David Francis [Mon, 27 Feb 2023 15:33:11 +0000 (10:33 -0500)]
drm/amdgpu/bu: add mtype_local as a module parameter
Selects the MTYPE to be used for local memory,
(0 = MTYPE_CC (default), 1 = MTYPE_NC, 2 = MTYPE_RW)
v2: squash in build fix (Alex)
Reviewed-by: Graham Sider <Graham.Sider@amd.com> Signed-off-by: David Francis <David.Francis@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Felix Kuehling [Tue, 21 Feb 2023 22:44:18 +0000 (17:44 -0500)]
drm/amdgpu: Override MTYPE per page on GFXv9.4.3 APUs
On GFXv9.4.3 NUMA APUs, system memory locality must be determined per
page to choose the correct MTYPE. This patch adds a GMC callback that
can provide this per-page override and implements it for native mode.
Carve-out mode is not yet supported and will use the safe default
(remote) MTYPE for system memory.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Reviewed-and-tested-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Felix Kuehling [Tue, 21 Feb 2023 22:31:32 +0000 (17:31 -0500)]
drm/amdgpu: Fix per-BO MTYPE selection for GFXv9.4.3
Treat system memory on NUMA systems as remote by default. Overriding with
a more efficient MTYPE per page will be implemented in the next patch.
No need for a special case for APP APUs. System memory is handled the same
for carve-out and native mode. And VRAM doesn't exist in native mode.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Reviewed-and-tested-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Graham Sider [Mon, 6 Feb 2023 19:04:42 +0000 (14:04 -0500)]
drm/amdgpu/bu: Add use_mtype_cc_wa module param
By default, set use_mtype_cc_wa to 1 to set PTE coherence flag MTYPE_CC
instead of MTYPE_RW by default. This is required for the time being to
mitigate a bug causing XCCs to hit stale data due to TCC marking fully
dirty lines as exclusive.
Signed-off-by: Graham Sider <Graham.Sider@amd.com> Reviewed-by: Joseph Greathouse <Joseph.Greathouse@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Graham Sider [Wed, 8 Feb 2023 16:10:57 +0000 (11:10 -0500)]
drm/amdgpu: Use legacy TLB flush for gfx943
Invalidate TLBs via a legacy flush request (flush_type=0) prior to the
heavyweight flush requests (flush_type=2) in gmc_v9_0.c. This is
temporarily required to mitigate a bug causing CPC UTCL1 to return stale
translations after invalidation requests in address range mode.
v2: squash in long term fix "drm/amdgpu: disable extra gfx943 legacy flush on rev1+"
Signed-off-by: Graham Sider <Graham.Sider@amd.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Philip Yang [Fri, 31 Mar 2023 15:13:40 +0000 (11:13 -0400)]
drm/amdkfd: Move pgmap to amdgpu_kfd_dev structure
VRAM pgmap resource is allocated every time when switching compute
partitions because kfd_dev is re-initialized by post_partition_switch,
As a result, it causes memory region resource leaking and system
memory usage accounting unbalanced.
pgmap resource should be allocated and registered only once when loading
driver and freed when unloading driver, move it from kfd_dev to
amdgpu_kfd_dev.
Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Mukul Joshi [Wed, 15 Mar 2023 18:04:33 +0000 (14:04 -0400)]
drm/amdkfd: Increase queue number per process to 255 on GFX9.4.3
Increase the maximum number of queues that can be created per process
to 255 on GFX 9.4.3. There is no HWS limitation restricting the number
queues that can be created.
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>
Hawking Zhang [Mon, 20 Mar 2023 09:51:30 +0000 (17:51 +0800)]
drm/amdgpu: Adjust the sequence to query ras error info
It turns out STATUS_VALID_FLAG needs to be checked
ahead of any other fields. ADDRESS_VALID_FLAG and
ERR_INFO_VALID_FLAG only manages ADDRESS and ERR_INFO
field respectively. driver should continue poll
ERR CNT field even ERR_INFO_VALD_FLAG is not set.
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Gavin Wan [Fri, 17 Mar 2023 22:42:30 +0000 (18:42 -0400)]
drm/amdgpu: Checked if the pointer NULL before use it.
For SRIOV on some parts, the host driver does not post VBIOS. So the guest
cannot get bios information. Therefore, adev->virt.fw_reserve.p_pf2vf
and adev->mode_info.atom_context are NULL.
Signed-off-by: Gavin Wan <Gavin.Wan@amd.com> Reviewed-by: Zhigang Luo <Zhigang.Luo@amd.com> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Gavin Wan [Mon, 10 Apr 2023 19:04:26 +0000 (15:04 -0400)]
drm/amdgpu: Set memory partitions to 1 for SRIOV.
For SRIOV, the memory partitions are set on host drover. Each VF only
has one memory partition. We need set the memory partitions to 1 on
guest driver for SRIOV.
V2: sqaush in fix ("drm/amdgpu: Fix memory range info of GC 9.4.3 VFs")
Signed-off-by: Gavin Wan <Gavin.Wan@amd.com> Acked-by: Zhigang Luo <zhigang.luo@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Gavin Wan [Mon, 3 Apr 2023 21:49:41 +0000 (17:49 -0400)]
drm/amdgpu: Skip using MC FB Offset when APU flag is set for SRIOV.
The MC_VM_FB_OFFSET is PF only register. It cannot be read on VF.
So, the driver should not use MC_VM_FB_OFFSET address to set the
address of dev->gmc.aper_base.
Signed-off-by: Gavin Wan <Gavin.Wan@amd.com> Reviewed-by: Zhigang Luo <zhigang.luo@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rajneesh Bhardwaj [Tue, 28 Feb 2023 01:08:29 +0000 (20:08 -0500)]
drm/amdkfd: Native mode memory partition support
For native mode, after amdgpu_bo is created on CPU domain, then call
amdgpu_ttm_tt_set_mem_pool to select the TTM pool using bo->mem_id.
ttm_bo_validate will allocate the memory to the correct memory partition
before mapping to GPUs.
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-and-tested-by: Mukul Joshi <mukul.joshi@amd.com> Signed-off-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Philip Yang [Mon, 27 Feb 2023 16:16:09 +0000 (11:16 -0500)]
drm/amdgpu: Set TTM pools for memory partitions
For native mode only, create TTM pool for each memory partition to store
the NUMA node id, then the TTM pool will be selected using memory
partition id to allocate memory from the correct partition.
Acked-by: Christian König <christian.koenig@amd.com>
(rajneesh: changed need_swiotlb and need_dma32 to false for pool init) Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-and-tested-by: Mukul Joshi <mukul.joshi@amd.com> Signed-off-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Mon, 13 Feb 2023 13:56:18 +0000 (19:26 +0530)]
drm/amdgpu: Add auto mode for compute partition
When auto mode is specified, driver will choose the right compute
partition mode.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Reviewed-by: Philip Yang <philip.yang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Mon, 13 Feb 2023 13:20:07 +0000 (18:50 +0530)]
drm/amdgpu: Check memory ranges for valid xcp mode
Check the memory ranges available to the device also for deciding a
valid partition mode. Only select combinations are valid for a
particular mode.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Reviewed-by: Philip Yang <philip.yang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Fri, 3 Feb 2023 11:42:10 +0000 (17:12 +0530)]
drm/amdgpu: Add callback to fill xcp memory id
Add callback in xcp interface to fill xcp memory id information. Memory
id is used to identify the range/partition of an XCP from the available
memory partitions in device. Also, fill the id information.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Tue, 14 Feb 2023 09:15:45 +0000 (14:45 +0530)]
drm/amdgpu: Initialize memory ranges for GC 9.4.3
GC 9.4.3 ASICS may have memory split into multiple partitions.Initialize
the memory partition information for each range. The information may be
in the form of a numa node id or a range of pages.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Tue, 14 Feb 2023 13:03:51 +0000 (18:33 +0530)]
drm/amdgpu: Store additional numa node information
Use a struct to store additional numa node information including size
and base address. Add numa_info pointer to xcc object to point to the
relevant structure based on its proximity domain.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Fri, 24 Feb 2023 12:31:38 +0000 (18:01 +0530)]
drm/amdgpu: Use apt name for FW reserved region
Use the generic term fw_reserved_memory for FW reserve region. This
region may also hold discovery TMR in addition to other reserve
regions. This region size could be larger than discovery tmr size, hence
don't change the discovery tmr size based on this.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Thu, 23 Feb 2023 14:43:56 +0000 (20:13 +0530)]
drm/amdgpu: Use GPU VA space for IH v4.4.2 in APU
For IH ring buffer and read/write pointers, use GPU VA space rather than
Guest PA on APU configs. Access through Guest PA doesn't work when IOMMU
is enabled. It is also beneficial in NUMA configs as it allocates from
the closest numa pool in a numa enabled system.
Lijo Lazar [Mon, 20 Feb 2023 06:34:30 +0000 (12:04 +0530)]
drm/amdgpu: Simplify aquavanjram instance mapping
Simplify so as to use the same sequence to assign logical to physical
ids for all IPs.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Tested-by: James Zhu <James.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Wed, 8 Mar 2023 07:11:22 +0000 (12:41 +0530)]
drm/amdgpu/vcn: Use buffer object's deletion logic
VCN DPG buffer object is intialized to NULL. If allotted, buffer object
deletion logic will take care of NULL check and delete accordingly. This
is useful for cases where indirect sram flag could be manipulated later
after buffer allocation.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Sonny Jiang [Mon, 20 Mar 2023 20:17:21 +0000 (16:17 -0400)]
drm/amdgpu: A workaround for JPEG_v4_0_3 ring test fail
The jpeg_v4_0_3 jpeg_pitch register uses UVD_JRBC_SCRATCH0. It needs to
move WREG() to after jpeg_start.
Switch to a posted register write when doing the ring test to make sure
the register write lands before we test the result.
Signed-off-by: Sonny Jiang <sonny.jiang@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Fri, 17 Feb 2023 14:11:06 +0000 (19:41 +0530)]
drm/amdgpu: Use logical ids for VCN/JPEG v4.0.3
Address VCN/JPEG instances using logical ids. Whenever register access is
required, get the physical instance using GET_INST.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Tested-by: James Zhu <James.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Fri, 17 Feb 2023 13:14:20 +0000 (18:44 +0530)]
drm/amdgpu: Add VCN logical to physical id mapping
Add mappings for logical to physical id for VCN/JPEG 4.0.3
v2: make local function static (Alex)
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Tested-by: James Zhu <James.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Fri, 17 Feb 2023 12:57:51 +0000 (18:27 +0530)]
drm/amdgpu: Add instance mask for VCN and JPEG
Keep an instance mask formed by physical instance numbers for VCN and JPEG
IPs. Populate the mask from discovery table information.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Tested-by: James Zhu <James.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Shiwu Zhang [Mon, 13 Feb 2023 07:45:42 +0000 (15:45 +0800)]
drm/amdgpu: preserve the num_links in case of reflection
For topology reflection, each socket to every other socket has the
exactly same topology info as the other way around. So it is safe
to keep the reflected num_links value otherwise it will be overriden
by the link info output of GET_PEER_LINKS command.
Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Philip Yang [Thu, 9 Feb 2023 23:23:16 +0000 (18:23 -0500)]
drm/amdkfd: Flush TLB after unmapping for GFX v9.4.3
kfd_flush_tlb_after_unmap should return true for GFX v9.4.3, to do TLB
heavyweight flush after unmapping from GPU to guarantee that the GPU
will not access pages after they have been unmapped. This also helps
improve the mapping to GPU performance.
Without this, KFD accidently flush TLB after mapping to GPU because the
vm update sequence number is increased by previous unmapping.
Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Mon, 30 Jan 2023 04:18:39 +0000 (09:48 +0530)]
drm/amdgpu: Add fallback path for discovery info
If SOC doesn't expose dedicated vram, discovery region may be
available through system memory. Rename the existing interface to
generic read_binary_from_mem and add a fallback path to read from system
memory.
Lijo Lazar [Mon, 30 Jan 2023 04:08:09 +0000 (09:38 +0530)]
drm/amdgpu: Read discovery info from system memory
On certain ASICs, discovery info is available at reserved region in system
memory. The location is available through ACPI interface. Add API to read
discovery info from there.
Mukul Joshi [Tue, 31 Jan 2023 16:23:50 +0000 (11:23 -0500)]
drm/amdkfd: Enable SVM on Native mode
This patch enables SVM capability on GFX9.4.3 when
run in Native mode. It also sets best_prefetch and
best_restore locations to CPU as there is no VRAM.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Acked-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Mon, 16 Jan 2023 05:25:38 +0000 (10:55 +0530)]
drm/amdgpu: Add flags for partition mode query
It's not required to take lock on all cases while querying partition
mode. Querying partition mode during KFD init process doesn't need to
take a lock. Init process after a switch will already be happening under
lock. Control the behaviour by adding flags to xcp_query_partition_mode.
Lijo Lazar [Thu, 9 Mar 2023 07:34:56 +0000 (13:04 +0530)]
drm/amdgpu: Add mode-2 reset in SMU v13.0.6
Modifications to mode-2 reset flow for SMU v13.0.6 ASICs.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Reviewed-by: Asad Kamal <asad.kamal@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Fri, 10 Mar 2023 06:03:37 +0000 (11:33 +0530)]
drm/amd/pm: Notify PMFW about driver unload cases
On SMU v13.0.6 APUs, FW will need to take some actions if driver is going
to halt RLC. Notify PMFW that driver is not going to manage device so
that FW takes care of the required actions.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Reviewed-by: Asad Kamal <asad.kamal@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Fri, 10 Mar 2023 12:41:25 +0000 (18:11 +0530)]
drm/amd/pm: Update PMFW headers for version 85.54
It adds message support for FW notification on driver unload.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Reviewed-by: Asad Kamal <asad.kamal@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Le Ma [Thu, 16 Mar 2023 09:42:49 +0000 (17:42 +0800)]
drm/amdgpu: change the print level to warn for ip block disabled
Avoid to mislead users as it's not a real error.
Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Amber Lin <Amber.Lin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rajneesh Bhardwaj [Sat, 28 Jan 2023 02:48:06 +0000 (21:48 -0500)]
drm/amdgpu: Create VRAM BOs on GTT for GFXIP9.4.3
On GFXIP9.4.3 APP APU where there is no dedicated VRAM domain handle
VRAM BO allocation requests on CPU domain and validate them on GTT.
Support for handling multi-socket and multi-numa partitions within a
socket will be added by future patches, this enables 1P NPS1 asic
bringup configuration.
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rajneesh Bhardwaj [Sat, 28 Jan 2023 02:57:00 +0000 (21:57 -0500)]
drm/amdgpu: Handle VRAM dependencies on GFXIP9.4.3
[For 1P NPS1 mode driver bringup]
Changes required to initialize the amdgpu driver with frontdoor firmware
loading and discovery=2 with the native mode SBIOS that enables CPU GPU
unified interleaved memory.
sudo modprobe amdgpu discovery=2
Once PSP TMR region is reported via the ACPI interface, the dependency
on the ip_discovery.bin will be removed.
Choice of where to allocate driver table is given to each IP version. In
general, both GTT and VRAM domains will be considered. If one of the
tables has a strict restriction for VRAM domain, then only VRAM domain
is considered.
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
(lijo: Modified the handling for SMU Tables) Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Wed, 28 Dec 2022 10:18:38 +0000 (18:18 +0800)]
drm/amdgpu: Add mmhub v1_8_0 ras err status registers
add new ras error status registers introduced in
mmhub v1_8_0 to log mmea and mm_cane ras err, including
MMEAx_UE|CE_ERR_STATUS_LO|HI
MM_CANE_UE|CE_ERR_STATUS_LO|HI
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
projects / users / jedix / linux-maple.git / log