<feed xmlns='http://www.w3.org/2005/Atom'>
<title>user/sven/linux.git/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h, branch v5.4.36</title>
<subtitle>Linux Kernel
</subtitle>
<id>https://git.stealer.net/cgit.cgi/user/sven/linux.git/atom?h=v5.4.36</id>
<link rel='self' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/atom?h=v5.4.36'/>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/'/>
<updated>2019-07-18T19:18:04Z</updated>
<entry>
<title>drm/amdkfd: Implement kfd2kgd_calls for Arcturus</title>
<updated>2019-07-18T19:18:04Z</updated>
<author>
<name>Oak Zeng</name>
<email>Oak.Zeng@amd.com</email>
</author>
<published>2019-07-09T14:59:30Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=3e205a0849a760166578b4d95b17e904f23d962e'/>
<id>urn:sha1:3e205a0849a760166578b4d95b17e904f23d962e</id>
<content type='text'>
Arcturus shares most of the kfd2kgd_calls with gfx9. But due to
SDMA register address change, it can't share SDMA related functions.
Export gfx9 kfd2kgd_calls and implement SDMA related functions
for Arcturus.

Signed-off-by: Oak Zeng &lt;Oak.Zeng@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: remove duplicated PCIE atomics request</title>
<updated>2019-07-01T19:54:47Z</updated>
<author>
<name>Jack Xiao</name>
<email>Jack.Xiao@amd.com</email>
</author>
<published>2019-05-29T05:52:17Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=aabf3a951c4e959a00e96e9ef11828149dfc81a8'/>
<id>urn:sha1:aabf3a951c4e959a00e96e9ef11828149dfc81a8</id>
<content type='text'>
Since amdgpu has always requested PCIE atomics, kfd don't
need duplicated PCIE atomics enablement. Referring to amdgpu
request result is enough.

Signed-off-by: Jack Xiao &lt;Jack.Xiao@amd.com&gt;
Reviewed-by: Hawking Zhang &lt;Hawking.Zhang@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: Add navi10 kfd support for amdgpu (v3)</title>
<updated>2019-06-21T23:59:23Z</updated>
<author>
<name>Hawking Zhang</name>
<email>Hawking.Zhang@amd.com</email>
</author>
<published>2019-03-05T11:59:30Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=6bdadb2072243154e6441bfd789dce1fc941976c'/>
<id>urn:sha1:6bdadb2072243154e6441bfd789dce1fc941976c</id>
<content type='text'>
KFD (Kernel Fusion Driver) is the compute backend driver
for AMD GPUs.

v2: squash in updates (Alex)
v3: fix warnings (Alex)

Signed-off-by: Oak Zeng &lt;Oak.Zeng@amd.com&gt;
Signed-off-by: Philip Cox &lt;Philip.Cox@amd.com&gt;
Signed-off-by: Hawking Zhang &lt;Hawking.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: Add function to add/remove gws to kfd process</title>
<updated>2019-05-28T19:44:18Z</updated>
<author>
<name>Oak Zeng</name>
<email>Oak.Zeng@amd.com</email>
</author>
<published>2019-05-08T21:14:45Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=71efab6a30ab392af6a3fe1ce06bcf25c82ff3cb'/>
<id>urn:sha1:71efab6a30ab392af6a3fe1ce06bcf25c82ff3cb</id>
<content type='text'>
GWS bo is shared between all kfd processes. Add function to add gws
to kfd process's bo list so gws can be evicted from and restored
for process.

Signed-off-by: Oak Zeng &lt;Oak.Zeng@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: Add interface to alloc gws from amdgpu</title>
<updated>2019-05-28T19:44:05Z</updated>
<author>
<name>Oak Zeng</name>
<email>Oak.Zeng@amd.com</email>
</author>
<published>2019-05-06T14:48:37Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=ca66fb8fbb9b9690591b4e85707a4f31cb042adf'/>
<id>urn:sha1:ca66fb8fbb9b9690591b4e85707a4f31cb042adf</id>
<content type='text'>
Add amdgpu_amdkfd interface to alloc and free gws
from amdgpu

Signed-off-by: Oak Zeng &lt;Oak.Zeng@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: Add gws number to kfd topology node properties</title>
<updated>2019-05-28T19:43:58Z</updated>
<author>
<name>Oak Zeng</name>
<email>Oak.Zeng@amd.com</email>
</author>
<published>2019-05-03T14:10:38Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=29e764621bbcd4a93745cf1b9fd5a2f5f45c3f72'/>
<id>urn:sha1:29e764621bbcd4a93745cf1b9fd5a2f5f45c3f72</id>
<content type='text'>
Add amdgpu_amdkfd interface to get num_gws and add num_gws
to /sys/class/kfd/kfd/topology/nodes/x/properties. Only report
num_gws if MEC FW support GWS barriers. Currently it is
determined by a module parameter which will be replaced
with MEC FW version check when firmware is ready.

Signed-off-by: Oak Zeng &lt;Oak.Zeng@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: replace get_user_pages with HMM mirror helpers</title>
<updated>2019-05-24T17:21:02Z</updated>
<author>
<name>Philip Yang</name>
<email>Philip.Yang@amd.com</email>
</author>
<published>2018-12-13T20:35:28Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=899fbde1464639e3d12eaffdad8481a59b367fcb'/>
<id>urn:sha1:899fbde1464639e3d12eaffdad8481a59b367fcb</id>
<content type='text'>
Use HMM helper function hmm_vma_fault() to get physical pages backing
userptr and start CPU page table update track of those pages. Then use
hmm_vma_range_done() to check if those pages are updated before
amdgpu_cs_submit for gfx or before user queues are resumed for kfd.

If userptr pages are updated, for gfx, amdgpu_cs_ioctl will restart
from scratch, for kfd, restore worker is rescheduled to retry.

HMM simplify the CPU page table concurrent update check, so remove
guptasklock, mmu_invalidations, last_set_pages fields from
amdgpu_ttm_tt struct.

HMM does not pin the page (increase page ref count), so remove related
operations like release_pages(), put_page(), mark_page_dirty().

Signed-off-by: Philip Yang &lt;Philip.Yang@amd.com&gt;
Reviewed-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: Implement get num of hops between two xgmi device</title>
<updated>2019-05-24T17:20:48Z</updated>
<author>
<name>shaoyunl</name>
<email>shaoyun.liu@amd.com</email>
</author>
<published>2019-04-17T18:28:18Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=da361dd13f4fb766445f55bbd2eec36c73a9b1bb'/>
<id>urn:sha1:da361dd13f4fb766445f55bbd2eec36c73a9b1bb</id>
<content type='text'>
KFD need to provide the info for upper level to determine the data path

Signed-off-by: shaoyunl &lt;shaoyun.liu@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: Expose HDP registers to user space</title>
<updated>2019-05-24T17:20:47Z</updated>
<author>
<name>Oak Zeng</name>
<email>Oak.Zeng@amd.com</email>
</author>
<published>2019-04-11T19:43:39Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=d8e408a82704c86ba87c3d58cfe69dcdb758aa07'/>
<id>urn:sha1:d8e408a82704c86ba87c3d58cfe69dcdb758aa07</id>
<content type='text'>
Introduce a new memory type (KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP) and
expose mmio page of HDP registers to user space through this new
memory type.

v2: moved remapped hdp regs to adev struct
v3: rename the new memory type to ALLOC_MEM_FLAGS_MMIO_REMAP
v4: use more generic function name
v5: Fail remapped mmio allocation for asics before gfx9

Signed-off-by: Oak Zeng &lt;Oak.Zeng@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: get_fw_version isn't ASIC specific</title>
<updated>2019-04-19T16:32:40Z</updated>
<author>
<name>Amber Lin</name>
<email>Amber.Lin@amd.com</email>
</author>
<published>2019-04-12T15:07:16Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=0da8b10e36e91b1c68b88cdcc6efb285a3b386ed'/>
<id>urn:sha1:0da8b10e36e91b1c68b88cdcc6efb285a3b386ed</id>
<content type='text'>
Method of getting firmware version is the same across ASICs, so remove
them from ASIC-specific files and create one in amdgpu_amdkfd.c. This new
created get_fw_version simply reads fw_version from adev-&gt;gfx than parsing
the ucode header.

Signed-off-by: Amber Lin &lt;Amber.Lin@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>
</feed>
