If use gart for FB translation, we will squeeze vram into
sysvm aperture. This requires 2 level gart table. Add
page table depth and page table block size parameters
to gmc. This is prepare work to 2-level gart table
construction
Signed-off-by: NOak Zeng <Oak.Zeng@amd.com>
Reviewed-by: NChristian Konig <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <felix.kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

If use GART for FB translation, place both vram and gart to sysvm
aperture. AGP aperture is not set up in this case because it
is not used
Signed-off-by: NOak Zeng <Oak.Zeng@amd.com>
Reviewed-by: NChristian Konig <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <felix.kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Modify the comment to reflect the fact that, if
use GART for vram address translation for vmid0,
[vram_start, vram_end] will be placed inside SYSVM
aperture, together with GART.
Signed-off-by: NOak Zeng <Oak.Zeng@amd.com>
Reviewed-by: NChristian Konig <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <felix.kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

In amdgpu_gmc_gart_location function, gart_size is adjusted
by a smu_prv_buffer_size. This logic shouldn't belong to
this function. Move the logic to the mc_init functions
Signed-off-by: NOak Zeng <Oak.Zeng@amd.com>
Reviewed-by: NChristian Konig <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <felix.kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

On A+A platform, CPU write page directory and page table in cached
mode. So it is necessary for page table walker to snoop CPU cache.
This setting is necessary for page walker to snoop page directory
and page table data out of CPU cache.
Signed-off-by: NOak Zeng <Oak.Zeng@amd.com>
Acked-by: NChristian Konig <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <felix.kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

On A+A platform, vram can be mapped as WB. Not necessarily
to always map vram as WC on such platform.

Calling function arch_io_reserve_memtype_wc will mark the
whole vram region as WC. So don't call it for A+A platform.
Signed-off-by: NOak Zeng <Oak.Zeng@amd.com>
Suggested-by: NAlex Deucher <alexander.deucher@amd.com>
Acked-by: NChristian Konig <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <felix.kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The psp supplies the link type in the upper 2 bits of the psp xgmi node
information num_hops field.  With a new link type, Aldebaran has these
bits set to a non-zero value (1 = xGMI3) so the KFD topology will report
the incorrect IO link weights without proper masking.
The actual number of hops is located in the 3 least significant bits of
this field so mask if off accordingly before passing it to the KFD.
Signed-off-by: NJonathan Kim <jonathan.kim@amd.com>
Reviewed-by: NAmber Lin <amber.lin@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

By default this timestamp is 32 bit counter. It gets
overflowed in around 10 minutes.
Signed-off-by: NAlex Sierra <alex.sierra@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

If xnack is on, VM retry fault interrupt send to IH ring1, and ring1
will be full quickly. IH cannot receive other interrupts, this causes
deadlock if migrating buffer using sdma and waiting for sdma done while
handling retry fault.

Remove VMC from IH storm client, enable ring1 write pointer overflow,
then IH will drop retry fault interrupts and be able to receive other
interrupts while driver is handling retry fault.

IH ring1 write pointer doesn't writeback to memory by IH, and ring1
write pointer recorded by self-irq is not updated, so always read
the latest ring1 write pointer from register.
Signed-off-by: NPhilip Yang <Philip.Yang@amd.com>
Signed-off-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

With the current kfd memory accounting scheme, kfd applications
can use up to 15/16 of total system memory. For system which
has small total system memory size it leaves small system memory
for OS. For example, if the system has totally 16GB of system
memory, this scheme leave OS and non-kfd applications only 1GB
of system memory. In many cases, this leads to OOM killer.

This patch changed the KFD system memory accounting scheme.
15/16 of free system memory when kfd driver load. This deduct
the system memory that OS already use.
Signed-off-by: NOak Zeng <Oak.Zeng@amd.com>
Suggested-by: NPhilip Yang <Philip.Yang@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Arcturus and onwards products should follow the same sequence
that have pmfw loading ahead of tmr setup
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: NKevin Wang <kevin1.wang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Aldebaran MMHUB CG/LS logic is controlled by VBIOS. Enable the state
change logic only if driver is used for control.
Signed-off-by: NLijo Lazar <lijo.lazar@amd.com>
Reviewed-by: NFeifei Xu <Feifei.Xu@amd.com>
Reviewed-by: NKevin Wang <kevin1.wang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

v1: The interrupts need to be enabled to move to DS clocks.
v2: Don't enable GFX IDLE interrupts if there are no GFX rings.
Signed-off-by: NLijo Lazar <lijo.lazar@amd.com>
Reviewed-by: NHawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Aldebaran clock gating support for GFX,SDMA,IH blocks
VCN/JPEG blocks are excluded in this patch, to be enabled later
Signed-off-by: NLijo Lazar <lijo.lazar@amd.com>
Acked-by: NFeifei Xu <Feifei.Xu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Add the mmhub client id table for aldebaran.
Reviewed-by: NHawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Enable dpg indirect sram mode on aldebaran.
Signed-off-by: NJames Zhu <James.Zhu@amd.com>
Reviewed-by: NLeo Liu <leo.liu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Enable vcn dpg mode on aldebaran
Signed-off-by: NJames Zhu <James.Zhu@amd.com>
Reviewed-by: NLeo Liu <leo.liu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Enable vcn and jpeg 2.6 on aldebaran.
Signed-off-by: NJames Zhu <James.Zhu@amd.com>
Reviewed-by: NLeo Liu <leo.liu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Enable smu13 block on aldebaran
Signed-off-by: NLijo Lazar <lijo.lazar@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

global noretry setting now is cached to gmc.noretry
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: NKevin Wang <kevin1.wang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

hdp read cache is removed in aldebaran. don't issue
an mmio write or write data packet to hardware.

v2: rebase
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: NFeifei Xu <Feifei.Xu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Simplify all Aldebaran DIDs into one ASIC type.
Signed-off-by: NAmber Lin <Amber.Lin@amd.com>
Reviewed-by: NKevin Wang <kevin1.wang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
This causes infinite retries on the UTCL1 RB, preventing
higher priority RB such as paging RB.

[How]
Set to one the SDMAx_UTLC1_TIMEOUT registers for all SDMAs.
Signed-off-by: NAlex Sierra <alex.sierra@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NHawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

0x61 is assigned to HBM2E in atom_dgpu_vram_type.
Signed-off-by: NFeifei Xu <Feifei.Xu@amd.com>
Reviewed-by: NHawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

driver should use the gfx_info atomfirmware interface
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: NFeifei Xu <Feifei.Xu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

For ASICs that don't support ip discovery feature, query
gfx configuration through atomfirmware interface, rather
than gpu_info firmware.
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: NFeifei Xu <Feifei.Xu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Temporarily add smu_pptable module parameter for aldebaran.This is used
to force soft PPTable use overriding any VBIOS PPTable.
Signed-off-by: NLijo Lazar <Lijo.Lazar@amd.com>
Reviewed-by: NKenneth Feng <kenneth.feng@amd.com>
Reviewed-by: NKevin Wang <kevin1.wang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Disable PCIe BAR resizing on A+A config. It's not needed because we won't use the
PCIe BAR, but it breaks the PCI BAR configuration with the current SBIOS.

Error message of FB BAR resize failure under A+A:

[  154.913731] [drm:amdgpu_device_resize_fb_bar [amdgpu]] *ERROR* Problem resizing BAR0 (-22).
Signed-off-by: NOak Zeng <Oak.Zeng@amd.com>
Reviewed-by: NAmber Lin <Amber.Lin@amd.com>
Reviewed-by: NFelix Kuehling <Felix.kuehling@amd.com>
Reviewed-by: NChristian Koenig <Christian.Koenig@amd.com>
Tested-by: NAmber Lin <Amber.Lin@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

For A+A configuration, device memory is supposed to be mapped as
cachable from CPU side. For kernel pre-map gpu device memory using
ioremap_cache
Signed-off-by: NOak Zeng <Oak.Zeng@amd.com>
Reviewed-by: NChristian Koenig <Christian.Koenig@amd.com>
Tested-by: NAmber Lin <Amber.Lin@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

shall revisit the change later
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

replace vega10 ih block with vega20 ih block for
aldebaran.
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Acked-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NDennis Li <Dennis.Li@amd.com>
Reviewed-by: NFeifei Xu <Feifei.Xu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

For aldebaran, psp firmware won't program IH_CHICKEN.
it now depends on driver to program it properly so
either bus address or gpu virtual address is just
working for ih ring.
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Acked-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NDennis Li <Dennis.Li@amd.com>
Reviewed-by: NFeifei Xu <Feifei.Xu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The callback will be invoked to query mmea error
status when needed.
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li<Dennis.Li@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The callback will be invoked to reset mmhub ras error
counters when needed.
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li<Dennis.Li@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The callback will be invoked to harvest all kinds
of mmhub ras error
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li<Dennis.Li@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The callback will be invoked to reset sdma ras error
counters when needed.
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li<Dennis.Li@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The callback will be invoked to harvest all kinds
of sdma ras error
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li<Dennis.Li@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

sdma ras function is the main structure to support
sdma ras on aldebaran. the patch initializes late_init
late_fini callbacks.
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li<Dennis.Li@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

perform one-time initialization for sdma registers
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: NKevin Wang <kevin1.wang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Similar as xgmi connected gpu nodes, physical_node_id
* segment_size should be used to calculate the offset
of aper_base.

The asic type check is redundant. once physical_node_id
and segment_size are initialized, it should be count
on.
Signed-off-by: NHawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: NKevin Wang <kevin1.wang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>