULL

v4:
* set amdkfd_fence_ops to be static
* Suggested by: Fengguang Wu &lt;fengguang.wu@intel.com&gt;

Signed-off-by: Harish Kasiviswanathan &lt;Harish.Kasiviswanathan@amd.com&gt;
Signed-off-by: Felix Kuehling &lt;Felix.Kuehling@amd.com&gt;
Reviewed-by: Oded Gabbay &lt;oded.gabbay@gmail.com&gt;
Signed-off-by: Oded Gabbay &lt;oded.gabbay@gmail.com&gt;
</content>
</entry>
</feed>
