<feed xmlns='http://www.w3.org/2005/Atom'>
<title>kernel/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c, branch linux-4.17.y</title>
<subtitle>Hosts the 0x221E linux distro kernel.</subtitle>
<id>https://universe.0xinfinity.dev/distro/kernel/atom?h=linux-4.17.y</id>
<link rel='self' href='https://universe.0xinfinity.dev/distro/kernel/atom?h=linux-4.17.y'/>
<link rel='alternate' type='text/html' href='https://universe.0xinfinity.dev/distro/kernel/'/>
<updated>2018-05-09T20:16:51Z</updated>
<entry>
<title>drm/amdgpu: Switch to interruptable wait to recover from ring hang.</title>
<updated>2018-05-09T20:16:51Z</updated>
<author>
<name>Andrey Grodzovsky</name>
<email>andrey.grodzovsky@amd.com</email>
</author>
<published>2018-04-30T14:04:42Z</published>
<link rel='alternate' type='text/html' href='https://universe.0xinfinity.dev/distro/kernel/commit/?id=e6a5b9f9aee145c2f2c24431d84edfbb0d49eea5'/>
<id>urn:sha1:e6a5b9f9aee145c2f2c24431d84edfbb0d49eea5</id>
<content type='text'>
v2:
Use dma_fence_wait instead of dma_fence_wait_timeout(...,MAX_SCHEDULE_TIMEOUT)
Avoid printing error message for ERESTARTSYS

Originally-by: David Panariti &lt;David.Panariti@amd.com&gt;
Signed-off-by: Andrey Grodzovsky &lt;andrey.grodzovsky@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: move amd_gpu_scheduler into common location</title>
<updated>2017-12-07T16:51:56Z</updated>
<author>
<name>Lucas Stach</name>
<email>l.stach@pengutronix.de</email>
</author>
<published>2017-12-06T16:49:39Z</published>
<link rel='alternate' type='text/html' href='https://universe.0xinfinity.dev/distro/kernel/commit/?id=1b1f42d8fde4fef1ed7873bf5aa91755f8c3de35'/>
<id>urn:sha1:1b1f42d8fde4fef1ed7873bf5aa91755f8c3de35</id>
<content type='text'>
This moves and renames the AMDGPU scheduler to a common location in DRM
in order to facilitate re-use by other drivers. This is mostly a straight
forward rename with no code changes.

One notable exception is the function to_drm_sched_fence(), which is no
longer a inline header function to avoid the need to export the
drm_sched_fence_ops_scheduled and drm_sched_fence_ops_finished structures.

Reviewed-by: Chunming Zhou &lt;david1.zhou@amd.com&gt;
Tested-by: Dieter Nützel &lt;Dieter@nuetzel-hh.de&gt;
Acked-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Signed-off-by: Lucas Stach &lt;l.stach@pengutronix.de&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdgpu:implement ctx query2</title>
<updated>2017-12-04T21:33:12Z</updated>
<author>
<name>Monk Liu</name>
<email>Monk.Liu@amd.com</email>
</author>
<published>2017-10-17T06:58:01Z</published>
<link rel='alternate' type='text/html' href='https://universe.0xinfinity.dev/distro/kernel/commit/?id=bc1b1bf6e347af908c9a994803e18e2e22cf84b3'/>
<id>urn:sha1:bc1b1bf6e347af908c9a994803e18e2e22cf84b3</id>
<content type='text'>
this query will give flag bits to indicate what happend
on the given context

Signed-off-by: Monk Liu &lt;Monk.Liu@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:don't change ctx-&gt;reset_couner upon query</title>
<updated>2017-12-04T21:33:11Z</updated>
<author>
<name>Monk Liu</name>
<email>Monk.Liu@amd.com</email>
</author>
<published>2017-10-17T06:39:23Z</published>
<link rel='alternate' type='text/html' href='https://universe.0xinfinity.dev/distro/kernel/commit/?id=668ca1b44de16a8182419cc1a5913d33cc6263bd'/>
<id>urn:sha1:668ca1b44de16a8182419cc1a5913d33cc6263bd</id>
<content type='text'>
reset_counter marks the reset counter number once the context
is created, shouldn't be changed due to query.

To keep U/K interface on the ctx_query and keep ctx's reset_counter
logic compatible with GPU RESET feature, now use another var named
"reset_counter_query" to replace the original checked &amp; updated in
amdgpu_ctx_query.

Signed-off-by: Monk Liu &lt;Monk.Liu@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:pass ctx-&gt;guilty address to entity init</title>
<updated>2017-12-04T21:33:09Z</updated>
<author>
<name>Monk Liu</name>
<email>Monk.Liu@amd.com</email>
</author>
<published>2017-10-23T04:25:24Z</published>
<link rel='alternate' type='text/html' href='https://universe.0xinfinity.dev/distro/kernel/commit/?id=1102900de082a7ef6991de0ad2ff88047ae859b3'/>
<id>urn:sha1:1102900de082a7ef6991de0ad2ff88047ae859b3</id>
<content type='text'>
this way the real interested guilty is connected to entity-&gt;guilty
pointer, and we can use entity-&gt;pointer later in gpu recovery procedure

Signed-off-by: Monk Liu &lt;Monk.Liu@amd.com&gt;
Reviewed-by: Chunming Zhou &lt;David1.Zhou@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amd/scheduler:introduce guilty pointer member</title>
<updated>2017-12-04T21:33:09Z</updated>
<author>
<name>Monk Liu</name>
<email>Monk.Liu@amd.com</email>
</author>
<published>2017-10-23T04:23:29Z</published>
<link rel='alternate' type='text/html' href='https://universe.0xinfinity.dev/distro/kernel/commit/?id=b3eebe3d899fdcc77297f1513983a5250f0bed80'/>
<id>urn:sha1:b3eebe3d899fdcc77297f1513983a5250f0bed80</id>
<content type='text'>
this member will be used later, it will points to
the real var inside of context and CS_SUBMIT &amp; gpu schdduler
can decide if skip a job depends on context-&gt;guilty or *entity-&gt;guilty

Signed-off-by: Monk Liu &lt;Monk.Liu@amd.com&gt;
Reviewed-by: Chunming Zhou &lt;David1.Zhou@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdgpu: move the VRAM lost counter per context</title>
<updated>2017-10-19T19:27:04Z</updated>
<author>
<name>Christian König</name>
<email>christian.koenig@amd.com</email>
</author>
<published>2017-10-09T13:18:43Z</published>
<link rel='alternate' type='text/html' href='https://universe.0xinfinity.dev/distro/kernel/commit/?id=e55f2b646df3318e24f12b8388ab6e5cccb3e92d'/>
<id>urn:sha1:e55f2b646df3318e24f12b8388ab6e5cccb3e92d</id>
<content type='text'>
Instead of per device track the VRAM lost per context and return ECANCELED
instead of ENODEV.

Signed-off-by: Christian König &lt;christian.koenig@amd.com&gt;
Reviewed-by: Nicolai Hähnle &lt;nicolai.haehnle@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdgpu: Move old fence waiting before reservation lock is aquired v2</title>
<updated>2017-10-19T19:26:59Z</updated>
<author>
<name>Andrey Grodzovsky</name>
<email>andrey.grodzovsky@amd.com</email>
</author>
<published>2017-10-10T20:50:17Z</published>
<link rel='alternate' type='text/html' href='https://universe.0xinfinity.dev/distro/kernel/commit/?id=0ae94444c08a0adf2fab4aab26be0646ee445a19'/>
<id>urn:sha1:0ae94444c08a0adf2fab4aab26be0646ee445a19</id>
<content type='text'>
Helps avoiding deadlock during GPU reset.
Added mutex to amdgpu_ctx to preserve order of fences on a ring.

v2:
Put waiting logic in a function in a seperate function in amdgpu_ctx.c

Signed-off-by: Andrey Grodzovsky &lt;andrey.grodzovsky@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: add interface for editing a foreign process's priority v3</title>
<updated>2017-10-09T20:30:24Z</updated>
<author>
<name>Andres Rodriguez</name>
<email>andresx7@gmail.com</email>
</author>
<published>2017-06-26T20:17:13Z</published>
<link rel='alternate' type='text/html' href='https://universe.0xinfinity.dev/distro/kernel/commit/?id=52c6a62c64fac03a434cdacf6ef671c6a9e9000f'/>
<id>urn:sha1:52c6a62c64fac03a434cdacf6ef671c6a9e9000f</id>
<content type='text'>
The AMDGPU_SCHED_OP_PROCESS_PRIORITY_OVERRIDE ioctls are used to set
the priority of a different process in the current system.

When a request is dropped, the process's contexts will be
restored to the priority specified at context creation time.

A request can be dropped by setting the override priority to
AMDGPU_CTX_PRIORITY_UNSET.

An fd is used to identify the remote process. This is simpler than
passing a pid number, which is vulnerable to re-use, etc.

This functionality is limited to DRM_MASTER since abuse of this
interface can have a negative impact on the system's performance.

v2: removed unused output structure
v3: change refcounted interface for a regular set operation

Signed-off-by: Andres Rodriguez &lt;andresx7@gmail.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: add plumbing for ctx priority changes v2</title>
<updated>2017-10-09T20:30:24Z</updated>
<author>
<name>Andres Rodriguez</name>
<email>andresx7@gmail.com</email>
</author>
<published>2017-06-07T00:20:38Z</published>
<link rel='alternate' type='text/html' href='https://universe.0xinfinity.dev/distro/kernel/commit/?id=c23be4ae1d50e8d93f805de7ae3e8ea49a4ad781'/>
<id>urn:sha1:c23be4ae1d50e8d93f805de7ae3e8ea49a4ad781</id>
<content type='text'>
Introduce amdgpu_ctx_priority_override(). A mechanism to override a
context's priority.

An override can be terminated by setting the override to
AMD_SCHED_PRIORITY_UNSET.

v2: change refcounted interface for a direct set

Signed-off-by: Andres Rodriguez &lt;andresx7@gmail.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>
</feed>
