commit 5ad7bbf3dba5c4a684338df1f285080f2588b535 upstream.
Currently amdgpu calls drm_sched_fini() from the fence driver sw fini
routine - such function is expected to be called only after the
respective init function - drm_sched_init() - was executed successfully.
Happens that we faced a driver probe failure in the Steam Deck
recently, and the function drm_sched_fini() was called even without
its counter-part had been previously called, causing the following oops:
amdgpu: probe of 0000:04:00.0 failed with error -110
BUG: kernel NULL pointer dereference, address: 0000000000000090
PGD 0 P4D 0
Oops: 0002 [#1] PREEMPT SMP NOPTI
CPU: 0 PID: 609 Comm: systemd-udevd Not tainted 6.2.0-rc3-gpiccoli #338
Hardware name: Valve Jupiter/Jupiter, BIOS F7A0113 11/04/2022
RIP: 0010:drm_sched_fini+0x84/0xa0 [gpu_sched]
[...]
Call Trace:
<TASK>
amdgpu_fence_driver_sw_fini+0xc8/0xd0 [amdgpu]
amdgpu_device_fini_sw+0x2b/0x3b0 [amdgpu]
amdgpu_driver_release_kms+0x16/0x30 [amdgpu]
devm_drm_dev_init_release+0x49/0x70
[...]
To prevent that, check if the drm_sched was properly initialized for a
given ring before calling its fini counter-part.
Notice ideally we'd use sched.ready for that; such field is set as the latest
thing on drm_sched_init(). But amdgpu seems to "override" the meaning of such
field - in the above oops for example, it was a GFX ring causing the crash, and
the sched.ready field was set to true in the ring init routine, regardless of
the state of the DRM scheduler. Hence, we ended-up using sched.ops as per
Christian's suggestion [0], and also removed the no_scheduler check [1].
[0] https://lore.kernel.org/amd-gfx/984ee981-2906-0eaf-ccec-9f80975cb136@amd.com/
[1] https://lore.kernel.org/amd-gfx/cd0e2994-f85f-d837-609f-7056d5fb7231@amd.com/
Fixes: 067f44c8b4 ("drm/amdgpu: avoid over-handle of fence driver fini in s3 test (v2)")
Suggested-by: Christian König <christian.koenig@amd.com>
Cc: Guchun Chen <guchun.chen@amd.com>
Cc: Luben Tuikov <luben.tuikov@amd.com>
Cc: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Guilherme G. Piccoli <gpiccoli@igalia.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 4b31b92b143f7d209f3d494c56d4c4673e9fc53d upstream.
change guarantees that gfxoff is allowed before moving further in
s2idle sequence to add more reliablity about gfxoff in amdgpu IP's
suspend flow
Signed-off-by: Harsh Jain <harsh.jain@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: "Limonciello, Mario" <Mario.Limonciello@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 3786a9bc0455ca58d953319f62daf96b6eb95490 upstream.
These have been at production level for a while. Drop
the flag.
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 1923bc5a56daeeabd7e9093bad2febcd6af2416a ]
Removing the firmware framebuffer from the driver means that even
if the driver doesn't support the IP blocks in a GPU it will no
longer be functional after the driver fails to initialize.
This change will ensure that unsupported IP blocks at least cause
the driver to work with the EFI framebuffer.
Cc: stable@vger.kernel.org
Suggested-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit d1acd68b2b8924c804e1e3cc1bc5fa4d6b76176c ]
Disable runtime power management on several sienna cichlid
cards, otherwise SMU will possibly fail to be resumed from
runtime suspend. Will drop this after a clean solution between
kernel driver and SMU FW is available.
amdgpu 0000:63:00.0: amdgpu: GECC is enabled
amdgpu 0000:63:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
amdgpu 0000:63:00.0: amdgpu: SMU is resuming...
amdgpu 0000:63:00.0: amdgpu: SMU: I'm not done with your command: SMN_C2PMSG_66:0x0000000E SMN_C2PMSG_82:0x00000080
amdgpu 0000:63:00.0: amdgpu: Failed to SetDriverDramAddr!
amdgpu 0000:63:00.0: amdgpu: Failed to setup smc hw!
[drm:amdgpu_device_ip_resume_phase2 [amdgpu]] *ERROR* resume of IP block <smu> failed -62
amdgpu 0000:63:00.0: amdgpu: amdgpu_device_ip_resume failed (-62)
v2: seperate to a function.
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Stable-dep-of: 1923bc5a56da ("drm/amd: Delay removal of the firmware framebuffer")
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 1d4624cd72b912b2680c08d0be48338a1629a858 upstream.
Some special polaris 10 chips overlap with the polaris11
DID range. Handle this properly in the driver.
v2: use local flags for other function calls.
Acked-by: Luben Tuikov <luben.tuikov@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>
[ Upstream commit f0d0f1087333714ee683cc134a95afe331d7ddd9 ]
With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
indirect call targets are validated against the expected function
pointer prototype to make sure the call target is valid to help mitigate
ROP attacks. If they are not identical, there is a failure at run time,
which manifests as either a kernel panic or thread getting killed. A
proposed warning in clang aims to catch these at compile time, which
reveals:
drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c:412:15: error: incompatible function pointer types initializing 'void (*)(struct amdgpu_device *, u32, u32, u32, u32)' (aka 'void (*)(struct amdgpu_device *, unsigned int, unsigned int, unsigned int, unsigned int)') with an expression of type 'void (struct amdgpu_device *, enum idh_request, u32, u32, u32)' (aka 'void (struct amdgpu_device *, enum idh_request, unsigned int, unsigned int, unsigned int)') [-Werror,-Wincompatible-function-pointer-types-strict]
.trans_msg = xgpu_ai_mailbox_trans_msg,
^~~~~~~~~~~~~~~~~~~~~~~~~
1 error generated.
drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c:435:15: error: incompatible function pointer types initializing 'void (*)(struct amdgpu_device *, u32, u32, u32, u32)' (aka 'void (*)(struct amdgpu_device *, unsigned int, unsigned int, unsigned int, unsigned int)') with an expression of type 'void (struct amdgpu_device *, enum idh_request, u32, u32, u32)' (aka 'void (struct amdgpu_device *, enum idh_request, unsigned int, unsigned int, unsigned int)') [-Werror,-Wincompatible-function-pointer-types-strict]
.trans_msg = xgpu_nv_mailbox_trans_msg,
^~~~~~~~~~~~~~~~~~~~~~~~~
1 error generated.
The type of the second parameter in the prototype should be 'enum
idh_request' instead of 'u32'. Update it to clear up the warnings.
Link: https://github.com/ClangBuiltLinux/linux/issues/1750
Reported-by: Sami Tolvanen <samitolvanen@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit ca54639c7752edf1304d92ff4d0c049d4efc9ba0 ]
As comment of pci_get_class() says, it returns a pci_device with its
refcount increased and decreased the refcount for the input parameter
@from if it is not NULL.
If we break the loop in amdgpu_atrm_get_bios() with 'pdev' not NULL, we
need to call pci_dev_put() to decrease the refcount. Add the missing
pci_dev_put() to avoid refcount leak.
Fixes: d38ceaf99e ("drm/amdgpu: add core driver (v4)")
Signed-off-by: Xiongfeng Wang <wangxiongfeng2@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit b85e285e3d6352b02947fc1b72303673dfacb0aa ]
As comment of pci_get_domain_bus_and_slot() says, it returns
a pci device with refcount increment, when finish using it,
the caller must decrement the reference count by calling
pci_dev_put().
So before returning from amdgpu_device_resume|suspend_display_audio(),
pci_dev_put() is called to avoid refcount leak.
Fixes: 3f12acc8d6 ("drm/amdgpu: put the audio codec into suspend state before gpu reset V3")
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 20543be93ca45968f344261c1a997177e51bd7e1 ]
drm_display_info is updated by drm_get_edid() or
drm_connector_update_edid_property(). In the amdgpu driver it is almost
always updated when the edid is read in amdgpu_connector_get_edid(),
but not always. Change amdgpu_connector_get_edid() and
amdgpu_connector_free_edid() to keep drm_display_info updated.
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Claudio Suarez <cssk@net-c.es>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Stable-dep-of: 602ad43c3cd8 ("drm/amdgpu: Partially revert "drm/amdgpu: update drm_display_info correctly when the edid is read"")
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit b9ab82da8804ec22c7e91ffd9d56c7a3abff0c8e upstream.
Allow user to know number of compute units (CU) that are in use at any
given moment. Enable access to the method kgd_gfx_v9_get_cu_occupancy
that computes CU occupancy.
Signed-off-by: Ramesh Errabolu <Ramesh.Errabolu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@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>
[ Upstream commit 65f8682b9aaae20c2cdee993e6fe52374ad513c9 ]
For asic with VF MMIO access protection avoid using CPU for VM table updates.
CPU pagetable updates have issues with HDP flush as VF MMIO access protection
blocks write to mmBIF_BX_DEV0_EPF0_VF0_HDP_MEM_COHERENCY_FLUSH_CNTL register
during sriov runtime.
v3: introduce virtualization capability flag AMDGPU_VF_MMIO_ACCESS_PROTECT
which indicates that VF MMIO write access is not allowed in sriov runtime
Signed-off-by: Danijel Slivka <danijel.slivka@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 90bfee142af0f0e9d3bec80e7acd5f49b230acf7 ]
If the number of pages from the userptr BO differs from the SG BO then the
allocated memory for the SG table doesn't get freed before returning
-EINVAL, which may lead to a memory leak in some error paths. Fix this by
checking the number of pages before allocating memory for the SG table.
Fixes: 264fb4d332 ("drm/amdgpu: Add multi-GPU DMA mapping helpers")
Signed-off-by: Rafael Mendonca <rafaelmendsr@gmail.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit d61e1d1d5225a9baeb995bcbdb904f66f70ed87e upstream.
In the S2idle suspend/resume phase the gfxoff is keeping functional so
some IP blocks will be likely to reinitialize at gfxoff entry and that
will result in failing to program GC registers.Therefore, let disallow
gfxoff until AMDGPU IPs reinitialized completely.
Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org # 5.15.x
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 50b0e4d4da09fa501e722af886f97e60a4f820d6 upstream.
Commit 8795e182b02d ("PCI/portdrv: Don't disable AER reporting in get_port_device_capability()")
uncovered a bug in amdgpu that required a reordering of the driver
init sequence to avoid accessing a special register on the GPU
before it was properly set up leading to an PCI AER error. This
reordering uncovered a different hw programming ordering dependency
in some APUs where the SDMA doorbells need to be programmed before
the GFX doorbells. To fix this, move the SDMA doorbell programming
back into the soc15 common code, but use the actual doorbell range
values directly rather than the values stored in the ring structure
since those will not be initialized at this point.
This is a partial revert, but with the doorbell assignment
fixed so the proper doorbell index is set before it's used.
Fixes: e3163bc8ffdfdb ("drm/amdgpu: move nbio sdma_doorbell_range() into sdma code for vega")
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: skhan@linuxfoundation.org
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 4bb71fce58f30df3f251118291d6b0187ce531e6 ]
This got lost somewhere along the way, This fixes
audio not working until set_property was called.
Signed-off-by: hongao <hongao@uniontech.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 3b7329cf5a767c1be38352d43066012e220ad43c upstream.
- Under SRIOV, we need to send REQ_GPU_FINI to the hypervisor
during the suspend time. Furthermore, we cannot request a
mode 1 reset under SRIOV as VF. Therefore, we will skip it
as it is called in suspend_noirq() function.
- In the resume code path, we need to send REQ_GPU_INIT to the
hypervisor and also resume PSP IP block under SRIOV.
Signed-off-by: Bokun Zhang <Bokun.Zhang@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>
[ Upstream commit abbc7a3dafb91b9d4ec56b70ec9a7520f8e13334 ]
Some asics still support non-atomic code paths.
Fixes: 66f99628eb2440 ("drm/amdgpu: use dirty framebuffer helper")
Reported-by: Arthur Marsh <arthur.marsh@internode.on.net>
Reviewed-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 66f99628eb24409cb8feb5061f78283c8b65f820 ]
Currently, we aren't handling DRM_IOCTL_MODE_DIRTYFB. So, use
drm_atomic_helper_dirtyfb() as the dirty callback in the amdgpu_fb_funcs
struct.
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@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>
commit 9a458402fb69bda886aa6cbe067311b6e3d9c52a upstream.
[Why]
This fixes 892deb48269c ("drm/amdgpu: Separate vf2pf work item init from virt data exchange").
we should read pf2vf data based at mman.fw_vram_usage_va after gmc
sw_init. commit 892deb48269c breaks this logic.
[How]
calling amdgpu_virt_exchange_data in amdgpu_virt_init_data_exchange to
set the right base in the right sequence.
v2:
call amdgpu_virt_init_data_exchange after gmc sw_init to make data
exchange workqueue run
v3:
clean up the code logic
v4:
add some comment and make the code more readable
Fixes: 892deb48269c ("drm/amdgpu: Separate vf2pf work item init from virt data exchange")
Signed-off-by: Jingwen Chen <Jingwen.Chen2@amd.com>
Reviewed-by: Horace Chen <horace.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit a8671493d2074950553da3cf07d1be43185ef6c6 ]
Move common IP init before GMC init so that HDP gets
remapped before GMC init which uses it.
This fixes the Unsupported Request error reported through
AER during driver load. The error happens as a write happens
to the remap offset before real remapping is done.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216373
The error was unnoticed before and got visible because of the commit
referenced below. This doesn't fix anything in the commit below, rather
fixes the issue in amdgpu exposed by the commit. The reference is only
to associate this commit with below one so that both go together.
Fixes: 8795e182b02d ("PCI/portdrv: Don't disable AER reporting in get_port_device_capability()")
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 892deb48269c65376f3eeb5b4c032ff2c2979bd7 ]
We want to be able to call virt data exchange conditionally
after gmc sw init to reserve bad pages as early as possible.
Since this is a conditional call, we will need
to call it again unconditionally later in the init sequence.
Refactor the data exchange function so it can be
called multiple times without re-initializing the work item.
v2: Cleaned up the code. Kept the original call to init_exchange_data()
inside early init to initialize the work item, afterwards call
exchange_data() when needed.
Signed-off-by: Victor Skvortsov <victor.skvortsov@amd.com>
Reviewed By: Shaoyun.liu <Shaoyun.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit e3163bc8ffdfdb405e10530b140135b2ee487f89 upstream.
This mirrors what we do for other asics and this way we are
sure the sdma doorbell range is properly initialized.
There is a comment about the way doorbells on gfx9 work that
requires that they are initialized for other IPs before GFX
is initialized. However, the statement says that it applies to
multimedia as well, but the VCN code currently initializes
doorbells after GFX and there are no known issues there. In my
testing at least I don't see any problems on SDMA.
This is a prerequisite for fixing the Unsupported Request error
reported through AER during driver load.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216373
The error was unnoticed before and got visible because of the commit
referenced below. This doesn't fix anything in the commit below, rather
fixes the issue in amdgpu exposed by the commit. The reference is only
to associate this commit with below one so that both go together.
Fixes: 8795e182b02d ("PCI/portdrv: Don't disable AER reporting in get_port_device_capability()")
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@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 dc1d85cb790f2091eea074cee24a704b2d6c4a06 upstream.
This mirrors what we do for other asics and this way we are
sure the ih doorbell range is properly initialized.
There is a comment about the way doorbells on gfx9 work that
requires that they are initialized for other IPs before GFX
is initialized. In this case IH is initialized before GFX,
so there should be no issue.
This is a prerequisite for fixing the Unsupported Request error
reported through AER during driver load.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216373
The error was unnoticed before and got visible because of the commit
referenced below. This doesn't fix anything in the commit below, rather
fixes the issue in amdgpu exposed by the commit. The reference is only
to associate this commit with below one so that both go together.
Fixes: 8795e182b02d ("PCI/portdrv: Don't disable AER reporting in get_port_device_capability()")
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@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 6c20490663553cd7e07d8de8af482012329ab9d6 upstream.
As per PCIE Base Spec r4.0 Section 6.18
'Software must not enable LTR in an Endpoint unless the Root Complex
and all intermediate Switches indicate support for LTR.'
This fixes the Unsupported Request error reported through AER during
ASPM enablement.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216455
The error was unnoticed before and got visible because of the commit
referenced below. This doesn't fix anything in the commit below, rather
fixes the issue in amdgpu exposed by the commit. The reference is only
to associate this commit with below one so that both go together.
Fixes: 8795e182b02d ("PCI/portdrv: Don't disable AER reporting in get_port_device_capability()")
Reported-by: Gustaw Smolarczyk <wielkiegie@gmail.com>
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: stable@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[ Upstream commit b8983d42524f10ac6bf35bbce6a7cc8e45f61e04 ]
The mmVM_L2_CNTL3 register is not assigned an initial value
Signed-off-by: Qu Huang <jinsdb@126.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 9d705d7741ae70764f3d6d87e67fad3b5c30ffd0 ]
V1:
The amdgpu_xgmi_remove_device function will send unload command
to psp through psp ring to terminate xgmi, but psp ring has been
destroyed in psp_hw_fini.
V2:
1. Change the commit title.
2. Restore amdgpu_xgmi_remove_device to its original calling location.
Move psp_xgmi_terminate call from amdgpu_xgmi_remove_device to
psp_hw_fini.
Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 373008bfc9cdb0f050258947fa5a095f0657e1bc ]
[Why]
During multi-vf executing benchmark (Luxmark) observed kiq error timeout.
It happenes because all of VFs do the tlb invalidation at the same time.
Although each VF has the invalidate register set, from hardware side
the invalidate requests are queue to execute.
[How]
In case of 12 VF increase timeout on 12*100ms
Signed-off-by: Dusica Milinkovic <Dusica.Milinkovic@amd.com>
Acked-by: Shaoyun Liu <shaoyun.liu@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>
The new vkms virtual display code is atomic so there is
no need to call drm_helper_disable_unused_functions()
when it is enabled. Doing so can result in a segfault.
When the driver switched from the old virtual display code
to the new atomic virtual display code, it was missed that
we enable virtual display unconditionally under SR-IOV
so the checks here missed that case. Add the missing
check for SR-IOV.
There is no equivalent of this patch for Linus' tree
because the relevant code no longer exists. This patch
is only relevant to kernels 5.15 and 5.16.
Fixes: 84ec374bd5 ("drm/amdgpu: create amdgpu_vkms (v4)")
Cc: stable@vger.kernel.org # 5.15.x
Cc: hgoffin@amazon.com
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f5ba14043621f4afdf3ad5f92ee2d8dbebbe4340 upstream.
When pinning a buffer, we should check to see if there are any
additional restrictions imposed by bo->preferred_domains. This will
prevent the BO from being moved to an invalid domain when pinning.
For example, this can happen if the user requests to create a BO in GTT
domain for display scanout. amdgpu_dm will allow pinning to either VRAM
or GTT domains, since DCN can scanout from either or. However, in
amdgpu_bo_pin_restricted(), pinning to VRAM is preferred if there is
adequate carveout. This can lead to pinning to VRAM despite the user
requesting GTT placement for the BO.
v2: Allow the kernel to override the domain, which can happen when
exporting a BO to a V4L camera (for example).
Signed-off-by: Leo Li <sunpeng.li@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@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>
[ Upstream commit aa482ddca85a3485be0e7b83a0789dc4d987670b ]
Active State Power Management (ASPM) feature is enabled since kernel 5.14.
There are some AMD Volcanic Islands (VI) GFX cards, such as the WX3200 and
RX640, that do not work with ASPM-enabled Intel Alder Lake based systems.
Using these GFX cards as video/display output, Intel Alder Lake based
systems will freeze after suspend/resume.
The issue was originally reported on one system (Dell Precision 3660 with
BIOS version 0.14.81), but was later confirmed to affect at least 4
pre-production Alder Lake based systems.
Add an extra check to disable ASPM on Intel Alder Lake based systems with
the problematic AMD Volcanic Islands GFX cards.
Fixes: 0064b0ce85 ("drm/amd/pm: enable ASPM by default")
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1885
Signed-off-by: Richard Gong <richard.gong@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 0ab5d711ec74d9e60673900974806b7688857947 ]
Evaluating `pcie_aspm_enabled` as part of driver probe has the implication
that if one PCIe bridge with an AMD GPU connected doesn't support ASPM
then none of them do. This is an invalid assumption as the PCIe core will
configure ASPM for individual PCIe bridges.
Create a new helper function that can be called by individual dGPUs to
react to the `amdgpu_aspm` module parameter without having negative results
for other dGPUs on the PCIe bus.
Suggested-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
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: Sasha Levin <sashal@kernel.org>
[ Upstream commit 425d7a87e54ee358f580eaf10cf28dc95f7121c1 ]
Some video card has more than one vcn instance, passing 0 to
vcn_v3_0_pause_dpg_mode is incorrect.
Error msg:
Register(1) [mmUVD_POWER_STATUS] failed to reach value
0x00000001 != 0x00000002
Reviewed-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: tiancyin <tianci.yin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 5cb0e3fb2c54eabfb3f932a1574bff1774946bc0 upstream.
amdgpu: [mmhub0] no-retry page fault (src_id:0 ring:40 vmid:8 pasid:32769, for process test_basic pid 3305 thread test_basic pid 3305)
amdgpu: in page starting at address 0x00007ff990003000 from IH client 0x12 (VMC)
amdgpu: VM_L2_PROTECTION_FAULT_STATUS:0x00840051
amdgpu: Faulty UTCL2 client ID: MP1 (0x0)
amdgpu: MORE_FAULTS: 0x1
amdgpu: WALKER_ERROR: 0x0
amdgpu: PERMISSION_FAULTS: 0x5
amdgpu: MAPPING_ERROR: 0x0
amdgpu: RW: 0x1
When memory is allocated by kfd, no one triggers the tlb flush for MMHUB0.
There is page fault from MMHUB0.
v2:fix indentation
v3:change subject and fix indentation
Signed-off-by: Ruili Ji <ruiliji2@amd.com>
Reviewed-by: Philip Yang <philip.yang@amd.com>
Reviewed-by: Aaron Liu <aaron.liu@amd.com>
Acked-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>
[ Upstream commit 4fac4fcf4500bce515b0f32195e7bb86aa0246c6 ]
The kfd_bo_list is used to restore process BOs after
evictions. As page tables could be destroyed during
evictions, we should also update pinned BOs' page tables
during restoring to make sure they are valid.
So for pinned BOs,
1, Validate them and update their page tables.
2, Don't add eviction fence for them.
v2:
- Don't handle pinned ones specially in BO validation.(Felix)
Signed-off-by: Lang Yu <Lang.Yu@amd.com>
Acked-by: Christian König <christian.koenig@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>
