Add helper to call the update function for both BO and shadow.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Add a function to get the parent of a PD/PT.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Slowly leaking memory one page at a time :)
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAndrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Retrieve hive_id from amdgpu device

v2: compile fix
Signed-off-by: NShaoyun Liu <Shaoyun.Liu@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Driver will save an array of XGMI hive info, each hive will have a list of devices
that have the same hive ID.
Signed-off-by: NShaoyun Liu <Shaoyun.Liu@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Add dummy function for xgmi function interface with psp
Signed-off-by: NShaoyun Liu <Shaoyun.Liu@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NHuang Rui <ray.huang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Place holder for XGMI support
Signed-off-by: NShaoyun Liu <Shaoyun.Liu@amd.com>
Reviewed-by: NHuang Rui <ray.huang@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

On hives with xgmi enabled, the fb_location aperture is a size
which defines the total framebuffer size of all nodes in the
hive.  Each GPU in the hive has the same view via the fb_location
aperture.  GPU0 starts at offset (0 * segment size),
GPU1 starts at offset (1 * segment size), etc.

For access to local vram on each GPU, we need to take this offset into
account. This including on setting up GPUVM page table and GART table

v2: squash in "drm/amdgpu: Init correct fb region for none XGMI configuration"
Acked-by: NHuang Rui <ray.huang@amd.com>
Acked-by: NSlava Abramov <slava.abramov@amd.com>
Signed-off-by: NShaoyun Liu <Shaoyun.Liu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Acked-by: NHuang Rui <ray.huang@amd.com>

Used to populate the xgmi info on vega20.

v2: PF_MAX_REGION is val - 1 (Ray)
Acked-by: NHuang Rui <ray.huang@amd.com>
Acked-by: NSlava Abramov <slava.abramov@amd.com>
Reviewed-by :Shaoyun liu <Shaoyun.liu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by :Shaoyun liu <Shaoyun.liu@amd.com>

Initial pass at a structure to store xgmi info.  xgmi is a high
speed cross gpu interconnect.
Acked-by: NHuang Rui <ray.huang@amd.com>
Acked-by: NSlava Abramov <slava.abramov@amd.com>
Reviewed-by :Shaoyun liu <Shaoyun.liu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NShaoyun Liu <Shaoyun.Liu@amd.com>

since we use PSP to program IH regs now
Signed-off-by: NMonk Liu <Monk.Liu@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NHuang Rui <ray.huang@amd.com>
Signed-off-by: NEmily Deng <Emily.Deng@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Otherwise we might run into a use after free during bulk move.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NHuang Rui <ray.huang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Fix SDMA hang in prt mode, clear XNACK_WATERMARK in reg SDMA0_UTCL1_WATERMK to avoid the issue

Affected ASICs: VEGA10 VEGA12 RV1 RV2

v2: add reg clear for SDMA1
Signed-off-by: NTao Zhou <tao.zhou1@amd.com>
Tested-by: NYukun Li <yukun1.li@amd.com>
Reviewed-by: NHawking Zhang <Hawking.Zhang@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This patch fixes following warnings.

./drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c:3011:
warning: Excess function parameter 'dev' description
in 'amdgpu_vm_get_task_info'

./drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c:3012:
warning: Function parameter or member 'adev' not
described in 'amdgpu_vm_get_task_info'

./drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c:3012:
warning: Excess function parameter 'dev' description
in 'amdgpu_vm_get_task_info'
Signed-off-by: NMasanari Iida <standby24x7@gmail.com>
Signed-off-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The intent of two commits was lost in the last rebase:

810955ba drm/amdgpu: Fix acquiring VM on large-BAR systems
b5d21aac drm/amdgpu: Don't use shadow BO for compute context

This commit restores the original behaviour:
* Don't set AMDGPU_GEM_CREATE_NO_CPU_ACCESS for page directories
  to allow them to be reused for compute VMs
* Don't create shadow BOs for page tables in compute VMs

v2: move more logic into amdgpu_vm_bo_param
Signed-off-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Tested-by: NKent Russell <Kent.Russell@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Enable the old AGP aperture to avoid GART mappings.

v2: don't enable it for SRIOV
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Only initialize KFD once by moving amdgpu_amdkfd_init from
amdgpu_pci_probe to amdgpu_init. This fixes kernel oopses and hangs
when booting multi-GPU systems.

Also removed some vestiges of KFD being its own module.
Signed-off-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Acked-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Otherwise we won't be able to use the AGP aperture.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Start to use the old AGP aperture for system memory access.

v2: Move that to amdgpu_ttm_alloc_gart
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Helper to figure out the location of the AGP BAR.

v2: fix a couple of bugs
v3: correctly add one to vram_end
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Correct sign extend the GMC addresses to 48bit.

v2: sign extending turned out easier than thought.
v3: clean up the defines and move them into amdgpu_gmc.h as well
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Since we have a lot of FAQ on the VM state machine try to improve the
documentation by adding functions for each state move.

v2: fix typo in amdgpu_vm_bo_invalidated, use amdgpu_vm_bo_relocated in
    one more place as well.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Avoid unlocking a lock we never locked.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Allows us to avoid taking the spinlock in more places.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

amdgpu_vm_bo_* functions should come much later.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Added vega20 overdrive support based on existing OD sysfs
APIs. However, the OD logics are simplified on vega20. So,
the behavior will be a little different and works only on
some limited levels.

V2: fix typo
    fix commit description
    revise error logs
    add support for clock OD

V3: separate clock from voltage OD settings
Signed-off-by: NEvan Quan <evan.quan@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Vega20 does not appear to be affected by the same issue
as vega10.  Enable the full stolen memory handling on
vega20.  Reserve the appropriate size at init time to avoid
display artifacts and then free it at the end of init once
the new FB is up and running.
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

vega12 does not appear to be affected by the same issue
as vega10.  Enable the full stolen memory handling on
vega12.  Reserve the appropriate size at init time to avoid
display artifacts and then free it at the end of init once
the new FB is up and running.
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Raven does not appear to be affected by the same issue
as vega10.  Enable the full stolen memory handling on
Raven.  Reserve the appropriate size at init time to avoid
display artifacts and then free it at the end of init once
the new FB is up and running.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=106639Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

No functional change, just rework it in order to adjust the
behavior on a per asic level.  The problem is that on vega10,
something corrupts the lower 8 MB of vram on the second
resume from S3.  This does not seem to affect Raven, other
gmc9 based asics need testing.
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Add BOs to the idle state again and correctly clear the flag when
new BOs are added.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Tested-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

First step to fix the LRU corruption, we accidentially tried to move things
on the LRU after dropping the lock.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Tested-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This at least allows to fail any subsequent IOCTLs with -ENODEV
after the device is gone.
Still this operation is not supported yet in graphic mode
and will lead at least to page faults and other issues.
Signed-off-by: NAndrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This reverts commit a7f91061.

Felix pointed out that we need to have the BOs mapped even before
amdgpu_vm_update_directories is called.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Acked-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Always try to put the GART away from where VRAM is.

v2: correctly handle the 4GB limitation
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This caused a confusing error message, but there is functionally
no problem since the default method is DIRECT.
Reviewed-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

Move some KFD-related (but used in amdgpu_drv.c) definitions from
kfd_priv.h to kgd_kfd_interface.h so we don't need to include kfd_priv.h
in amdgpu_drv.c. This fixes a build failure when AMDGPU is enabled but
MMU_NOTIFIER is not.
This patch also disables KFD-related module options when HSA_AMD is not
enabled.

v2: rebase (Alex)
Signed-off-by: NAmber Lin <Amber.Lin@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

kvmalloc_array uses __GFP_ZERO flag ensures that the returned address
is zeroed already, memset it to zero again afterwards is unnecessary,
and in this case buggy because we only clear the first entry.
Signed-off-by: NPhilip Yang <Philip.Yang@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Instead of the larger one use the smaller hole in the MC address
space for the GART mappings.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This reverts commit 31625ccae4464b61ec8cdb9740df848bbc857a5b.

It triggered various badness on my development machine when running the
piglit gpu profile with radeonsi on Bonaire, looks like memory
corruption due to insufficiently protected list manipulations.
Signed-off-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>