<feed xmlns='http://www.w3.org/2005/Atom'>
<title>kernel/linux.git/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h, branch linux-4.13.y</title>
<subtitle>Linux kernel stable tree (mirror)</subtitle>
<id>https://git.radix-linux.su/kernel/linux.git/atom?h=linux-4.13.y</id>
<link rel='self' href='https://git.radix-linux.su/kernel/linux.git/atom?h=linux-4.13.y'/>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/'/>
<updated>2017-06-09T15:29:38+00:00</updated>
<entry>
<title>drm/amdgpu: Add vm context module param</title>
<updated>2017-06-09T15:29:38+00:00</updated>
<author>
<name>Harish Kasiviswanathan</name>
<email>Harish.Kasiviswanathan@amd.com</email>
</author>
<published>2017-06-09T15:26:57+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=9a4b7d4c769e7513dec8f441de1f521ec4ead4b6'/>
<id>urn:sha1:9a4b7d4c769e7513dec8f441de1f521ec4ead4b6</id>
<content type='text'>
Add VM update mode module param (amdgpu.vm_update_mode) that can used to
control how VM pde/pte are updated for Graphics and Compute.

BIT0 controls Graphics and BIT1 Compute.
 BIT0 [= 0] Graphics updated by SDMA [= 1] by CPU
 BIT1 [= 0] Compute updated by SDMA [= 1] by CPU

By default, only for large BAR system vm_update_mode = 2, indicating
that Graphics VMs will be updated via SDMA and Compute VMs will be
updated via CPU. And for all all other systems (by default)
vm_update_mode = 0

Signed-off-by: Harish Kasiviswanathan &lt;Harish.Kasiviswanathan@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: Move compute vm bug logic to amdgpu_vm.c</title>
<updated>2017-06-01T20:00:20+00:00</updated>
<author>
<name>Alex Xie</name>
<email>AlexBin.Xie@amd.com</email>
</author>
<published>2017-06-01T13:42:59+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=e59c020598666ffc22c627910667e44ac2412304'/>
<id>urn:sha1:e59c020598666ffc22c627910667e44ac2412304</id>
<content type='text'>
  In review, Christian would like to keep the logic
  inside amdgpu_vm.c with a cost of slightly slower.
  The loop is still optimized out with this patch.

v2: remove the if statement. Now it is not slower.

Signed-off-by: Alex Xie &lt;AlexBin.Xie@amd.com&gt;
Reviewed-by: Christian König &lt;christian.koeng@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdgpu: cleanup VM manager init/fini</title>
<updated>2017-05-24T22:17:58+00:00</updated>
<author>
<name>Christian König</name>
<email>christian.koenig@amd.com</email>
</author>
<published>2017-05-11T14:21:20+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=05ec3eda8b213ccc7df9645119c06f4d34a6fe2c'/>
<id>urn:sha1:05ec3eda8b213ccc7df9645119c06f4d34a6fe2c</id>
<content type='text'>
VM is mandatory for all hw amdgpu supports. So remove the leftovers
to make it optionally.

Signed-off-by: Christian König &lt;christian.koenig@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>
<entry>
<title>drm/amdgpu: make pipeline sync be in same place v2</title>
<updated>2017-05-24T21:40:35+00:00</updated>
<author>
<name>Chunming Zhou</name>
<email>David1.Zhou@amd.com</email>
</author>
<published>2017-05-11T18:52:48+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=b9bf33d5ac55aa9f23b60b4d03017b2e59d02f02'/>
<id>urn:sha1:b9bf33d5ac55aa9f23b60b4d03017b2e59d02f02</id>
<content type='text'>
v2: directly return for 'if' case.

Signed-off-by: Chunming Zhou &lt;David1.Zhou@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 limitation for dedicated vm number v4</title>
<updated>2017-05-24T21:40:09+00:00</updated>
<author>
<name>Chunming Zhou</name>
<email>David1.Zhou@amd.com</email>
</author>
<published>2017-04-21T07:51:04+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=c350577073de8fe21e6cdd798c4e4746d670bb47'/>
<id>urn:sha1:c350577073de8fe21e6cdd798c4e4746d670bb47</id>
<content type='text'>
Limit reserved vmids to 1 to avoid taking too many
out of commission and starving the system.

v2: move #define to amdgpu_vm.h
v3: move reserved vmid counter to id_manager,
and increase counter before allocating vmid
v4: rename to reserved_vmid_num

Signed-off-by: Chunming Zhou &lt;David1.Zhou@amd.com&gt;
Reviewed-by: Junwei Zhang &lt;Jerry.Zhang@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 reserved vmid field in vm struct v2</title>
<updated>2017-05-24T21:40:08+00:00</updated>
<author>
<name>Chunming Zhou</name>
<email>David1.Zhou@amd.com</email>
</author>
<published>2017-04-20T08:17:34+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=36bbf3bf9b23a9fe66558e2aa235f7d81c4a0727'/>
<id>urn:sha1:36bbf3bf9b23a9fe66558e2aa235f7d81c4a0727</id>
<content type='text'>
v2: rename dedicated_vmid to reserved_vmid

Signed-off-by: Chunming Zhou &lt;David1.Zhou@amd.com&gt;
Reviewed-by: Junwei Zhang &lt;Jerry.Zhang@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 vm ioctl</title>
<updated>2017-05-24T21:40:07+00:00</updated>
<author>
<name>Chunming Zhou</name>
<email>David1.Zhou@amd.com</email>
</author>
<published>2017-04-24T03:09:04+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=cfbcacf42803a690be40068325d20d74b6093c8c'/>
<id>urn:sha1:cfbcacf42803a690be40068325d20d74b6093c8c</id>
<content type='text'>
It will be used for reserving vmid for shader debugging
that requires a fixed vmid.

v2: fix warning (Alex)

Signed-off-by: Chunming Zhou &lt;David1.Zhou@amd.com&gt;
Reviewed-by: Junwei Zhang &lt;Jerry.Zhang@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 fundamental suspend/resume issue</title>
<updated>2017-05-24T21:39:30+00:00</updated>
<author>
<name>Christian König</name>
<email>christian.koenig@amd.com</email>
</author>
<published>2017-05-10T18:06:58+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=32601d48d783ac66e22fb506409a414c426784c0'/>
<id>urn:sha1:32601d48d783ac66e22fb506409a414c426784c0</id>
<content type='text'>
Reinitializing the VM manager during suspend/resume is a very very bad
idea since all the VMs are still active and kicking.

This can lead to random VM faults after resume when new processes
become the same client ID assigned.

Signed-off-by: Christian König &lt;christian.koenig@amd.com&gt;
Reviewed-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Cc: stable@vger.kernel.org
</content>
</entry>
<entry>
<title>drm/amdgpu: PRT support for gfx9 (v3)</title>
<updated>2017-04-28T21:32:47+00:00</updated>
<author>
<name>Zhang, Jerry</name>
<email>Jerry.Zhang@amd.com</email>
</author>
<published>2017-04-19T01:53:29+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=d0766e981b36608f9fe9b29985d4cd696099c3f8'/>
<id>urn:sha1:d0766e981b36608f9fe9b29985d4cd696099c3f8</id>
<content type='text'>
Fix PRT handling on gfx9

v2: unify PRT bit for all ASICs
v3: move PRT flag checking in amdgpu_vm_bo_split_mapping()

Signed-off-by: Junwei Zhang &lt;Jerry.Zhang@amd.com&gt;
Acked-by: David Zhou &lt;david1.zhou@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: split VMID management by VMHUB</title>
<updated>2017-04-28T21:32:18+00:00</updated>
<author>
<name>Christian König</name>
<email>christian.koenig@amd.com</email>
</author>
<published>2017-04-06T15:52:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=7645670decdb677e2f415ff91609d31e5d4777d8'/>
<id>urn:sha1:7645670decdb677e2f415ff91609d31e5d4777d8</id>
<content type='text'>
This way GFX and MM won't fight for VMIDs any more.

Initially disabled since we need to stop flushing all HUBS
at the same time as well.

Signed-off-by: Christian König &lt;christian.koenig@amd.com&gt;
Reviewed-by: Andres Rodriguez &lt;andresx7@gmail.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>
</feed>
