]> git.proxmox.com Git - mirror_ubuntu-kernels.git/log
mirror_ubuntu-kernels.git
6 years agodrm/amdgpu: fix shadow BO restoring
Christian König [Tue, 11 Sep 2018 09:50:57 +0000 (11:50 +0200)]
drm/amdgpu: fix shadow BO restoring

Don't grab the reservation lock any more and simplify the handling quite
a bit.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: always recover VRAM during GPU recovery
Christian König [Tue, 11 Sep 2018 08:36:16 +0000 (10:36 +0200)]
drm/amdgpu: always recover VRAM during GPU recovery

It shouldn't add much overhead and we should make sure that critical
VRAM content is always restored.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: shadow BOs don't need any alignment
Christian König [Tue, 11 Sep 2018 08:31:54 +0000 (10:31 +0200)]
drm/amdgpu: shadow BOs don't need any alignment

They aren't directly used by the hardware.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: always enable shadow BOs v2
Christian König [Tue, 11 Sep 2018 08:30:31 +0000 (10:30 +0200)]
drm/amdgpu: always enable shadow BOs v2

Even when GPU recovery is disabled we could run into a manually
triggered recovery.

v2: keep accidental removed comments

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Emily Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: stop pipelining VM PDs/PTs moves
Christian König [Tue, 11 Sep 2018 07:30:46 +0000 (09:30 +0200)]
drm/amdgpu: stop pipelining VM PDs/PTs moves

We are going to need this for recoverable page fault handling and it
makes shadow handling during GPU reset much more easier.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: remove fence fallback
Christian König [Tue, 18 Sep 2018 08:38:09 +0000 (10:38 +0200)]
drm/amdgpu: remove fence fallback

DC doesn't seem to have a fallback path either.

So when interrupts doesn't work any more we are pretty much busted no
matter what.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/powerplay: retrieve the updated clock table after OD
Evan Quan [Mon, 17 Sep 2018 06:59:54 +0000 (14:59 +0800)]
drm/amd/powerplay: retrieve the updated clock table after OD

With OD settings applied, the clock table will be updated accordingly.
We need to retrieve the new clock tables then.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/powerplay: update OD to take voltage value instead of offset
Evan Quan [Thu, 13 Sep 2018 08:14:33 +0000 (16:14 +0800)]
drm/amd/powerplay: update OD to take voltage value instead of offset

With the latest SMC fw, we are able to get the voltage value for
specific frequency point. So, we update the OD relates to take
absolute voltage instead of offset.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/powerplay: update OD feature judgement
Evan Quan [Wed, 12 Sep 2018 03:45:01 +0000 (11:45 +0800)]
drm/amd/powerplay: update OD feature judgement

Update the conditions to judge whether an OD feature
should be supported on vega20.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: fix unknown vram mem type for vega20
Hawking Zhang [Mon, 17 Sep 2018 12:19:48 +0000 (20:19 +0800)]
drm/amdgpu: fix unknown vram mem type for vega20

vega20 should use umc_info v3_3 instead of v3_1. There are
serveral versions of umc_info for vega series. Compared to
various versions of these structures, vram_info strucure is
unified for vega series. The patch switch to query mem_type
from vram_info structure for all the vega series dGPU.

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>
6 years agodrm/amdgpu: update vram_info structure in atomfirmware.h
Hawking Zhang [Mon, 17 Sep 2018 12:25:03 +0000 (20:25 +0800)]
drm/amdgpu: update vram_info structure in atomfirmware.h

atomfirmware has structure changes in varm_info. Updated it
to the latest one.

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>
6 years agodrm/radeon: change function signature to pass full range
Mathieu Malaterre [Thu, 12 Apr 2018 19:33:33 +0000 (21:33 +0200)]
drm/radeon: change function signature to pass full range

In function ‘radeon_process_i2c_ch’ a comparison of a u8 value against
255 is done. Since it is always false, change the signature of this
function to use an `int` instead, which match the type used in caller:
`radeon_atom_hw_i2c_xfer`.

Fix the following warning triggered with W=1:

  CC [M]  drivers/gpu/drm/radeon/atombios_i2c.o
  drivers/gpu/drm/radeon/atombios_i2c.c: In function ‘radeon_process_i2c_ch’:
  drivers/gpu/drm/radeon/atombios_i2c.c:71:11: warning: comparison is always false due to limited range of data type [-Wtype-limits]
   if (num > ATOM_MAX_HW_I2C_READ) {
           ^

Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Mathieu Malaterre <malat@debian.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: use processed values for counting
A. Wilcox [Mon, 2 Jul 2018 03:44:52 +0000 (22:44 -0500)]
drm/amdgpu: use processed values for counting

adev->gfx.rlc has the values from rlc_hdr already processed by
le32_to_cpu.  Using the rlc_hdr values on big-endian machines causes
a kernel Oops due to writing well outside of the array (0x24000000
instead of 0x24).

Signed-off-by: A. Wilcox <AWilcox@Wilcox-Tech.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/amdgpu: Avoid fault when allocating an empty buffer object
Tom St Denis [Mon, 17 Sep 2018 18:07:00 +0000 (14:07 -0400)]
drm/amd/amdgpu: Avoid fault when allocating an empty buffer object

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: drop size check
Christian König [Fri, 14 Sep 2018 19:06:50 +0000 (21:06 +0200)]
drm/amdgpu: drop size check

We no don't allocate zero sized kernel BOs any longer.

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>
6 years agodrm/amdgpu: don't allocate zero sized kernel BOs
Christian König [Fri, 14 Sep 2018 19:03:37 +0000 (21:03 +0200)]
drm/amdgpu: don't allocate zero sized kernel BOs

Just free the BO if the size should be zero.

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>
6 years agodrm/amdgpu: stop crashing on GDS/GWS/OA eviction
Christian König [Fri, 14 Sep 2018 18:44:17 +0000 (20:44 +0200)]
drm/amdgpu: stop crashing on GDS/GWS/OA eviction

Simply ignore any copying here.

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>
6 years agodrm/amdgpu: add GDS, GWS and OA debugfs files
Christian König [Fri, 14 Sep 2018 13:43:57 +0000 (15:43 +0200)]
drm/amdgpu: add GDS, GWS and OA debugfs files

Additional to the existing files for VRAM and GTT.

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>
6 years agodrm/amdgpu: fix parameter documentation for amdgpu_vm_free_pts
Christian König [Sat, 15 Sep 2018 08:04:54 +0000 (10:04 +0200)]
drm/amdgpu: fix parameter documentation for amdgpu_vm_free_pts

The function was modified without updating the documentation.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add amdgpu_vm_entries_mask v2
Christian König [Sat, 15 Sep 2018 08:02:13 +0000 (10:02 +0200)]
drm/amdgpu: add amdgpu_vm_entries_mask v2

We can't get the mask for the root directory from the number of entries.

So add a new function to avoid that problem.

v2: fix typo in mask

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display: Drop amdgpu_dm_prev_state struct
Leo Li [Wed, 5 Sep 2018 15:28:29 +0000 (11:28 -0400)]
drm/amd/display: Drop amdgpu_dm_prev_state struct

[Why]
It's not being used

[How]
Nuke it

Signed-off-by: Leo Li <sunpeng.li@amd.com>
Reviewed-by: David Francis <David.Francis@amd.com>
Acked-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display: Drop amdgpu_display_manager.dal member
Leo Li [Wed, 5 Sep 2018 15:19:42 +0000 (11:19 -0400)]
drm/amd/display: Drop amdgpu_display_manager.dal member

[Why]
It's not being used anymore.

[How]
Nuke it

Signed-off-by: Leo Li <sunpeng.li@amd.com>
Reviewed-by: David Francis <David.Francis@amd.com>
Acked-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display: add query HPD interface.
Chiawen Huang [Wed, 5 Sep 2018 12:34:57 +0000 (20:34 +0800)]
drm/amd/display: add query HPD interface.

[Why]
current dc_link_detect function is not only detection but also update some link data.

[How]
added a pure get HPD state function.

Signed-off-by: Chiawen Huang <chiawen.huang@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display: dc 3.1.66
Tony Cheng [Mon, 27 Aug 2018 17:35:31 +0000 (13:35 -0400)]
drm/amd/display: dc 3.1.66

Signed-off-by: Tony Cheng <tony.cheng@amd.com>
Reviewed-by: Steven Chiu <Steven.Chiu@amd.com>
Acked-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display: add aux i2c event log.
Chiawen Huang [Wed, 29 Aug 2018 10:39:38 +0000 (18:39 +0800)]
drm/amd/display: add aux i2c event log.

[Why]
support i2c transition event log

[How]
refined aux REQ and REP events in aux flow.
commented REQ and REP events in i2c flow.

note: i2c event log is currently commented out. more work is required
to find an portocol parser to and generate event for the parser

Signed-off-by: Chiawen Huang <chiawen.huang@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display: stop using switch for different CS revisions
Dmytro Laktyushkin [Wed, 29 Aug 2018 20:23:59 +0000 (16:23 -0400)]
drm/amd/display: stop using switch for different CS revisions

Clock sources currently have support for asic specific
function pointers. But actual separation into functions
was never performed, leaving us with giant functions that
rely on switch.

This change creates separate functions, removing switch use.

Signed-off-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display: Fix 3D stereo issues.
Charlene Liu [Mon, 27 Aug 2018 15:31:08 +0000 (11:31 -0400)]
drm/amd/display: Fix 3D stereo issues.

We were not providing the correct pixel clocks to DML for marks
calculation.

Signed-off-by: Charlene Liu <charlene.liu@amd.com>
Reviewed-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Acked-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: revert "stop using gart_start as offset for the GTT domain"
Christian König [Fri, 14 Sep 2018 10:54:33 +0000 (12:54 +0200)]
drm/amdgpu: revert "stop using gart_start as offset for the GTT domain"

Turned out the commit is incomplete and since we remove using the AGP
mapping from the GTT manager it is also not necessary any more.

This reverts commit 22d8bfafcc12dfa17b91d2e8ae4e1898e782003a.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: James Zhu <James.Zhu@amd.com>
Tested-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: fix mask in GART location calculation
Christian König [Fri, 14 Sep 2018 08:17:24 +0000 (10:17 +0200)]
drm/amdgpu: fix mask in GART location calculation

We need to mask the lower bits not the upper one.

Fixes: ec210e3226dc0 drm/amdgpu: put GART away from VRAM v2
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: James Zhu <James.Zhu@amd.com>
Tested-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu/display: return proper error codes in dm
Alex Deucher [Thu, 13 Sep 2018 16:01:28 +0000 (11:01 -0500)]
drm/amdgpu/display: return proper error codes in dm

Replace -1 with proper error codes.

Acked-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: simplify Raven, Raven2, and Picasso handling
Alex Deucher [Thu, 13 Sep 2018 20:41:57 +0000 (15:41 -0500)]
drm/amdgpu: simplify Raven, Raven2, and Picasso handling

Treat them all as Raven rather than adding a new picasso
asic type.  This simplifies a lot of code and also handles the
case of rv2 chips with the 0x15d8 pci id.  It also fixes dmcu
fw handling for picasso.

Acked-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu/soc15: clean up picasso support
Alex Deucher [Thu, 13 Sep 2018 20:05:22 +0000 (15:05 -0500)]
drm/amdgpu/soc15: clean up picasso support

It's the same as raven so remove the duplicate case.

Acked-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: Add DMCU to firmware query interface
David Francis [Thu, 13 Sep 2018 19:37:50 +0000 (15:37 -0400)]
drm/amdgpu: Add DMCU to firmware query interface

DMCU firmware version can be read using the AMDGPU_INFO ioctl
or the amdgpu_firmware_info debugfs entry

Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display: Add DMCU firmware version
David Francis [Thu, 13 Sep 2018 19:36:27 +0000 (15:36 -0400)]
drm/amd/display: Add DMCU firmware version

Read the version number from the common firmware header and store
it in the dm struct

Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: Style fixes to PRIME code documentation
Vijetha Malkai [Thu, 13 Sep 2018 18:47:39 +0000 (14:47 -0400)]
drm/amdgpu: Style fixes to PRIME code documentation

* Use consistent capitalization in the description of function arguments
* Define and consistently use the BO acronym for buffer objects
* Some minor wording improvements

Signed-off-by: Vijetha Malkai <vijetha.malkai@amd.com>
[ Michel Dänzer: Made commit log more specific ]

Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: Initialize fences array entries in amdgpu_sa_bo_next_hole
Michel Dänzer [Wed, 12 Sep 2018 16:07:10 +0000 (18:07 +0200)]
drm/amdgpu: Initialize fences array entries in amdgpu_sa_bo_next_hole

The entries were only initialized once in amdgpu_sa_bo_new. If a fence
wasn't signalled yet in the first amdgpu_sa_bo_next_hole call, but then
got signalled before a later amdgpu_sa_bo_next_hole call, it could
destroy the fence but leave its pointer in the array, resulting in
use-after-free in amdgpu_sa_bo_new.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: set CG flags for raven2 (v2)
Huang Rui [Tue, 19 Jun 2018 15:46:42 +0000 (10:46 -0500)]
drm/amdgpu: set CG flags for raven2 (v2)

Raven2 does not enable all of the CG flags that raven1 does.

v2: rebase (Alex)

Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/powerplay: disable raven2 force dpm level support (v2)
Evan Quan [Tue, 19 Jun 2018 15:41:00 +0000 (10:41 -0500)]
drm/amd/powerplay: disable raven2 force dpm level support (v2)

It's not supported yet.

v2: rebase (Alex)

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/powerplay: round up the Mhz convertion (v2)
Evan Quan [Tue, 19 Jun 2018 15:32:50 +0000 (10:32 -0500)]
drm/amd/powerplay: round up the Mhz convertion (v2)

Since the clock value there may be like 29999 10Khz.

v2: rebase (Alex)

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/powerplay: update smu10_verify_smc_interface() to be raven2 compatible
Evan Quan [Fri, 15 Jun 2018 22:22:38 +0000 (17:22 -0500)]
drm/amd/powerplay: update smu10_verify_smc_interface() to be raven2 compatible

Check the raven2 version number as well.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display: Add DC config flag for Raven2 (v2)
Bhawanpreet Lakha [Mon, 22 Jan 2018 22:40:50 +0000 (17:40 -0500)]
drm/amd/display: Add DC config flag for Raven2 (v2)

Add DRM_AMD_DC_DCN1_01 config flag for Raven2

v2: Make DC select DRM_AMD_DC_DCN1_01 (Alex)

Signed-off-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display: Add Raven2 definitions in dc
Bhawanpreet Lakha [Mon, 22 Jan 2018 21:12:27 +0000 (16:12 -0500)]
drm/amd/display: Add Raven2 definitions in dc

Add Raven2 definitions in the dc code

Signed-off-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: fix the VM fault while write at the top of the invisible vram
Huang Rui [Tue, 16 Jan 2018 02:42:58 +0000 (10:42 +0800)]
drm/amdgpu: fix the VM fault while write at the top of the invisible vram

Raven2 has a HW issue that it is unable to use the vram which is out of
MC_VM_SYSTEM_APERTURE_HIGH_ADDR. So here is the workaround that increase system
aperture high address to get rid of the VM fault and hardware hang.

Signed-off-by: Huang Rui <ray.huang@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>
6 years agodrm/amdgpu/gfx9: add raven2 golden setting
Feifei Xu [Thu, 4 Jan 2018 10:36:40 +0000 (18:36 +0800)]
drm/amdgpu/gfx9: add raven2 golden setting

Golden register settings from the hw team.

Signed-off-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu/gfx9: add support for raven2 gfx firmware
Feifei Xu [Thu, 4 Jan 2018 10:33:49 +0000 (18:33 +0800)]
drm/amdgpu/gfx9: add support for raven2 gfx firmware

use raven2 gfx firmware.

Signed-off-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu/sdma4: Add raven2 golden setting
Feifei Xu [Thu, 4 Jan 2018 10:13:41 +0000 (18:13 +0800)]
drm/amdgpu/sdma4: Add raven2 golden setting

Golden register settings from the hw team.

Signed-off-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu/sdma4: specify raven2 firmware.
Feifei Xu [Thu, 4 Jan 2018 10:05:35 +0000 (18:05 +0800)]
drm/amdgpu/sdma4: specify raven2 firmware.

use raven2 sdma firmware.

Signed-off-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add psp support for raven2
Likun Gao [Tue, 5 Jun 2018 06:05:45 +0000 (14:05 +0800)]
drm/amdgpu: add psp support for raven2

Modified for using raven2_asd.bin to replace raven_asd.bin for raven2

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add raven2 vcn firmware support
Feifei Xu [Fri, 15 Jun 2018 21:01:41 +0000 (16:01 -0500)]
drm/amdgpu: add raven2 vcn firmware support

Specify raven2 vcn firmware on amdgpu_vce_sw_init.

Signed-off-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add raven2 to gpu_info firmware
Feifei Xu [Thu, 4 Jan 2018 09:26:00 +0000 (17:26 +0800)]
drm/amdgpu: add raven2 to gpu_info firmware

Add gpu_info firmware for raven2.

Signed-off-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: set external rev id for raven2
Alex Deucher [Fri, 15 Jun 2018 22:28:44 +0000 (17:28 -0500)]
drm/amdgpu: set external rev id for raven2

It's different from raven1.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: move get_rev_id at first before load gpu_info firmware
Huang Rui [Fri, 15 Jun 2018 21:05:48 +0000 (16:05 -0500)]
drm/amdgpu: move get_rev_id at first before load gpu_info firmware

Rev id is used for identifying Raven2 series of chips. So we would better to
initialize it at first.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: use IP presence to free uvd and vce handles
Alex Deucher [Wed, 12 Sep 2018 18:51:25 +0000 (13:51 -0500)]
drm/amdgpu: use IP presence to free uvd and vce handles

Rather than checking the asic type, check whether the UVD
or VCE IP blocks exist.  This way we don't have to update
the check with new asics that use VCN.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: enable gfxoff in non-sriov and stutter mode by default
Kenneth Feng [Thu, 6 Sep 2018 06:56:19 +0000 (14:56 +0800)]
drm/amdgpu: enable gfxoff in non-sriov and stutter mode by default

enable gfxoff in non-sriov and stutter mode by default

Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add new raven series device
Huang Rui [Mon, 9 Jul 2018 11:51:19 +0000 (19:51 +0800)]
drm/amdgpu: add new raven series device

This patch is to add new pci device for raven series.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add ip blocks for picasso (v2)
Huang Rui [Mon, 9 Jul 2018 12:00:05 +0000 (20:00 +0800)]
drm/amdgpu: add ip blocks for picasso (v2)

Add PCO IPs.

V2: enable VCN as well

Signed-off-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: enable vcn powergating for PCO
Kenneth Feng [Fri, 24 Aug 2018 08:44:11 +0000 (16:44 +0800)]
drm/amdgpu: enable vcn powergating for PCO

enable vcn pg

Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: enable mmhub power gating
Kenneth Feng [Mon, 20 Aug 2018 07:39:32 +0000 (15:39 +0800)]
drm/amdgpu: enable mmhub power gating

Remove some functions due to the design change.
All the mmhub power gating sequence is moved to
smu fw.Driver sends the message to enable mmhub
powergating.We can also skip the fw version check
since the old fw version is in a very early stage
and we don't use that fw for release.

Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: Enable SDMA power gating for PCO
Kenneth Feng [Fri, 10 Aug 2018 08:22:26 +0000 (16:22 +0800)]
drm/amdgpu: Enable SDMA power gating for PCO

Enable SDMA power gating

Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Acked-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: Add pg support for gfxoff for PCO
Kenneth Feng [Tue, 7 Aug 2018 09:05:22 +0000 (17:05 +0800)]
drm/amdgpu: Add pg support for gfxoff for PCO

Add pg support for gfxoff.

Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add picasso for amdgpu kms
Likun Gao [Tue, 10 Jul 2018 12:34:10 +0000 (20:34 +0800)]
drm/amdgpu: add picasso for amdgpu kms

Add picasso for amdgpu kms

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add picasso support for sdma_v4
Likun Gao [Tue, 10 Jul 2018 12:30:42 +0000 (20:30 +0800)]
drm/amdgpu: add picasso support for sdma_v4

Add sdma support to picasso

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add picasso support for gfx_v9_0
Likun Gao [Tue, 10 Jul 2018 12:29:12 +0000 (20:29 +0800)]
drm/amdgpu: add picasso support for gfx_v9_0

Add gfx support to picasso

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add picasso support for gmc
Likun Gao [Tue, 10 Jul 2018 12:26:41 +0000 (20:26 +0800)]
drm/amdgpu: add picasso support for gmc

Same as raven.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add clockgating support for picasso
Likun Gao [Tue, 10 Jul 2018 12:25:24 +0000 (20:25 +0800)]
drm/amdgpu: add clockgating support for picasso

Treat it the same as raven for now.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add picasso support for vcn
Likun Gao [Tue, 10 Jul 2018 12:17:13 +0000 (20:17 +0800)]
drm/amdgpu: add picasso support for vcn

Add vcn support for picasso.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add picasso ucode loading method
Likun Gao [Tue, 10 Jul 2018 12:15:45 +0000 (20:15 +0800)]
drm/amdgpu: add picasso ucode loading method

Same as raven.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: initilize picasso psp firmwares support
Likun Gao [Tue, 10 Jul 2018 12:12:38 +0000 (20:12 +0800)]
drm/amdgpu: initilize picasso psp firmwares support

Same as raven for now.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: Add support of powerplay for picasso
Likun Gao [Tue, 10 Jul 2018 11:25:29 +0000 (19:25 +0800)]
drm/amdgpu: Add support of powerplay for picasso

add powerplay support for picasso, treat it the same as raven now.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display/dm: add picasso support
Likun Gao [Tue, 10 Jul 2018 12:32:06 +0000 (20:32 +0800)]
drm/amd/display/dm: add picasso support

Add support for picasso to the display manager.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add picasso support for vm
Likun Gao [Tue, 10 Jul 2018 12:20:16 +0000 (20:20 +0800)]
drm/amdgpu: add picasso support for vm

Add vm support for picasso.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add soc15 support for picasso
Likun Gao [Tue, 10 Jul 2018 12:22:36 +0000 (20:22 +0800)]
drm/amdgpu: add soc15 support for picasso

Add the IP blocks, clock and powergating flags, and common clockgating support.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add picasso to asic_type enum
Likun Gao [Tue, 10 Jul 2018 12:10:05 +0000 (20:10 +0800)]
drm/amdgpu: add picasso to asic_type enum

Add picasso to amd_asic_type enum and amdgpu_asic_name[].

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: use a single linked list for amdgpu_vm_bo_base
Christian König [Mon, 10 Sep 2018 18:02:46 +0000 (20:02 +0200)]
drm/amdgpu: use a single linked list for amdgpu_vm_bo_base

Instead of the double linked list. Gets the size of amdgpu_vm_pt down to
64 bytes again.

We could even reduce it down to 32 bytes, but that would require some
rather extreme hacks.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: remove amdgpu_bo_list_entry.robj (v2)
Christian König [Mon, 10 Sep 2018 14:07:57 +0000 (16:07 +0200)]
drm/amdgpu: remove amdgpu_bo_list_entry.robj (v2)

We can get that just by casting tv.bo.

v2: squash in kfd fix (Alex)

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/ttm: once more fix ttm_bo_bulk_move_lru_tail
Christian König [Wed, 12 Sep 2018 19:19:57 +0000 (21:19 +0200)]
drm/ttm: once more fix ttm_bo_bulk_move_lru_tail

While cutting the lists we sometimes accidentally added a list_head from
the stack to the LRUs, effectively corrupting the list.

Remove the list cutting and use explicit list manipulation instead.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-and-Tested: Huang Rui <ray.huang@amd.com>
Tested-by: Mike Lothian <mike@fireburn.co.uk>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display: Fix pflip IRQ status after gpu reset.
Andrey Grodzovsky [Wed, 12 Sep 2018 20:38:57 +0000 (16:38 -0400)]
drm/amd/display: Fix pflip IRQ status after gpu reset.

Problem:
After GPU reset pflip completion IRQ is disabled and hence
any subsequent mode set or plane update leads to hang.

Fix:
Unless acrtc->otg_inst is initialized to -1 during display
block initializtion then durng resume from GPU reset
amdgpu_irq_gpu_reset_resume_helper will override CRTC 0 pflip
IRQ value with whatever value was on every other unused CRTC because
dm_irq_state will do irq_source = dal_irq_type + acrtc->otg_inst
where acrtc->otg_inst will be 0 for every unused CRTC.

Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: allow fragment processing for invalid PTEs
Christian König [Fri, 7 Sep 2018 18:34:17 +0000 (20:34 +0200)]
drm/amdgpu: allow fragment processing for invalid PTEs

That should improve the PRT performance on Vega quite a bit.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Acked-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: use the maximum possible fragment size on Vega/Raven
Christian König [Fri, 7 Sep 2018 12:21:15 +0000 (14:21 +0200)]
drm/amdgpu: use the maximum possible fragment size on Vega/Raven

The fragment size controls only the L1 on Vega/Raven and we now don't
have any extra overhead any more because of larger fragments.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: meld together VM fragment and huge page handling
Christian König [Thu, 6 Sep 2018 15:13:06 +0000 (17:13 +0200)]
drm/amdgpu: meld together VM fragment and huge page handling

This optimizes the generating of PTEs by walking the hierarchy only once
for a range and making changes as necessary.

It allows for both huge (2MB) as well giant (1GB) pages to be used on
Vega and Raven.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Acked-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: use leaf iterator for filling PTs
Christian König [Mon, 3 Sep 2018 12:34:51 +0000 (14:34 +0200)]
drm/amdgpu: use leaf iterator for filling PTs

Less overhead and is the starting point for further cleanups and
improvements.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: use the DFS iterator in amdgpu_vm_invalidate_pds v2
Christian König [Sat, 8 Sep 2018 11:05:34 +0000 (13:05 +0200)]
drm/amdgpu: use the DFS iterator in amdgpu_vm_invalidate_pds v2

Less code and easier to maintain.

v2: rename the function as well

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: use dfs iterator to free PDs/PTs
Christian König [Thu, 6 Sep 2018 13:35:13 +0000 (15:35 +0200)]
drm/amdgpu: use dfs iterator to free PDs/PTs

Allows us to free all PDs/PTs without recursion.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: use leaf iterator for allocating PD/PT
Christian König [Sat, 1 Sep 2018 10:03:37 +0000 (12:03 +0200)]
drm/amdgpu: use leaf iterator for allocating PD/PT

Less code and allows for easier error handling.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add some VM PD/PT iterators v2
Christian König [Sat, 1 Sep 2018 08:36:48 +0000 (10:36 +0200)]
drm/amdgpu: add some VM PD/PT iterators v2

Both a leaf as well as dfs iterator to walk over all the PDs/PTs.

v2: update comments and fix for_each_amdgpu_vm_pt_dfs_safe

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu:Add error message when register failed to reach expected value
James Zhu [Mon, 10 Sep 2018 16:53:25 +0000 (12:53 -0400)]
drm/amdgpu:Add error message when register failed to reach expected value

Add error message when register failed to reach expected value, It will
help discover potential issue.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd/display: fix ptr_ret.cocci warnings
kbuild test robot [Wed, 12 Sep 2018 00:59:07 +0000 (08:59 +0800)]
drm/amd/display: fix ptr_ret.cocci warnings

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c:771:1-3: WARNING: PTR_ERR_OR_ZERO can be used

 Use PTR_ERR_OR_ZERO rather than if(IS_ERR(...)) + PTR_ERR

Generated by: scripts/coccinelle/api/ptr_ret.cocci

Fixes: e498eb713604 ("drm/amd/display: Add support for hw_state logging via debugfs")
CC: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd: Add DM DMCU support
David Francis [Tue, 11 Sep 2018 17:49:49 +0000 (13:49 -0400)]
drm/amd: Add DM DMCU support

DMCU (Display Microcontroller Unit) is a GPU chip involved in
eDP features like Adaptive Backlight Modulation and Panel Self
Refresh.

DC is already fully equipped to initialize DMCU as long as the
firmware is loaded.

At the moment only the raven firmware is available.

A single .bin file is loaded by the kernel's loading mechanism
and split into two ucodes according to the header.

DMCU is optional, so if the firmware is not found, no error or
warning is raised.

Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd: Add PSP DMCU support
David Francis [Tue, 11 Sep 2018 17:46:41 +0000 (13:46 -0400)]
drm/amd: Add PSP DMCU support

DMCU (Display Microcontroller Unit) is a GPU chip involved in
eDP features like Adaptive Backlight Modulation and Panel Self
Refresh.

PSP is already equipped to handle DMCU firmware loading, all
that is needed is to translate between the new DMCU ucode ID and
the equivalent psp_gfx_fw_type.

Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amd: Add ucode DMCU support
David Francis [Tue, 11 Sep 2018 17:41:01 +0000 (13:41 -0400)]
drm/amd: Add ucode DMCU support

DMCU (Display Microcontroller Unit) is a GPU chip involved in
eDP features like Adaptive Backlight Modulation and Panel Self
Refresh.

DMCU has two pieces of firmware: the ERAM and the interrupt
vectors, which must be loaded seperately.

To this end, the DMCU firmware has a custom header and parsing
logic similar to MEC, to extract the two ucodes from a single
struct firmware.

Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: Move fault hash table to amdgpu vm
Oak Zeng [Thu, 6 Sep 2018 03:51:23 +0000 (23:51 -0400)]
drm/amdgpu: Move fault hash table to amdgpu vm

In stead of share one fault hash table per device, make it
per vm. This can avoid inter-process lock issue when fault
hash table is full.

Change-Id: I5d1281b7c41eddc8e26113e010516557588d3708
Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
Suggested-by: Christian Konig <Christian.Koenig@amd.com>
Suggested-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian Konig <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: move cs dependencies front a bit
Chunming Zhou [Tue, 11 Sep 2018 09:22:40 +0000 (17:22 +0800)]
drm/amdgpu: move cs dependencies front a bit

cs dependencies handling doesn't need in vm resv

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>
6 years agodrm/amdgpu: try allocating VRAM as power of two
Christian König [Fri, 7 Sep 2018 12:27:05 +0000 (14:27 +0200)]
drm/amdgpu: try allocating VRAM as power of two

Try to allocate VRAM in power of two sizes and only fallback to vram
split sizes if that fails.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: Fix SDMA TO after GPU reset v3
Andrey Grodzovsky [Mon, 10 Sep 2018 22:43:58 +0000 (18:43 -0400)]
drm/amdgpu: Fix SDMA TO after GPU reset v3

After GPU reset amdgpu_vm_clear_bo triggers VM flush
but job->vm_pd_addr is not set causing SDMA TO.

v2:
Per advise by Christian König avoid flushing VM for jobs where
job->vm_pd_addr wasn't explicitly set.

v3:
Shortcut vm_flush_needed early.

Fixes cbd5285 drm/amdgpu: move setting the GART addr into TTM.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add amdgpu_vm_update_func
Christian König [Thu, 30 Aug 2018 13:55:54 +0000 (15:55 +0200)]
drm/amdgpu: add amdgpu_vm_update_func

Add helper to call the update function for both BO and shadow.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: add amdgpu_vm_pt_parent helper
Christian König [Thu, 30 Aug 2018 13:55:54 +0000 (15:55 +0200)]
drm/amdgpu: add amdgpu_vm_pt_parent helper

Add a function to get the parent of a PD/PT.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdgpu: fix error handling in amdgpu_cs_user_fence_chunk
Christian König [Mon, 10 Sep 2018 13:52:55 +0000 (15:52 +0200)]
drm/amdgpu: fix error handling in amdgpu_cs_user_fence_chunk

Slowly leaking memory one page at a time :)

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdkfd: Only add bi-directional iolink on GPU with XGMI or largebar (v2)
shaoyunl [Fri, 7 Sep 2018 16:00:07 +0000 (12:00 -0400)]
drm/amdkfd: Only add bi-directional iolink on GPU with XGMI or largebar (v2)

v2: compile fix

Signed-off-by: shaoyunl <Shaoyun.Liu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdkfd: Generate xGMI direct iolink
Shaoyun Liu [Mon, 13 Aug 2018 18:04:11 +0000 (14:04 -0400)]
drm/amdkfd: Generate xGMI direct iolink

Generate xGMI iolink for upper level usage

Signed-off-by: Shaoyun Liu <Shaoyun.Liu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
6 years agodrm/amdkfd: Add new iolink type defines
Shaoyun Liu [Mon, 13 Aug 2018 18:02:17 +0000 (14:02 -0400)]
drm/amdkfd: Add new iolink type defines

Update the iolink type defines according to the new thunk spec

Signed-off-by: Shaoyun Liu <Shaoyun.Liu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>