linux-stable.git/drivers/gpu/drm/amd/amdgpu, branch v6.5.7 Linux kernel stable tree drm/amd: Fix detection of _PR3 on the PCIe root port 2023-10-10T20:02:57+00:00 Mario Limonciello mario.limonciello@amd.com 2023-09-26T22:59:53+00:00 12c92dd1d1d195311788a18063650c511c45d9a1 commit 134b8c5d8674e7cde380f82e9aedfd46dcdd16f7 upstream. On some systems with Navi3x dGPU will attempt to use BACO for runtime PM but fails to resume properly. This is because on these systems the root port goes into D3cold which is incompatible with BACO. This happens because in this case dGPU is connected to a bridge between root port which causes BOCO detection logic to fail. Fix the intent of the logic by looking at root port, not the immediate upstream bridge for _PR3. Cc: stable@vger.kernel.org Suggested-by: Jun Ma <Jun.Ma2@amd.com> Tested-by: David Perry <David.Perry@amd.com> Fixes: b10c1c5b3a4e ("drm/amdgpu: add check for ACPI power resources") Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 134b8c5d8674e7cde380f82e9aedfd46dcdd16f7 upstream.

On some systems with Navi3x dGPU will attempt to use BACO for runtime
PM but fails to resume properly.  This is because on these systems
the root port goes into D3cold which is incompatible with BACO.

This happens because in this case dGPU is connected to a bridge between
root port which causes BOCO detection logic to fail.  Fix the intent of
the logic by looking at root port, not the immediate upstream bridge for
_PR3.

Cc: stable@vger.kernel.org
Suggested-by: Jun Ma <Jun.Ma2@amd.com>
Tested-by: David Perry <David.Perry@amd.com>
Fixes: b10c1c5b3a4e ("drm/amdgpu: add check for ACPI power resources")
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drm/amdgpu: Handle null atom context in VBIOS info ioctl 2023-10-06T11:16:12+00:00 David Francis David.Francis@amd.com 2023-09-05T14:13:51+00:00 109063d6b222613daacbcde57b3dc99ff40fee15 [ Upstream commit 5e7e82254270c8cf8b107451c5de01cee2f135ae ] On some APU systems, there is no atom context and so the atom_context struct is null. Add a check to the VBIOS_INFO branch of amdgpu_info_ioctl to handle this case, returning all zeroes. Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: David Francis <David.Francis@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 5e7e82254270c8cf8b107451c5de01cee2f135ae ]

On some APU systems, there is no atom context and so the
atom_context struct is null.

Add a check to the VBIOS_INFO branch of amdgpu_info_ioctl
to handle this case, returning all zeroes.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: David Francis <David.Francis@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drm/amdgpu: fallback to old RAS error message for aqua_vanjaram 2023-10-06T11:16:12+00:00 Hawking Zhang Hawking.Zhang@amd.com 2023-09-08T13:21:55+00:00 ea3f2bff5ab7a719f8caab28384304bf97fb7ef4 [ Upstream commit ffd6bde302061aeee405ab364403af30210f0b99 ] So driver doesn't generate incorrect message until the new format is settled down for aqua_vanjaram Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Yang Wang <kevinyang.wang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit ffd6bde302061aeee405ab364403af30210f0b99 ]

So driver doesn't generate incorrect message until
the new format is settled down for aqua_vanjaram

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drm/amdgpu/nbio4.3: set proper rmmio_remap.reg_offset for SR-IOV 2023-10-06T11:16:12+00:00 Alex Deucher alexander.deucher@amd.com 2023-09-07T19:44:54+00:00 a663c199b4d63ea1f90a37f6329c1a0af9437390 [ Upstream commit ab43213e7afd08ac68d4282060bacf309e70fd14 ] Needed for HDP flush to work correctly. Reviewed-by: Timmy Tsai <timmtsai@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit ab43213e7afd08ac68d4282060bacf309e70fd14 ]

Needed for HDP flush to work correctly.

Reviewed-by: Timmy Tsai <timmtsai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drm/amdgpu/soc21: don't remap HDP registers for SR-IOV 2023-10-06T11:16:11+00:00 Alex Deucher alexander.deucher@amd.com 2023-09-06T15:35:04+00:00 c634d78245b2a4ada1110837808925a1904e9a68 [ Upstream commit 1832403cd41ca6b19b24e9d64f79cb08d920ca44 ] This matches the behavior for soc15 and nv. Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Timmy Tsai <timmtsai@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 1832403cd41ca6b19b24e9d64f79cb08d920ca44 ]

This matches the behavior for soc15 and nv.

Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Timmy Tsai <timmtsai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drm/amdgpu: Store CU info from all XCCs for GFX v9.4.3 2023-10-06T11:16:11+00:00 Mukul Joshi mukul.joshi@amd.com 2023-08-25T15:59:09+00:00 830807d1fb11e3cbf7cf9091e820aa34a8b9c02d [ Upstream commit 97e3c6a853f2af9145daf0c6ca25bcdf55c759d4 ] Currently, we store CU info only for a single XCC assuming that it is the same for all XCCs. However, that may not be true. As a result, store CU info for all XCCs. This info is later used for CU masking. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 97e3c6a853f2af9145daf0c6ca25bcdf55c759d4 ]

Currently, we store CU info only for a single XCC assuming
that it is the same for all XCCs. However, that may not be
true. As a result, store CU info for all XCCs. This info is
later used for CU masking.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drm/amdgpu: set completion status as preempted for the resubmission 2023-10-06T11:16:00+00:00 Jiadong Zhu Jiadong.Zhu@amd.com 2023-07-26T07:21:48+00:00 2c4cc4d787a5f332f2c61f12cdb31e01da386439 [ Upstream commit 8cbbd11547f61b90b33a4ef70c4614eb2e789c49 ] The driver's CSA buffer is shared by all the ibs. When the high priority ib is submitted after the preempted ib, CP overrides the ib_completion_status as completed in the csa buffer. After that the preempted ib is resubmitted, CP would clear some locals stored for ib resume when reading the completed status, which causes gpu hang in some cases. Always set status as preempted for those resubmitted ib instead of reading everything from the CSA buffer. Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2535 Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2717 Signed-off-by: Jiadong Zhu <Jiadong.Zhu@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 8cbbd11547f61b90b33a4ef70c4614eb2e789c49 ]

The driver's CSA buffer is shared by all the ibs. When the high priority ib
is submitted after the preempted ib, CP overrides the ib_completion_status
as completed in the csa buffer. After that the preempted ib is resubmitted,
CP would clear some locals stored for ib resume when reading the completed
status, which causes gpu hang in some cases.

Always set status as preempted for those resubmitted ib instead of reading
everything from the CSA buffer.

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2535
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2717
Signed-off-by: Jiadong Zhu <Jiadong.Zhu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drm/amdgpu: fix amdgpu_cs_p1_user_fence 2023-09-23T09:14:38+00:00 Christian König christian.koenig@amd.com 2023-08-25T13:28:00+00:00 2575ef6c874c8ff28f71bafc436cf49a7e034f40 commit 35588314e963938dfdcdb792c9170108399377d6 upstream. The offset is just 32bits here so this can potentially overflow if somebody specifies a large value. Instead reduce the size to calculate the last possible offset. The error handling path incorrectly drops the reference to the user fence BO resulting in potential reference count underflow. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 35588314e963938dfdcdb792c9170108399377d6 upstream.

The offset is just 32bits here so this can potentially overflow if
somebody specifies a large value. Instead reduce the size to calculate
the last possible offset.

The error handling path incorrectly drops the reference to the user
fence BO resulting in potential reference count underflow.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Revert "drm/amd: Disable S/G for APUs when 64GB or more host memory" 2023-09-23T09:14:35+00:00 Hamza Mahfooz hamza.mahfooz@amd.com 2023-09-08T14:36:44+00:00 f8aad58031460eb5eedef62d7aa12b357a56b72e commit 169ed4ece8373f02f10642eae5240e3d1ef5c038 upstream. This reverts commit 70e64c4d522b732e31c6475a3be2349de337d321. Since, we now have an actual fix for this issue, we can get rid of this workaround as it can cause pin failures if enough VRAM isn't carved out by the BIOS. Cc: stable@vger.kernel.org # 6.1+ Acked-by: Harry Wentland <harry.wentland@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 169ed4ece8373f02f10642eae5240e3d1ef5c038 upstream.

This reverts commit 70e64c4d522b732e31c6475a3be2349de337d321.

Since, we now have an actual fix for this issue, we can get rid of this
workaround as it can cause pin failures if enough VRAM isn't carved out
by the BIOS.

Cc: stable@vger.kernel.org # 6.1+
Acked-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drm/amd: Make fence wait in suballocator uninterruptible 2023-09-23T09:14:35+00:00 Simon Pilkington simonp.git@gmail.com 2023-09-01T07:17:38+00:00 ecccfc5334bf6d9db6205af6f6a4e1974aafe213 commit e2884fe84a83c562346eb9d92783a3576ce67177 upstream. Commit c103a23f2f29 ("drm/amd: Convert amdgpu to use suballocation helper.") made the fence wait in amdgpu_sa_bo_new() interruptible but there is no code to handle an interrupt. This caused the kernel to randomly explode in high-VRAM-pressure situations so make it uninterruptible again. Signed-off-by: Simon Pilkington <simonp.git@gmail.com> Fixes: c103a23f2f29 ("drm/amd: Convert amdgpu to use suballocation helper.") Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Christian König <christian.koenig@amd.com> Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2761 CC: stable@vger.kernel.org # 6.4+ Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit e2884fe84a83c562346eb9d92783a3576ce67177 upstream.

Commit c103a23f2f29
("drm/amd: Convert amdgpu to use suballocation helper.")
made the fence wait in amdgpu_sa_bo_new() interruptible but there is no
code to handle an interrupt. This caused the kernel to randomly explode
in high-VRAM-pressure situations so make it uninterruptible again.

Signed-off-by: Simon Pilkington <simonp.git@gmail.com>
Fixes: c103a23f2f29 ("drm/amd: Convert amdgpu to use suballocation helper.")
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2761
CC: stable@vger.kernel.org # 6.4+
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>