linux-stable.git/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c, branch v6.7.8

drm/amdgpu: skip to program GFXDEC registers for suspend abort

2024-03-01T12:41:43+00:00

[ Upstream commit 93bafa32a6918154aa0caf9f66679a32c2431357 ]

In the suspend abort cases, the gfx power rail doesn't turn off so
some GFXDEC registers/CSB can't reset to default value and at this
moment reinitialize GFXDEC/CSB will result in an unexpected error.
So let skip those program sequence for the suspend abort case.

Signed-off-by: Prike Liang 
Reviewed-by: Alex Deucher 
Signed-off-by: Alex Deucher 
Signed-off-by: Sasha Levin

drm/amdgpu: Enable tunneling on high-priority compute queues

2024-02-01T00:21:17+00:00

[ Upstream commit 91963397c49aa2907aeafa52d929555dcbc9cd07 ]

This improves latency if the GPU is already busy with other work.
This is useful for VR compositors that submit highly latency-sensitive
compositing work on high-priority compute queues while the GPU is busy
rendering the next frame.

Userspace merge request:
https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26462

v2: bump driver version (Alex)

Reviewed-by: Marek Olšák 
Signed-off-by: Friedrich Vock 
Signed-off-by: Alex Deucher 
Stable-dep-of: 03ff6d7238b7 ("drm/amdgpu/gfx10: set UNORD_DISPATCH in compute MQDs")
Signed-off-by: Sasha Levin

Revert "drm/amd: Enable PCIe PME from D3"

2024-02-01T00:21:04+00:00

This reverts commit 05f7a3475af0faa8bf77f8637c4a40349db4f78f.

duplicated a change made in 6.7
6967741d26c87300a51b5e50d4acd104bc1a9759

Cc: stable@vger.kernel.org # 6.7
Signed-off-by: Jonathan Gray 
Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: revert "Adjust removal control flow for smu v13_0_2"

2024-01-25T23:45:16+00:00

commit fb1c93c2e9604a884467a773790016199f78ca08 upstream.

Calling amdgpu_device_ip_resume_phase1() during shutdown leaves the
HW in an active state and is an unbalanced use of the IP callbacks.

Using the IP callbacks like this can lead to memory leaks, double
free and imbalanced reference counters.

Leaving the HW in an active state can lead to DMA accesses to memory now
freed by the driver.

Both is a complete no-go for driver unload so completely revert the
workaround for now.

This reverts commit f5c7e7797060255dbc8160734ccc5ad6183c5e04.

Signed-off-by: Christian König 
Acked-by: Alex Deucher 
Signed-off-by: Alex Deucher 
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman

drm/amd: Enable PCIe PME from D3

2024-01-25T23:45:14+00:00

commit bd1f6a31e7762ebc99b97f3eda5e5ea3708fa792 upstream.

When dGPU is put into BOCO it may be in D3cold but still able send
PME on display hotplug event. For this to work it must be enabled
as wake source from D3.

When runpm is enabled use pci_wake_from_d3() to mark wakeup as
enabled by default.

Cc: stable@vger.kernel.org # 6.1+
Signed-off-by: Mario Limonciello 
Acked-by: Alex Deucher 
Signed-off-by: Alex Deucher 
Signed-off-by: Greg Kroah-Hartman

drm/amd: Enable PCIe PME from D3

2023-11-29T23:09:34+00:00

When dGPU is put into BOCO it may be in D3cold but still able send
PME on display hotplug event. For this to work it must be enabled
as wake source from D3.

When runpm is enabled use pci_wake_from_d3() to mark wakeup as
enabled by default.

Cc: stable@vger.kernel.org # 6.1+
Signed-off-by: Mario Limonciello 
Acked-by: Alex Deucher 
Signed-off-by: Alex Deucher

drm/amdgpu: add a module parameter to control the AGP aperture

2023-11-17T05:58:20+00:00

Add a module parameter to control the AGP aperture.  The AGP
aperture is an aperture in the GPU's internal address space
which provides direct non-paged access to the platform address
space.  This access is non-snooped so only uncached memory
can be accessed.

Add a knob so that we can toggle this for debugging.

Fixes: 67318cb84341 ("drm/amdgpu/gmc11: set gart placement GC11")
Acked-by: Christian König 
Reviewed-by: Mario Limonciello 
Tested-by: Mario Limonciello  # PHX & Navi33
Signed-off-by: Alex Deucher

drm/amdgpu: Optimize the asic type fix code

2023-11-03T16:18:32+00:00

Use a new struct array to define the asic information which
asic type needs to be fixed.

Signed-off-by: Ma Jun 
Reviewed-by: Kenneth Feng 
Signed-off-by: Alex Deucher

drm/amd: Evict resources during PM ops prepare() callback

2023-10-13T15:00:18+00:00

Linux PM core has a prepare() callback run before suspend.

If the system is under high memory pressure, the resources may need
to be evicted into swap instead.  If the storage backing for swap
is offlined during the suspend() step then such a call may fail.

So move this step into prepare() to move evict majority of
resources and update all non-pmops callers to call the same callback.

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2362
Reviewed-by: Christian König 
Reviewed-by: Alex Deucher 
Signed-off-by: Mario Limonciello 
Signed-off-by: Alex Deucher

drm/amdgpu: update ib start and size alignment

2023-10-09T20:51:39+00:00

Update IB starting address alignment and size alignment with correct values
for decode and encode IPs.

Decode IB starting address alignment: 256 bytes
Decode IB size alignment: 64 bytes
Encode IB starting address alignment: 256 bytes
Encode IB size alignment: 4 bytes

Also bump amdgpu driver version for this update.

Signed-off-by: Boyuan Zhang 
Reviewed-by: Christian König 
Acked-by: Alex Deucher 
Signed-off-by: Alex Deucher