<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux.git/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h, branch v6.15</title>
<subtitle>Linux kernel source tree</subtitle>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/'/>
<entry>
<title>drm/amd/amdgpu: Revert "drm/amd/amdgpu: shorten the gfx idle worker timeout"</title>
<updated>2025-03-21T16:15:08+00:00</updated>
<author>
<name>Kenneth Feng</name>
<email>kenneth.feng@amd.com</email>
</author>
<published>2025-03-19T06:08:14+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=a67f0094c9d7a0310321d154e5b370a098759661'/>
<id>a67f0094c9d7a0310321d154e5b370a098759661</id>
<content type='text'>
This reverts commit 55ff973fe1c053de143969cfc8b34baff084084a.

Reason for revert: this causes some tests fail with call trace.

Signed-off-by: Kenneth Feng &lt;kenneth.feng@amd.com&gt;
Acked-by: Yang Wang &lt;kevinyang.wang@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This reverts commit 55ff973fe1c053de143969cfc8b34baff084084a.

Reason for revert: this causes some tests fail with call trace.

Signed-off-by: Kenneth Feng &lt;kenneth.feng@amd.com&gt;
Acked-by: Yang Wang &lt;kevinyang.wang@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/amdgpu/gfx: fix ref counting for ring based profile handling</title>
<updated>2025-03-19T19:52:56+00:00</updated>
<author>
<name>Alex Deucher</name>
<email>alexander.deucher@amd.com</email>
</author>
<published>2025-03-12T13:44:19+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=553673a3e1bc0abbb994d9884f772189c739e3a0'/>
<id>553673a3e1bc0abbb994d9884f772189c739e3a0</id>
<content type='text'>
We need to make sure the workload profile ref counts are
balanced.  This isn't currently the case because we can
increment the count on submissions, but the decrement may
be delayed as work comes in.  Track when we enable the
workload profile so the references are balanced.

v2: switch to a mutex and active flag
v3: fix mutex init

Fixes: 8fdb3958e396 ("drm/amdgpu/gfx: add ring helpers for setting workload profile")
Cc: Yang Wang &lt;kevinyang.wang@amd.com&gt;
Cc: Kenneth Feng &lt;kenneth.feng@amd.com&gt;
Tested-by: Kenneth Feng &lt;kenneth.feng@amd.com&gt;
Reviewed-by: Kenneth Feng &lt;kenneth.feng@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
We need to make sure the workload profile ref counts are
balanced.  This isn't currently the case because we can
increment the count on submissions, but the decrement may
be delayed as work comes in.  Track when we enable the
workload profile so the references are balanced.

v2: switch to a mutex and active flag
v3: fix mutex init

Fixes: 8fdb3958e396 ("drm/amdgpu/gfx: add ring helpers for setting workload profile")
Cc: Yang Wang &lt;kevinyang.wang@amd.com&gt;
Cc: Kenneth Feng &lt;kenneth.feng@amd.com&gt;
Tested-by: Kenneth Feng &lt;kenneth.feng@amd.com&gt;
Reviewed-by: Kenneth Feng &lt;kenneth.feng@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/amd/amdgpu: shorten the gfx idle worker timeout</title>
<updated>2025-03-18T18:03:38+00:00</updated>
<author>
<name>Kenneth Feng</name>
<email>kenneth.feng@amd.com</email>
</author>
<published>2025-03-11T08:46:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=55ff973fe1c053de143969cfc8b34baff084084a'/>
<id>55ff973fe1c053de143969cfc8b34baff084084a</id>
<content type='text'>
Shorten the gfx idle worker timeout. This is to sync with
DAL when there is no activity on the screen. Original 1
second can not sync with DAL, so DAL can not apply MALL
when the workload type is not bootup default.

Signed-off-by: Kenneth Feng &lt;kenneth.feng@amd.com&gt;
Reviewed-by: Yang Wang &lt;kevinyang.wang@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Shorten the gfx idle worker timeout. This is to sync with
DAL when there is no activity on the screen. Original 1
second can not sync with DAL, so DAL can not apply MALL
when the workload type is not bootup default.

Signed-off-by: Kenneth Feng &lt;kenneth.feng@amd.com&gt;
Reviewed-by: Yang Wang &lt;kevinyang.wang@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/amdgpu/gfx: add amdgpu_gfx_off_ctrl_immediate()</title>
<updated>2025-02-13T02:04:07+00:00</updated>
<author>
<name>Alex Deucher</name>
<email>alexander.deucher@amd.com</email>
</author>
<published>2025-01-31T02:20:57+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=250d9769eeee6cfebbafaca3e4f34a907b5b166f'/>
<id>250d9769eeee6cfebbafaca3e4f34a907b5b166f</id>
<content type='text'>
Same as amdgpu_gfx_off_ctrl(), but without the delay
for gfxoff disallow.

Reviewed-by: Lijo Lazar &lt;lijo.lazar@amd.com&gt;
Suggested-by: Błażej Szczygieł &lt;mumei6102@gmail.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Same as amdgpu_gfx_off_ctrl(), but without the delay
for gfxoff disallow.

Reviewed-by: Lijo Lazar &lt;lijo.lazar@amd.com&gt;
Suggested-by: Błażej Szczygieł &lt;mumei6102@gmail.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/amdgpu/gfx: add ring helpers for setting workload profile</title>
<updated>2025-02-13T02:02:55+00:00</updated>
<author>
<name>Alex Deucher</name>
<email>alexander.deucher@amd.com</email>
</author>
<published>2025-01-08T22:15:56+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=8fdb3958e3965fcd11408e1cd5380895cdc96e06'/>
<id>8fdb3958e3965fcd11408e1cd5380895cdc96e06</id>
<content type='text'>
Add helpers to switch the workload profile dynamically when
commands are submitted.  This allows us to switch to
the FULLSCREEN3D or COMPUTE profile when work is submitted.
Add a delayed work handler to delay switching out of the
selected profile if additional work comes in.  This works
the same as the VIDEO profile for VCN.  This lets dynamically
enable workload profiles on the fly and then move back
to the default when there is no work.

Reviewed-by: Lijo Lazar &lt;lijo.lazar@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Add helpers to switch the workload profile dynamically when
commands are submitted.  This allows us to switch to
the FULLSCREEN3D or COMPUTE profile when work is submitted.
Add a delayed work handler to delay switching out of the
selected profile if additional work comes in.  This works
the same as the VIDEO profile for VCN.  This lets dynamically
enable workload profiles on the fly and then move back
to the default when there is no work.

Reviewed-by: Lijo Lazar &lt;lijo.lazar@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/amdgpu: Add sysfs interface for gc reset mask</title>
<updated>2024-11-08T16:08:01+00:00</updated>
<author>
<name>Jesse.zhang@amd.com</name>
<email>Jesse.zhang@amd.com</email>
</author>
<published>2024-11-05T07:22:56+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=6c8d1f4b042e706ccd7575beb0397a75d545d71b'/>
<id>6c8d1f4b042e706ccd7575beb0397a75d545d71b</id>
<content type='text'>
Add two sysfs interfaces for gfx and compute:
gfx_reset_mask
compute_reset_mask

These interfaces are read-only and show the resets supported by the IP.
For example, full adapter reset (mode1/mode2/BACO/etc),
soft reset, queue reset, and pipe reset.

V2: the sysfs node returns a text string instead of some flags (Christian)
v3: add a generic helper which takes the ring as parameter
    and print the strings in the order they are applied (Christian)

    check amdgpu_gpu_recovery  before creating sysfs file itself,
    and initialize supported_reset_types in IP version files (Lijo)
v4: Fixing uninitialized variables (Tim)

Signed-off-by: Jesse Zhang &lt;Jesse.Zhang@amd.com&gt;
Suggested-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;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Add two sysfs interfaces for gfx and compute:
gfx_reset_mask
compute_reset_mask

These interfaces are read-only and show the resets supported by the IP.
For example, full adapter reset (mode1/mode2/BACO/etc),
soft reset, queue reset, and pipe reset.

V2: the sysfs node returns a text string instead of some flags (Christian)
v3: add a generic helper which takes the ring as parameter
    and print the strings in the order they are applied (Christian)

    check amdgpu_gpu_recovery  before creating sysfs file itself,
    and initialize supported_reset_types in IP version files (Lijo)
v4: Fixing uninitialized variables (Tim)

Signed-off-by: Jesse Zhang &lt;Jesse.Zhang@amd.com&gt;
Suggested-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;
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/amdgpu: Group gfx sysfs functions</title>
<updated>2024-11-04T17:06:23+00:00</updated>
<author>
<name>Lijo Lazar</name>
<email>lijo.lazar@amd.com</email>
</author>
<published>2024-10-29T05:17:26+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=047767ddc93666704026c79c01554597375beb50'/>
<id>047767ddc93666704026c79c01554597375beb50</id>
<content type='text'>
Make amdgpu_gfx_sysfs_init/fini functions as common entry points for all
gfx related sysfs nodes.

Signed-off-by: Lijo Lazar &lt;lijo.lazar@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>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Make amdgpu_gfx_sysfs_init/fini functions as common entry points for all
gfx related sysfs nodes.

Signed-off-by: Lijo Lazar &lt;lijo.lazar@amd.com&gt;
Reviewed-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/amdgpu: add amdgpu_gfx_sched_mask and amdgpu_compute_sched_mask debugfs</title>
<updated>2024-11-04T17:05:53+00:00</updated>
<author>
<name>Jesse Zhang</name>
<email>jesse.zhang@amd.com</email>
</author>
<published>2024-10-29T02:11:05+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=c5c63d9cb5d3bbb2fc5973757616b17629795829'/>
<id>c5c63d9cb5d3bbb2fc5973757616b17629795829</id>
<content type='text'>
compute/gfx may have multiple rings on some hardware.
In some cases, userspace wants to run jobs on a specific ring for validation purposes.
This debugfs entry helps to disable or enable submitting jobs to a specific ring.
This entry is populated only if there are at least two or more cores in the gfx/compute ip.

Signed-off-by: Jesse Zhang &lt;jesse.zhang@amd.com&gt;
Suggested-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;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
compute/gfx may have multiple rings on some hardware.
In some cases, userspace wants to run jobs on a specific ring for validation purposes.
This debugfs entry helps to disable or enable submitting jobs to a specific ring.
This entry is populated only if there are at least two or more cores in the gfx/compute ip.

Signed-off-by: Jesse Zhang &lt;jesse.zhang@amd.com&gt;
Suggested-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;
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/amdgpu: fix fairness in enforce isolation handling</title>
<updated>2024-11-04T16:27:04+00:00</updated>
<author>
<name>Alex Deucher</name>
<email>alexander.deucher@amd.com</email>
</author>
<published>2024-10-14T15:58:34+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=efe6a8774375ddcbdd46fb920be55cc2d0120836'/>
<id>efe6a8774375ddcbdd46fb920be55cc2d0120836</id>
<content type='text'>
Make sure KFD gets a turn when serializing access to
the GC IP.  Currently non-KFD jobs can starve KFD if they
submit often enough.  This patch prevents that by stalling
non-KFD if its time period has elapsed.

v2: fix units
v3: check enablement properly

Acked-by: Srinivasan Shanmugam &lt;srinivasan.shanmugam@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Make sure KFD gets a turn when serializing access to
the GC IP.  Currently non-KFD jobs can starve KFD if they
submit often enough.  This patch prevents that by stalling
non-KFD if its time period has elapsed.

v2: fix units
v3: check enablement properly

Acked-by: Srinivasan Shanmugam &lt;srinivasan.shanmugam@amd.com&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>drm/amdgpu: Remove unused amdgpu_gfx_bit_to_me_queue</title>
<updated>2024-09-26T21:06:33+00:00</updated>
<author>
<name>Dr. David Alan Gilbert</name>
<email>linux@treblig.org</email>
</author>
<published>2024-09-23T01:24:45+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=9d7a8bdb909e9b34bced6f56d4f2d320b0021697'/>
<id>9d7a8bdb909e9b34bced6f56d4f2d320b0021697</id>
<content type='text'>
amdgpu_gfx_bit_to_me_queue has been unused since it was added in
commit 7470bfcf2014 ("drm/amdgpu: add helper function for gfx queue/bitmap
transition")

Remove it.

Signed-off-by: Dr. David Alan Gilbert &lt;linux@treblig.org&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
amdgpu_gfx_bit_to_me_queue has been unused since it was added in
commit 7470bfcf2014 ("drm/amdgpu: add helper function for gfx queue/bitmap
transition")

Remove it.

Signed-off-by: Dr. David Alan Gilbert &lt;linux@treblig.org&gt;
Signed-off-by: Alex Deucher &lt;alexander.deucher@amd.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
