Commit Graph

780 Commits

Author SHA1 Message Date
Jammy Zhou
1060029fae drm/amd/powerplay: Add Tonga SMU support
The SMU manager handles firmware loading for other IP
blocks (GFX, SDMA, etc.).  This implements it for Tonga.

v3: delete peci sub-module
v2: use cgs interface directly

Signed-off-by: Young Yang <Young.Yang@amd.com>
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:14 -05:00
yanyang1
306d8db3e7 drm/amd/powerplay: add header file for tonga smu and dpm
These headers provide the SMU interface used by the driver.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: yanyang1 <young.yang@amd.com>
2015-12-21 16:42:14 -05:00
yanyang1
7ff1d70a40 drm/amd/powerplay: Move smu7*.h from amdgpu to powerplay.
Move smu7.h, smu7_discrete.h and smu7_fusion.h from amdgpu to powerplay.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: yanyang1 <young.yang@amd.com>
2015-12-21 16:42:13 -05:00
yanyang1
3a287055ae drm/amd/powerplay: Add ixSWRST_COMMAND_1 in bif_5_0_d.h
Add ixSWRST_COMMAND_1 in bif_5_0_d.h.  Required by
new powerplay code for tonga and fiji.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: yanyang1 <young.yang@amd.com>
2015-12-21 16:42:12 -05:00
Rex Zhu
577bbe0183 drm/amd/powerplay: implement functions of amd_powerplay_func
This is the common interface for interacting with the powerplay
module.

v2: squash in fixes

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Acked-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:12 -05:00
Rex Zhu
e92a037057 drm/amd/powerplay: add event manager sub-component
The event manager handles power related driver events.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:11 -05:00
Rex Zhu
28a18bab2e drm/amd/powerplay: add CG and PG support for carrizo
This adds clock and powergating support for CZ.

v2: squash in fixes

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:11 -05:00
Jammy Zhou
bdecc20a98 drm/amd/powerplay: add Carrizo dpm support
This patch enables basic DPM support for Carrizo.
DPM handles dynamic clock and voltage scaling.

v3: delete peci sub-module
v2: use cgs interface directly
    correct define SMU_EnabledFeatureScoreboard_SclkDpmOn

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:10 -05:00
Jammy Zhou
4630f0faae drm/amd/powerplay: add Carrizo smu support
This implements the SMU firmware manager interface for CZ.
Some header files are moved from amdgpu folder to powerplay as well.

v3: delete peci sub-module.
v2: use cgs interface directly
    add load_mec_firmware function

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:10 -05:00
Jammy Zhou
3bace35914 drm/amd/powerplay: add hardware manager sub-component
The hwmgr handles all hardware related calls, including clock/power
gating control, DPM, read and parse PPTable, etc.

v5: squash in fixes
v4: implement acpi's atcs function use cgs interface
v3: fix code style error and add big-endian mode support.
v2: use cgs interface directly in hwmgr sub-module

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:09 -05:00
Jammy Zhou
ac885b3a20 drm/amd/powerplay: add SMU manager sub-component
The SMUMGR is one sub-component of powerplay for SMU firmware support.
The SMU handles firmware loading for other IP blocks (GFX, SDMA, etc.)
on VI parts.  The adds the core powerplay infrastructure to handle that.

v3: direct use printk in powerplay module.
v2: direct use cgs_read/write_register functions in smu-modules

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:09 -05:00
Rex Zhu
1b5708ffb1 drm/amdgpu: export amd_powerplay_func to amdgpu and other ip block
Update amdgpu to deal with the new powerplay module properly.

v2: squash in fixes
v3: squash in Rex's power state reporting fix

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Acked-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:08 -05:00
Rex Zhu
ba5c2a87b0 drm/amdgpu: disable legacy path of firmware check if powerplay is enabled
Powerplay will use a different interface once it's integrated.  These
legacy pathes will be removed once powerplay is enabled by default.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:08 -05:00
Alex Deucher
1f7371b2a5 drm/amd/powerplay: add basic powerplay framework
amdgpu_pp_ip_funcs is introduced to handle the two code paths,
the legacy one and the new powerplay implementation.

CONFIG_DRM_AMD_POWERPLAY kernel configuration option is
introduced for the powerplay component.

v4: squash in fixes
v3: register debugfs file when powerplay module enable
v2: add amdgpu_ucode_init_bo in hw init when amdgpu_powerplay enable.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:07 -05:00
Rex Zhu
47bf18b5b2 drm/amdgpu: add new cgs interface to get display info (v2)
Add new CGS interfaces to query display info across modules.
This is nedded by the powerplay module for synchronizing with
the display module.

v2: (agd): fold in refresh rate fix, rebase

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:07 -05:00
Rex Zhu
5e6186991a drm/amdgpu: implement cgs interface to query system info
Add a query to get the bus number and function of the
device.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
2015-12-21 16:42:06 -05:00
Rex Zhu
3f1d35a03b drm/amdgpu: implement new cgs interface for acpi function
Add a new driver internal interface for accessing ACPI
methods.  These will be used by various new components
including powerplay.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:06 -05:00
Rex Zhu
66dc0ddd02 drm/amdgpu: mv amdgpu_acpi.h to amd/include/amd_acpi.h
This will be shared with the new powerplay module.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:05 -05:00
Rex Zhu
7e85be9948 drm/amdgpu: mv some definition from amdgpu_acpi.c to amdgpu_acpi.h
These will be shared with the new powerplay module.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:05 -05:00
Rex Zhu
3a2c788d95 drm/amdgpu: share struct amdgpu_pm_state_type with powerplay module
rename amdgpu_pm_state_type to amd_pm_state_type

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Acked-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:42:04 -05:00
Felix Kuehling
005ae95e6e drm/amdgpu: Fix off-by-one errors in amdgpu_vm_bo_map
eaddr is sometimes treated as the last address inside the address
range, and sometimes as the first address outside the range. This
was resulting in errors when a test filled up the entire address
space. Make it consistent to always be the last address within the
range.

Signed-off-by: Felix.Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Cc: stable@vger.kernel.org
2015-12-21 16:39:14 -05:00
Alex Deucher
41869c1c7f drm/amdgpu: fix dp link rate selection (v2)
Need to properly handle the max link rate in the dpcd.
This prevents some cases where 5.4 Ghz is selected when
it shouldn't be.

v2: simplify logic, add array bounds check

Reviewed-by: Tom St Denis <tom.stdenis@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-21 16:38:51 -05:00
Christian König
ee1782c3f2 drm/amdgpu: keep the PTs validation list in the VM v2
This avoids allocating it on the fly.

v2: fix grammar in comment

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
2015-12-18 17:29:46 -05:00
Christian König
56467ebfb2 drm/amdgpu: split VM PD and PT handling during CS
This way we avoid the extra allocation for the page directory entry.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
2015-12-18 17:29:45 -05:00
Christian König
3c0eea6c35 drm/amdgpu: put VM page tables directly into duplicates list
They share the reservation object with the page directory anyway.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
2015-12-18 17:29:45 -05:00
Chunming Zhou
5b0112356c drm/amdgpu: restrict the sched jobs number to power of two
Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
CC: stable@vger.kernel.org
2015-12-18 17:29:44 -05:00
Alex Deucher
a1493cd575 drm/amdgpu: limit visible vram if it's smaller than the BAR
In some cases the amount of vram may be less than the BAR size,
if so, limit visible vram to the amount of actual vram, not the
BAR size.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-17 11:56:41 -05:00
Chunming Zhou
b70f014d58 drm/amdgpu: change default sched jobs to 32
Change the default scheduler queue size from 16 to 32.

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-14 19:47:17 -05:00
Chunming Zhou
37cd0ca204 drm/amdgpu: unify AMDGPU_CTX_MAX_CS_PENDING and amdgpu_sched_jobs
Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-14 19:45:24 -05:00
Chunming Zhou
c648ed7c5c drm/amdgpu: handle error case for ctx
Properly handle ctx init failure.

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-14 19:42:11 -05:00
Chunming Zhou
e8deea2d4b drm/amdgpu: add entity only when first job come
umd somtimes will create a context for every ring,
that means some entities wouldn't be used at all.

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-14 19:41:19 -05:00
Alex Deucher
2c1a278403 drm/amdgpu: add more debugging output for driver failures
Add more fine grained debugging output for init/fini/suspend/
resume failures.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-11 11:13:42 -05:00
Flora Cui
abdfb850ca drm/amdgpu: update rev id register for VI
Change-Id: I2ae9bb4a929f7c0c8783e0be563ae04be77596e2
Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-11 11:13:41 -05:00
Flora Cui
c27816a883 drm/amdgpu/gfx8: update PA_SC_RASTER_CONFIG:PKR_MAP only
Use default value as a base.

Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
2015-12-11 11:13:41 -05:00
Flora Cui
3b55ddadef drm/amdgpu/gfx8: Enable interrupt on ME1_PIPE3
Otherwise FW cannot see the RLC ACK for the memory clean request
It's for Stoney.

Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
2015-12-11 11:13:40 -05:00
jimqu
9c4153b1ee drm/amdgpu: add spin lock to protect freed list in vm (v2)
there is a protection fault about freed list when OCL test.
add a spin lock to protect it.

v2: drop changes in vm_fini

Signed-off-by: JimQu <jim.qu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
2015-12-04 12:23:38 -05:00
Tom St Denis
eb64526f5a amdgpu/gfxv8: Remove magic numbers from function gfx_v8_0_tiling_mode_table_init()
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-04 11:26:52 -05:00
Nicolai Hähnle
786b521908 drm/amdgpu: fix race condition in amd_sched_entity_push_job
As soon as we leave the spinlock after the job has been added to the job
queue, we can no longer rely on the job's data to be available.

I have seen a null-pointer dereference due to sched == NULL in
amd_sched_wakeup via amd_sched_entity_push_job and
amd_sched_ib_submit_kernel_helper. Since the latter initializes
sched_job->sched with the address of the ring scheduler, which is
guaranteed to be non-NULL, this race appears to be a likely culprit.

Signed-off-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Bugzilla: https://bugs.freedesktop.org/attachment.cgi?bugid=93079
Reviewed-by: Christian König <christian.koenig@amd.com>
2015-12-04 11:26:52 -05:00
Chunming Zhou
ba98f9e56c drm/amdgpu: add err check for pin userptr
Missing error check if the operation failed.

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
2015-12-04 11:26:51 -05:00
Tom St Denis
0d07db7e10 amdgpu/gfxv8: Simplification in gfx_v8_0_enable_gui_idle_interrupt()
Simplified the function by folding the two paths into one.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-04 11:26:50 -05:00
Tom St Denis
544b8a74c7 amdgpu/gfxv8: Simplification of gfx_v8_0_create_bitmask()
Simplification of the function gfx_v8_0_create_bitmask().

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-04 11:26:50 -05:00
Tom St Denis
90bea0abf6 amdgpu/gfxv8: Cleanup of gfx_v8_0_tiling_mode_table_init() (v2)
Simplification and LOC reduction of function gfx_v8_0_tiling_mode_table_init()

v2: remove spurious break
bug: https://bugs.freedesktop.org/show_bug.cgi?id=93236

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-04 11:26:49 -05:00
Tom St Denis
8cdacf4457 amdgpu/gfxv8: Add missing break to switch statement from states init code
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
2015-12-02 15:54:33 -05:00
Chunming Zhou
d033a6de80 drm/amd: abstract kernel rq and normal rq to priority of run queue
Allows us to set priorities in the scheduler.

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
2015-12-02 15:54:33 -05:00
Alex Deucher
ccba7691a5 drm/amdgpu: add EDC support for CZ (v3)
This adds EDC support for CZ.
EDC = Error Correction and Detection
This code properly initializes the EDC hardware and
resets the error counts.  This is done in late_init
since it requires the IB pool which is not initialized
during hw_init.

v2: fix the IB size as noted by Felix, fix shader pgm
register programming
v3: use the IB for the shaders as suggested by Christian

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-02 15:54:26 -05:00
Alex Deucher
aa5e24e5f8 drm/amd: add new gfx8 register definitions for EDC
EDC is a RAS feature for on chip memory.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-02 15:54:18 -05:00
Daniel Vetter
a9906fde57 drm/amdgpu: Use unlocked gem unreferencing
For drm_gem_object_unreference callers are required to hold
dev->struct_mutex, which these paths don't. Enforcing this requirement
has become a bit more strict with

commit ef4c6270bf
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Thu Oct 15 09:36:25 2015 +0200

    drm/gem: Check locking in drm_gem_object_unreference

Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-02 12:46:00 -05:00
monk.liu
f930b2e862 drm/amdgpu: Use new read bios from rom callback
Read the vbios directly from the rom.  In some cases,
e.g., virtualization, the rom is not available via
the BAR or other means.  Access it directly.

This is an updated version of Monks original patch which
uses family specific callbacks and unifies some of the
validation checking.

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-02 12:45:59 -05:00
Alex Deucher
95addb2ae0 drm/amdgpu: add read_bios_from_rom callback for VI parts
Read the vbios image directly from the rom.

Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-02 12:45:58 -05:00
Alex Deucher
1eb22bd38a drm/amdgpu: add read_bios_from_rom callback for CI parts
Read the vbios image directly from the rom.

Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2015-12-02 12:45:58 -05:00