<feed xmlns='http://www.w3.org/2005/Atom'>
<title>kernel.git/drivers/gpu/drm/amd/amdkfd, branch linux-5.19.y</title>
<subtitle>Hosts the 0x221E linux distro kernel.
</subtitle>
<id>https://git.0xinfinity.dev/distro/kernel.git/atom?h=linux-5.19.y</id>
<link rel='self' href='https://git.0xinfinity.dev/distro/kernel.git/atom?h=linux-5.19.y'/>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/'/>
<updated>2022-10-24T07:58:20Z</updated>
<entry>
<title>drm/amdkfd: Fix UBSAN shift-out-of-bounds warning</title>
<updated>2022-10-24T07:58:20Z</updated>
<author>
<name>Felix Kuehling</name>
</author>
<published>2022-09-21T21:45:59Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=5861c094874dc8cdb271f4b40c567afb9267bd24'/>
<id>urn:sha1:5861c094874dc8cdb271f4b40c567afb9267bd24</id>
<content type='text'>
[ Upstream commit b292cafe2dd02d96a07147e4b160927e8399d5cc ]

This was fixed in initialize_cpsch before, but not in initialize_nocpsch.
Factor sdma bitmap initialization into a helper function to apply the
correct implementation in both cases without duplicating it.

v2: Added a range check

Reported-by: Ellis Michael &lt;ellis@ellismichael.com&gt;
Signed-off-by: Felix Kuehling &lt;Felix.Kuehling@amd.com&gt;
Reviewed-by: Graham Sider &lt;Graham.Sider@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Signed-off-by: Sasha Levin &lt;sashal@kernel.org&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: Handle restart of kfd_ioctl_wait_events</title>
<updated>2022-09-05T08:31:32Z</updated>
<author>
<name>Felix Kuehling</name>
</author>
<published>2022-08-04T22:19:38Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=21e3891abc83f2d5482c67cca2fb7f228635daaf'/>
<id>urn:sha1:21e3891abc83f2d5482c67cca2fb7f228635daaf</id>
<content type='text'>
[ Upstream commit bea9a56afbc4b5a41ea579b8b0dc5e189b439504 ]

When kfd_ioctl_wait_events needs to restart due to a signal, we need to
update the timeout to account for the time already elapsed. We also need
to undo auto_reset of events that have signaled already, so that the
restarted ioctl will be able to count those signals again.

This fixes infinite hangs when kfd_ioctl_wait_events is interrupted by a
signal.

Signed-off-by: Felix Kuehling &lt;Felix.Kuehling@amd.com&gt;
Reviewed-and-tested-by: Xiaogang Chen &lt;Xiaogang.Chen@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Signed-off-by: Sasha Levin &lt;sashal@kernel.org&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: Fix isa version for the GC 10.3.7</title>
<updated>2022-08-31T15:18:18Z</updated>
<author>
<name>Prike Liang</name>
</author>
<published>2022-08-24T03:16:51Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=43365c8fbb3ca6d60ecb32b5c0f91e1563dd0ac1'/>
<id>urn:sha1:43365c8fbb3ca6d60ecb32b5c0f91e1563dd0ac1</id>
<content type='text'>
commit ee8086dbc1585d9f4020a19447388246a5cff5c8 upstream.

Correct the isa version for handling KFD test.

Fixes: 7c4f4f197e0c ("drm/amdkfd: Add GC 10.3.6 and 10.3.7 KFD definitions")
Signed-off-by: Prike Liang &lt;Prike.Liang@amd.com&gt;
Reviewed-by: Aaron Liu &lt;aaron.liu@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: correct sdma queue number of sdma 6.0.1</title>
<updated>2022-08-17T13:14:33Z</updated>
<author>
<name>Yifan Zhang</name>
</author>
<published>2022-06-20T08:24:58Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=ca456c5104d47e021f749553ac25627e803c2d59'/>
<id>urn:sha1:ca456c5104d47e021f749553ac25627e803c2d59</id>
<content type='text'>
[ Upstream commit efb4fd107cfd9748f777a4e9015d803d3c9db68b ]

sdma 6.0.1 has 8 queues instead of 2.

Fixes: 26776a7031c423 ("drm/amdkfd: add GC 11.0.1 KFD support")
Signed-off-by: Yifan Zhang &lt;yifan1.zhang@amd.com&gt;
Acked-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Reviewed-by: Tim Huang &lt;Tim.Huang@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Signed-off-by: Sasha Levin &lt;sashal@kernel.org&gt;
</content>
</entry>
<entry>
<title>drm/amdkfd: correct the MEC atomic support firmware checking for GC 10.3.7</title>
<updated>2022-07-13T16:20:36Z</updated>
<author>
<name>Prike Liang</name>
</author>
<published>2022-07-11T08:03:08Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=c0044865480a162146b9dfe7783e73a08e97b2b9'/>
<id>urn:sha1:c0044865480a162146b9dfe7783e73a08e97b2b9</id>
<content type='text'>
On the GC 10.3.7 platform the initial MEC release version #3 can support
atomic operation,so need correct and set its MEC atomic support version to #3.

Signed-off-by: Prike Liang &lt;Prike.Liang@amd.com&gt;
Reviewed-by: Aaron Liu &lt;aaron.liu@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Cc: stable@vger.kernel.org # 5.18.x
</content>
</entry>
<entry>
<title>drm/amdkfd:Fix fw version for 10.3.6</title>
<updated>2022-06-08T15:24:00Z</updated>
<author>
<name>Jesse Zhang</name>
</author>
<published>2022-06-07T02:44:57Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=a956a11ee669d069047525c8ec897b4c21a9cda1'/>
<id>urn:sha1:a956a11ee669d069047525c8ec897b4c21a9cda1</id>
<content type='text'>
fix fw error when loading fw for 10.3.6

Signed-off-by: Jesse Zhang &lt;Jesse.Zhang@amd.com&gt;
Reviewed-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Reviewed-by: Mario Limonciello &lt;mario.limonciello@amd.com&gt;
Reviewed-by: Felix Kuehling &lt;Felix.Kuehling@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Cc: stable@vger.kernel.org # 5.18.x
</content>
</entry>
<entry>
<title>drm/amdkfd: Fix partial migration bugs</title>
<updated>2022-06-03T20:43:22Z</updated>
<author>
<name>Philip Yang</name>
</author>
<published>2022-06-03T13:19:34Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=88467db6e2f46a2e79b1b67ce6873c284e4cf417'/>
<id>urn:sha1:88467db6e2f46a2e79b1b67ce6873c284e4cf417</id>
<content type='text'>
Migration range from system memory to VRAM, if system page can not be
locked or unmapped, we do partial migration and leave some pages in
system memory. Several bugs found to copy pages and update GPU mapping
for this situation:

1. copy to vram should use migrate-&gt;npage which is total pages of range
as npages, not migrate-&gt;cpages which is number of pages can be migrated.

2. After partial copy, set VRAM res cursor as j + 1, j is number of
system pages copied plus 1 page to skip copy.

3. copy to ram, should collect all continuous VRAM pages and copy
together.

4. Call amdgpu_vm_update_range, should pass in offset as bytes, not
as number of pages.

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;
Cc: stable@vger.kernel.org
</content>
</entry>
<entry>
<title>drm/amdkfd: Add GC 10.3.6 and 10.3.7 KFD definitions</title>
<updated>2022-06-03T20:26:36Z</updated>
<author>
<name>Mario Limonciello</name>
</author>
<published>2022-05-31T23:56:41Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=7c4f4f197e0c5c93a70329627f17fcc5883f3593'/>
<id>urn:sha1:7c4f4f197e0c5c93a70329627f17fcc5883f3593</id>
<content type='text'>
Loading amdgpu on GC 10.3.7 shows an ERR level message:
`kfd kfd: amdgpu: GC IP 0a0307 not supported in kfd`

Add these targets to match yellow carp structures.

Reported-by: David Chang &lt;david.chang@amd.com&gt;
Reviewed-by: Felix Kuehling &lt;Felix.Kuehling@amd.com&gt;
Tested-by: Jesse(Jie) Zhang &lt;Jesse.Zhang@amd.com&gt;
Signed-off-by: Mario Limonciello &lt;mario.limonciello@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Cc: stable@vger.kernel.org # 5.18.x
</content>
</entry>
<entry>
<title>drm/amdkfd: Use mmget_not_zero in MMU notifier</title>
<updated>2022-06-01T20:04:41Z</updated>
<author>
<name>Philip Yang</name>
</author>
<published>2022-05-26T20:15:38Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=fa582c6f3684ac0098a9d02ddf0ed52a02b37127'/>
<id>urn:sha1:fa582c6f3684ac0098a9d02ddf0ed52a02b37127</id>
<content type='text'>
MMU notifier callback may pass in mm with mm-&gt;mm_users==0 when process
is exiting, use mmget_no_zero to avoid accessing invalid mm in deferred
list work after mm is gone.

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/amdgpu: add AMDGPU_GEM_CREATE_DISCARDABLE</title>
<updated>2022-05-26T18:56:34Z</updated>
<author>
<name>Christian König</name>
</author>
<published>2022-05-06T11:08:34Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=fab2cc8335839867a3db38f195441b9c7c6460f6'/>
<id>urn:sha1:fab2cc8335839867a3db38f195441b9c7c6460f6</id>
<content type='text'>
Add a AMDGPU_GEM_CREATE_DISCARDABLE flag to note that the content of a BO
doesn't needs to be preserved during eviction.

KFD was already using a similar functionality for SVM BOs so replace the
internal flag with the new UAPI.

Signed-off-by: Christian König &lt;christian.koenig@amd.com&gt;
Reviewed-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Reviewed-by: Marek Olšák &lt;marek.olsak@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
</entry>
</feed>
