Kevin Wang [Fri, 21 Jun 2019 15:49:22 +0000 (10:49 -0500)]
drm/amd/powerplay: implement smc firmware v2.1 for smu11
1.add smc_firmware_header_v2_1 hfirmware support, support more pptable in smc firmware.
2.optimization current pptable load framework.
3.rename read_pptable_from_vbios with setup_pptable.
Signed-off-by: Kevin Wang <kevin1.wang@amd.com> Reviewed-by: Huang Rui <ray.huang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Kenneth Feng [Mon, 11 Mar 2019 08:06:03 +0000 (16:06 +0800)]
drm/amd/powerplay: enable power features
the below smu related power features can be enabled now.
1.Prefetcher
2.GFX DPM
3.SOCCLK DPM
4.MP0CLK DPM
5.LCLK DPM
6.GFX ULV
7.CG
8.PPT
9.TDC
10.GFX EDC
11.VR0HOT
12.Fan Control
13.Thermal Control
14.LED Display
15.MMHub PG
16.ATHub PG
Huang Rui [Wed, 20 Feb 2019 12:05:11 +0000 (20:05 +0800)]
drm/amd/powerplay: smu needs to be initialized after rlc in direct mode
For gfx 10, rlc firmware loading relies on smu firmware is loaded firstly, so in
direct type, it has to load smc ucode here before rlc. And meanwhile, the smu
initialization has to move after rlc, otherwise, smu message will get failure
during the handshake with rlc and smu.
Huang Rui [Thu, 31 Jan 2019 11:46:26 +0000 (19:46 +0800)]
drm/amd/powerplay: introduce the navi10 pptable implementation
This patch introduces the navi10 pptable implementation, so far it is already
has firmware loading, pptable side loading, writing back to smc, and feature
mask enabling.
Philip Cox [Thu, 30 May 2019 04:03:45 +0000 (23:03 -0500)]
drm/amdkfd: Add navi10 support to amdkfd. (v3)
KFD (kernel fusion driver) is the kernel driver
for the compute backend for usermode compute
stack.
v2: squash in updates (Alex)
v3: squash in rebase fixes (Alex)
Signed-off-by: Oak Zeng <Oak.Zeng@amd.com> Signed-off-by: Philip Cox <Philip.Cox@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Tue, 5 Mar 2019 11:59:30 +0000 (19:59 +0800)]
drm/amdgpu: Add navi10 kfd support for amdgpu (v3)
KFD (Kernel Fusion Driver) is the compute backend driver
for AMD GPUs.
v2: squash in updates (Alex)
v3: fix warnings (Alex)
Signed-off-by: Oak Zeng <Oak.Zeng@amd.com> Signed-off-by: Philip Cox <Philip.Cox@amd.com> Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Mon, 4 Mar 2019 06:07:37 +0000 (14:07 +0800)]
drm/amdgpu: add navi10 common ip block (v3)
This adds the core SOC code for navi asics.
v1: add place holder and initial basic function (Ray)
v2: add new introduced functions to avoid reference
NULL pointer (Hawking)
v3L squash in updates (Alex)
Signed-off-by: Huang Rui <ray.huang@amd.com> Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Mon, 4 Mar 2019 06:41:42 +0000 (14:41 +0800)]
drm/amdgpu: add gfx v10 implementation (v10)
GFX is the graphics and compute block on the GPU.
v1: add initial gfx v10 implementation (Ray)
v2: convert to new get_vm_pde function in emit_vm_flush (Hawking)
v3: switch to new emit ib interfaces (Hawking)
v4: squash in updates (Alex)
v5: remove unused variables (Alex)
v6: v6: some golden regs moved to vbios (Alex)
v7: squash in some cleanups (Alex)
v8: squash in golden settings update (Alex)
v9: squash in whitespace fixes (Ernst Sjöstrand, Alex)
v10: squash in GDS backup size fix and GDS/GWS/OA removal rebase fixes (Hawking)
Signed-off-by: Huang Rui <ray.huang@amd.com> Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Jack Xiao [Fri, 25 Jan 2019 06:11:35 +0000 (14:11 +0800)]
drm/amdgpu/mes: add amdgpu_mes driver parameter
amdgpu_mes, which is a driver scope parameter, is used
to whether enable mes or not.
MES (Micro Engine Scheduler) is the new on chip hw scheduling
microcontroller. It can be used to handle queue scheduling and
preemption and priorities.
Acked-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Jack Xiao <Jack.Xiao@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Leo Liu [Mon, 15 Oct 2018 15:38:59 +0000 (11:38 -0400)]
drm/amdgpu: add initial VCN2.0 support (v2)
VCN (Video Core Next) is the video encode/decode block.
Porting over the same functions from VCN1.0
v2: squash in updates (Alex)
Signed-off-by: Leo Liu <leo.liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: James Zhu <James.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Sun, 31 Mar 2019 14:13:57 +0000 (22:13 +0800)]
drm/amdgpu: add initial support for sdma v5.0 (v6)
SDMA (System DMA) is a general purpose DMA engine usable
by UMDs for transfers or the kernel for paging or GPUVM
updates.
v1: support basic funcitonalites includes rb, ib, vm,
copy buffer and trap irq
v2: convert to use new get_vm_pde in emit_vm_flush
v3: retire amdgpu_ttm_set_active_vram_size from sdma v5
v4: retire the redundant hdp_invalidate implementation
v5: squash in updates
v6: some golden regs moved to vbios
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Christian König [Thu, 7 Feb 2019 11:10:29 +0000 (12:10 +0100)]
drm/amdgpu: disable concurrent flushes for Navi10 v2
Navi10 have a bug in the SDMA which can theoretically cause memory
corruption with concurrent VMID flushes
v2: explicitely check Navi10
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>
Xiaojie Yuan [Wed, 27 Mar 2019 04:21:40 +0000 (12:21 +0800)]
drm/amdgpu/discovery: stop taking psp header into account
psp will write a header to vram, but the value exposed in
RCC_CONFIG_MEMSIZE does not include the memory that this header is
written to. Therefore, the interpretation of the table does not need to
take the psp header into account.
Xiaojie Yuan [Thu, 20 Jun 2019 15:18:50 +0000 (10:18 -0500)]
drm/amdgpu/discovery: add ip discovery initial support
The IP discovery table lists is populated by the psp at power on
and includes all of the hw details on the board:
- List of IPs and MMIO offsets
- IP harvest details
- IP configuration details
v2: prefix struct and function names with 'amdgpu'
v3: read table binary from vram using mmMM_INDEX and mmMM_DATA
update TABLE_BINARY_MAX_SIZE to 64kb (1 TMR)
add 'instance_number' field per ip info
consider endianness and replace uint8/16/32_t with u8/16/32
initialize register base addresses
initialize adev->gfx.config and adev->gfx.cu_info to replace gpu info fw
get major and minor version using a single api
don't expose internal data structures in amdgpu_discovery.h
v4: RCC_CONFIG_MEMSIZE is in MB units
hold mmio_idx_lock while reading ip discovery binary
v5: pick out discovery.h as a cross-OS header
do structure pointer cast directly
consider endianness while using the member of structure
convert base addresses to dword
at boot up, PSP BL copies ip discovery binary from VBIOS(SPIROM) image to the
top of the frame buffer (just below the reserved regions for PSP & SMU).
ip discovery data table includes the collection of each ip's identification
number, base addresses, version number, and harvest setting placeholder.
gc data table includes gfx info structure.
Signed-off-by: Xiaojie Yuan <xiaojie.yuan@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Jack Xiao [Wed, 23 Jan 2019 05:54:26 +0000 (13:54 +0800)]
drm/amdgpu: mark the partial job as preempted in mcbp unit test
In mcbp unit test, the test should detect the preempted job which may
be a partial execution ib and mark it as preempted; so that the gfx
block can correctly generate PM4 frame.
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Jack Xiao <Jack.Xiao@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Wed, 12 Dec 2018 17:29:53 +0000 (01:29 +0800)]
drm/amdgpu/psp: skip mec jt when autoload is enabled
When autoload is enabled, there is no need to load mec jt,
RLC will handle it automatically
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Jack Xiao <Jack.Xiao@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Wed, 12 Dec 2018 17:21:30 +0000 (01:21 +0800)]
drm/amd/amdgpu: add flag to mark whether autoload is supported or not
rlc autoload is supported since navi10
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Jack Xiao <Jack.Xiao@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Wed, 12 Dec 2018 17:23:56 +0000 (01:23 +0800)]
drm/amdgpu/psp: initialize autoload_supported flag in psp_sw_init
RLC autoload is supported since from Navi10
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Jack Xiao <Jack.Xiao@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Fri, 10 May 2019 15:58:44 +0000 (10:58 -0500)]
drm/amdgpu/psp: update psp gfx interface to match with psp fw (v2)
new psp gfx cmd is introuduced for rlc autoload
v2: rebase (Alex)
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Jack Xiao <Jack.Xiao@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Mon, 12 Nov 2018 08:33:08 +0000 (16:33 +0800)]
drm/amdgpu/psp: perform tmr_init and asd_init after loading sysdrv/sos
Since from navi10, the tmr_size should be decided by psp sos according to
toc header. Driver should issue LOAD_TOC to psp sos to get the tmr_size needed.
The allocation of tmr_size then should be done only when sos/sysdrv loading
completed
Accordingly, asd_init also move to psp_hw_start after sos fw loading to make
calling sequence consistent.
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Fri, 19 Oct 2018 13:46:05 +0000 (21:46 +0800)]
drm/amdgpu: add structure to support build-in toc to psp sos
Table Of Content (TOC) is used by RLC to auto load gc firmwares.
PSP need to parse the toc to calculate the tmr size needed and
load gc firmwares to tmr for RLC to auto load them finally
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tao Zhou [Mon, 21 May 2018 08:32:05 +0000 (16:32 +0800)]
drm/amdgpu: Add psp 11.0 support for navi10.
Add psp 11.0 code for navi10. psp 11.0 is not enabled for now.
Will enable it when psp 11.0 firmware is available.
Signed-off-by: Tao Zhou <Tao.Zhou1@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Mon, 4 Mar 2019 05:49:28 +0000 (13:49 +0800)]
drm/amdgpu: add mmhub v2 block for navi10 (v4)
mmhub is the memory controller hub for multi-media (VCN).
v1: add place holder and initial functions (Ray)
v2: replace legacy amdgpu_mc structure with amdgpu_gmc (Hawking)
v3: switch to use amdgpu_gmc_pd_addr (Hawking)
v4: squash in updates (Alex)
Signed-off-by: Huang Rui <ray.huang@amd.com> Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hawking Zhang [Mon, 4 Mar 2019 08:18:27 +0000 (16:18 +0800)]
drm/amdgpu: add gfxhub v2.0 block for navi10 (v4)
gfxhub is the memory controller hub for gfx and sdma.
v1: add place holder and initial basic functions (Ray)
v2: replace the refernce to legacy mc structure with gmc structure
remove the direct use of gart.table_addr (Hawking)
v3: switch to use amdgpu_gmc_pd_addr (Hawking)
v4: squash in updates (Alex)
Signed-off-by: Huang Rui <ray.huang@amd.com> Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>