<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux/drivers/gpu/drm/amd, branch v3.19</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=v3.19</id>
<link rel='self' href='https://git.shady.money/linux/atom?h=v3.19'/>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/'/>
<updated>2015-02-02T07:45:24Z</updated>
<entry>
<title>drm/amdkfd: Don't create BUG due to incorrect user parameter</title>
<updated>2015-02-02T07:45:24Z</updated>
<author>
<name>Oded Gabbay</name>
<email>oded.gabbay@amd.com</email>
</author>
<published>2015-01-20T12:57:19Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=b9dce23ddc949faa8ba138f8a0811e177c991bf1'/>
<id>urn:sha1:b9dce23ddc949faa8ba138f8a0811e177c991bf1</id>
<content type='text'>
This patch changes a BUG_ON() statement to pr_debug, in case the user tries to
update a non-existing queue.

Signed-off-by: Oded Gabbay &lt;oded.gabbay@amd.com&gt;
Reviewed-by: Ben Goz &lt;ben.goz@amd.com&gt;
Reviewed-by: Jammy Zhou &lt;Jammy.Zhou@amd.com&gt;
Reviewed-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: max num of queues can't be 0</title>
<updated>2015-02-02T07:45:24Z</updated>
<author>
<name>Oded Gabbay</name>
<email>oded.gabbay@amd.com</email>
</author>
<published>2015-01-29T08:33:34Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=ca400b2a1a6c768416725e5f873829e23300d3eb'/>
<id>urn:sha1:ca400b2a1a6c768416725e5f873829e23300d3eb</id>
<content type='text'>
Signed-off-by: Oded Gabbay &lt;oded.gabbay@amd.com&gt;
Reviewed-by: Jammy Zhou &lt;Jammy.Zhou@amd.com&gt;
Reviewed-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: Fix bug in accounting of queues</title>
<updated>2015-02-02T07:45:24Z</updated>
<author>
<name>Oded Gabbay</name>
<email>oded.gabbay@amd.com</email>
</author>
<published>2015-01-29T08:32:25Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=8b58f261113c442717b9d205ab187e51c3823597'/>
<id>urn:sha1:8b58f261113c442717b9d205ab187e51c3823597</id>
<content type='text'>
Signed-off-by: Oded Gabbay &lt;oded.gabbay@amd.com&gt;
Reviewed-by: Jammy Zhou &lt;Jammy.Zhou@amd.com&gt;
Reviewed-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: Fix bug in call to init_pipelines()</title>
<updated>2015-01-22T10:50:37Z</updated>
<author>
<name>Oded Gabbay</name>
<email>oded.gabbay@amd.com</email>
</author>
<published>2015-01-22T10:50:37Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=9fa843e76d9092e5348aac7252cbb71f09902507'/>
<id>urn:sha1:9fa843e76d9092e5348aac7252cbb71f09902507</id>
<content type='text'>
This patch fixes a bug where the first_pipe index passed into init_pipelines()
was a #define instead of the value that is passed into amdkfd by radeon

Signed-off-by: Oded Gabbay &lt;oded.gabbay@amd.com&gt;
Reviewed-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Reviewed-by: Jammy Zhou &lt;Jammy.Zhou@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: Fix bug in pipelines initialization</title>
<updated>2015-01-22T09:03:42Z</updated>
<author>
<name>Oded Gabbay</name>
<email>oded.gabbay@amd.com</email>
</author>
<published>2015-01-22T09:03:42Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=749042b012b82616ed15a28138e180c5e5da2067'/>
<id>urn:sha1:749042b012b82616ed15a28138e180c5e5da2067</id>
<content type='text'>
This patch fixes a bug when calling to init_pipeline() interface.
The index that was passed to that function didn't take into account the
first_pipe value, which represents the first pipe index that is under amdkfd's
responsibility.

Signed-off-by: Oded Gabbay &lt;oded.gabbay@amd.com&gt;
Reviewed-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Reviewed-by: Jammy Zhou &lt;Jammy.Zhou@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: Allow user to limit only queues per device</title>
<updated>2015-01-18T11:18:01Z</updated>
<author>
<name>Oded Gabbay</name>
<email>oded.gabbay@amd.com</email>
</author>
<published>2015-01-18T11:18:01Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=b8cbab042cd69f3918a51620986514681c6cbad0'/>
<id>urn:sha1:b8cbab042cd69f3918a51620986514681c6cbad0</id>
<content type='text'>
This patch replaces the two current amdkfd module parameters with a new one.

The current parameters that are being replaced are:

- Maximum number of HSA processes
- Maximum number of queues per process

The new parameter that replaces them is called "Maximum queues per device"

This replacement achieves two goals:

- Allows the user to have as many HSA processes as it wants (until
  a maximum of 512 HSA processes in Kaveri).

- Removes the limitation the user had on maximum number of queues per HSA
  process. E.g. the user can now have processes which only have one queue and
  other processes which have hundreds of queues, while before the user
  couldn't have more than 128 queues per process (as default).

The default value of the new parameter is 4096 (32 * 128, which were the
defaults of the old parameters). There is almost no additional GART memory
required for the default case. As a reminder, this amount of queues requires a
little bit below 4MB of GART memory.

v2:
In addition, This patch defines a new counter for queues accounting in the DQM
structure. This is done because the current counter only counts active queues
which allows the user to create more queues than the
max_num_of_queues_per_device module parameter allows.

However, we need the current counter for the runlist packet build process, so
the solution is to have a dedicated counter for this accounting.

Signed-off-by: Oded Gabbay &lt;oded.gabbay@amd.com&gt;
Reviewed-by: Ben Goz &lt;ben.goz@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: PQM handle queue creation fault</title>
<updated>2015-01-15T15:14:47Z</updated>
<author>
<name>Ben Goz</name>
<email>ben.goz@amd.com</email>
</author>
<published>2015-01-15T15:14:47Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=f046bfdf735794d5893963427eb54302b7940a86'/>
<id>urn:sha1:f046bfdf735794d5893963427eb54302b7940a86</id>
<content type='text'>
If the first queue created was failed on DQM then PQM should
unregister the process from DQM.

Signed-off-by: Ben Goz &lt;ben.goz@amd.com&gt;
Signed-off-by: Oded Gabbay &lt;oded.gabbay@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: Fix sparse warning (different address space)</title>
<updated>2015-01-08T14:46:16Z</updated>
<author>
<name>Oded Gabbay</name>
<email>oded.gabbay@amd.com</email>
</author>
<published>2015-01-08T14:46:16Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=8dfe58b2063811b415626060316672741049d4d4'/>
<id>urn:sha1:8dfe58b2063811b415626060316672741049d4d4</id>
<content type='text'>
Signed-off-by: Oded Gabbay &lt;oded.gabbay@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: Drop interrupt SW ring buffer</title>
<updated>2015-01-08T04:27:15Z</updated>
<author>
<name>Michel Dänzer</name>
<email>michel.daenzer@amd.com</email>
</author>
<published>2015-01-08T04:27:15Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=6ee0ad2a7f27f7dc365576b748bc98684f02882d'/>
<id>urn:sha1:6ee0ad2a7f27f7dc365576b748bc98684f02882d</id>
<content type='text'>
The work queue couldn't reliably prevent the SW ring buffer from
overflowing, so dmesg was spammed by

 kfd kfd: Interrupt ring overflow, dropping interrupt.

messages when running e.g. the Atlantis Substance demo from
https://wiki.unrealengine.com/Linux_Demos on Kaveri.

Since the SW ring buffer doesn't actually do anything at this point, just
remove it for now. When actual interrupt processing code is added to
amdkfd, it should try to do things immediately and only defer to work
queues when necessary.

Signed-off-by: Michel Dänzer &lt;michel.daenzer@amd.com&gt;
Reviewed-by: Christian König &lt;christian.koenig@amd.com&gt;
Signed-off-by: Oded Gabbay &lt;oded.gabbay@amd.com&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: rewrite kfd_ioctl() according to drm_ioctl()</title>
<updated>2015-01-06T17:44:36Z</updated>
<author>
<name>Oded Gabbay</name>
<email>oded.gabbay@amd.com</email>
</author>
<published>2014-12-29T12:20:05Z</published>
<link rel='alternate' type='text/html' href='https://git.shady.money/linux/commit/?id=76baee6c733bfef30fcf86cbd121e336b839e408'/>
<id>urn:sha1:76baee6c733bfef30fcf86cbd121e336b839e408</id>
<content type='text'>
This patch changes kfd_ioctl() to be very similar to drm_ioctl().

The patch defines an array of amdkfd_ioctls, which maps IOCTL definition to the
ioctl function.

The kfd_ioctl() uses that mapping to call the appropriate ioctl function,
through a function pointer.

This patch also declares a new typedef for the ioctl function pointer.

v2: Renamed KFD_COMMAND_(START|END) to AMDKFD_...

Signed-off-by: Oded Gabbay &lt;oded.gabbay@amd.com&gt;
Acked-by: Christian König &lt;christian.koenig@amd.com&gt;
</content>
</entry>
</feed>
