openwrt/staging/blogic.git
8 years agodrm/radeon: skip suspend/resume on DRM_SWITCH_POWER_DYNAMIC_OFF
Alex Deucher [Wed, 31 Aug 2016 21:31:30 +0000 (17:31 -0400)]
drm/radeon: skip suspend/resume on DRM_SWITCH_POWER_DYNAMIC_OFF

Missed this case previously. No need to do anything if the
device is already off.  runtime pm will handle it.

Acked-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/radeon: set runtime pm state to active on resume
Alex Deucher [Wed, 31 Aug 2016 21:34:23 +0000 (17:34 -0400)]
drm/radeon: set runtime pm state to active on resume

The sbios always powers up the dGPU on resume.

Acked-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: skip suspend/resume on DRM_SWITCH_POWER_DYNAMIC_OFF
Alex Deucher [Sat, 27 Aug 2016 16:57:20 +0000 (12:57 -0400)]
drm/amdgpu: skip suspend/resume on DRM_SWITCH_POWER_DYNAMIC_OFF

Missed this case previously. No need to do anything if the
device is already off.  runtime pm will handle it.

Acked-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: set runtime pm state to active on resume
Alex Deucher [Sat, 27 Aug 2016 18:53:08 +0000 (14:53 -0400)]
drm/amdgpu: set runtime pm state to active on resume

The sbios always powers up the dGPU on resume.

Acked-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu/gfx6: drop some dead code
Alex Deucher [Thu, 1 Sep 2016 18:10:46 +0000 (14:10 -0400)]
drm/amdgpu/gfx6: drop some dead code

The mqd is only used on CI and newer.

Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add gart recovery by gtt list V2
Chunming Zhou [Tue, 30 Aug 2016 08:36:25 +0000 (16:36 +0800)]
drm/amdgpu: add gart recovery by gtt list V2

V2:
a. gart recovery should be ahead of ring test.
b. rename to amdgpu_ttm_recover_gart

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>
8 years agodrm/amdgpu: link all gtt when binding them V2
Chunming Zhou [Tue, 30 Aug 2016 08:13:10 +0000 (16:13 +0800)]
drm/amdgpu: link all gtt when binding them V2

V2:
spin lock instead of mutex for gtt list

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>
8 years agodrm/amdgpu: add si pciids v2
Ken Wang [Thu, 21 Jan 2016 09:33:00 +0000 (17:33 +0800)]
drm/amdgpu: add si pciids v2

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu/si: Add updated smc firmware for SI kickers
Alex Deucher [Mon, 1 Aug 2016 20:05:47 +0000 (16:05 -0400)]
drm/amdgpu/si: Add updated smc firmware for SI kickers

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: Add SI Family information
Ken Wang [Tue, 24 May 2016 13:02:53 +0000 (21:02 +0800)]
drm/amdgpu: Add SI Family information

Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: enable SI DPM
Maruthi Srinivas Bayyavarapu [Tue, 26 Apr 2016 15:05:36 +0000 (20:35 +0530)]
drm/amdgpu: enable SI DPM

Signed-off-by: Maruthi Bayyavarapu <maruthi.bayyavarapu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: use DRM print instead of printk
Huang Rui [Wed, 31 Aug 2016 05:23:18 +0000 (13:23 +0800)]
drm/amdgpu: use DRM print instead of printk

Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add SI DPM support (v4)
Maruthi Bayyavarapu [Mon, 1 Aug 2016 16:42:32 +0000 (12:42 -0400)]
drm/amdgpu: add SI DPM support (v4)

v2: corrected register offset shift
v3: rebase fixes
v4: fix firmware paths
    add SI smc firmware versions for sysfs dump
    remove unused function forward define
    fix the tahiti specific value of DEEP_SLEEP_CLK_SEL field
    fix to miss adding thermal controller
    use vram_type instead of checking mem_gddr5 flag
    fix incorrect index of CG_FFCT_0 register
    fix incorrect reading method at si_get_current_pcie_speed

Signed-off-by: Maruthi Bayyavarapu <maruthi.bayyavarapu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add SI SMC support
Maruthi Srinivas Bayyavarapu [Tue, 26 Apr 2016 15:01:54 +0000 (20:31 +0530)]
drm/amdgpu: add SI SMC support

Signed-off-by: Maruthi Bayyavarapu <maruthi.bayyavarapu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add si dpm support in amdgpu_atombios
Maruthi Srinivas Bayyavarapu [Tue, 26 Apr 2016 14:54:38 +0000 (20:24 +0530)]
drm/amdgpu: add si dpm support in amdgpu_atombios

v2: renamed _atom_ to _atombios_ for consistency
    added ulClockParams to _COMPUTE_MEMORY_ENGINE_PLL_PARAMETERS_V3 and
    _COMPUTE_MEMORY_ENGINE_PLL_PARAMETERS_V5 to avoid build break

Signed-off-by: Maruthi Bayyavarapu <maruthi.bayyavarapu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add si specific logic into the device initialize function v3
Ken Wang [Thu, 21 Jan 2016 11:08:55 +0000 (19:08 +0800)]
drm/amdgpu: add si specific logic into the device initialize function v3

v3: guard doorbell_fini as well

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add si ip blocks setup v3
Ken Wang [Thu, 21 Jan 2016 09:29:41 +0000 (17:29 +0800)]
drm/amdgpu: add si ip blocks setup v3

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add all the components for si into Makefile/kconfig v3
Ken Wang [Tue, 19 Jan 2016 06:19:53 +0000 (14:19 +0800)]
drm/amdgpu: add all the components for si into Makefile/kconfig v3

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: introduce pcie port read/write entry
Huang Rui [Wed, 31 Aug 2016 05:23:25 +0000 (13:23 +0800)]
drm/amdgpu: introduce pcie port read/write entry

This patch adds pcie port read/write entry, because it will be also
used on si dpm part.

Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add si implementation v10
Ken Wang [Tue, 19 Jan 2016 06:08:49 +0000 (14:08 +0800)]
drm/amdgpu: add si implementation v10

v5: rebase fixes
v6: add mgcg arrays
v7: rebase fixes
v8: rebase fixes
v9: add get_disabled_bios(), make get_xclk static
v10: fix oland and hainan asic specific handle at si_program_aspm

Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add DMA implementation for si v8
Ken Wang [Tue, 19 Jan 2016 06:05:23 +0000 (14:05 +0800)]
drm/amdgpu: add DMA implementation for si v8

v4: rebase fixes
v5: use the generic nop fill
v6: rebase fixes
v7: rebase fixes
    copy count fixes from Jonathan
    general cleanup
    add fill buffer implementation
v8: adapt write_pte and copy_pte to latest changes

Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add graphic pipeline implementation for si v8
Ken Wang [Tue, 19 Jan 2016 06:04:28 +0000 (14:04 +0800)]
drm/amdgpu: add graphic pipeline implementation for si v8

v5: rebase fixes
v6: rebase fixes
v7: rebase fixes
    fix tile reg offset as noticed by Jonathan
    Drop some debugging remnants
v8: add gfx v6 firmware versions for sysfs dump

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: atombios change for dce6 to work v3
Ken Wang [Tue, 19 Jan 2016 07:07:29 +0000 (15:07 +0800)]
drm/amdgpu: atombios change for dce6 to work v3

v3: white space fixes

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add display controller implementation for si v10
Ken Wang [Tue, 19 Jan 2016 06:03:24 +0000 (14:03 +0800)]
drm/amdgpu: add display controller implementation for si v10

v4: rebase fixups
v5: more fixes based on dce8 code
v6: squash in dmif offset fix
v7: rebase fixups
v8: rebase fixups, drop some debugging remnants
v9: fix BE build
v10: include Marek's tiling fixes, add support for
     page_flip_target, set MASTER_UDPATE_MODE=0,
     fix cursor

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add interupt handler implementation for si v3
Ken Wang [Tue, 19 Jan 2016 06:02:14 +0000 (14:02 +0800)]
drm/amdgpu: add interupt handler implementation for si v3

v3: rebase fixups

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add graphic memory controller implementation for si v7
Ken Wang [Tue, 19 Jan 2016 06:01:19 +0000 (14:01 +0800)]
drm/amdgpu: add graphic memory controller implementation for si v7

v4: rebase fixups
v5: rebase fixups
v5: rebase fixups
v6: rebase fixups for gart size changes
v7: add gmc v6 firmware versions for sysfs dump

Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add si header files v4
Ken Wang [Tue, 19 Jan 2016 05:53:10 +0000 (13:53 +0800)]
drm/amdgpu: add si header files v4

v4: drop unused DCE6 macro

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add SI asics types v2
Ken Wang [Thu, 21 Jan 2016 09:00:06 +0000 (17:00 +0800)]
drm/amdgpu: add SI asics types v2

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Ken Wang <Qingqing.Wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu:add switch buffer to end of CS (v2)
Monk Liu [Fri, 26 Aug 2016 06:12:37 +0000 (14:12 +0800)]
drm/amdgpu:add switch buffer to end of CS (v2)

sync switch buffer scheme with windows kmd for gfx v8,
step1: append a switch_buffer to the end of CS

v2:rebase on latest staging

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/amdgpu: debugfs SMC addresses are byte addresses
Tom St Denis [Mon, 29 Aug 2016 12:39:29 +0000 (08:39 -0400)]
drm/amd/amdgpu: debugfs SMC addresses are byte addresses

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>
8 years agodrm/amd/powerplay: Remove unused variable causing compile warning
Jordan Lazare [Fri, 26 Aug 2016 21:10:28 +0000 (17:10 -0400)]
drm/amd/powerplay: Remove unused variable causing compile warning

If treating warnings as errors this causes a build error

Signed-off-by: Jordan Lazare <Jordan.Lazare@amd.com>
Reviewed-by: Alex Deucher <Alexander.Deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/powerplay: Only load SDMA0/MEC firmware once on Stoney (v2)
Tom St Denis [Fri, 26 Aug 2016 16:44:01 +0000 (12:44 -0400)]
drm/amd/powerplay: Only load SDMA0/MEC firmware once on Stoney (v2)

Only load the SDMA0/MEC1 firmware once in the Carrizo SMU manager
driver.

(v2) Avoid loading SDMA0 twice too.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/powerplay: Fix up return codes in cz SMU manager
Tom St Denis [Fri, 26 Aug 2016 16:43:14 +0000 (12:43 -0400)]
drm/amd/powerplay: Fix up return codes in cz SMU manager

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/powerplay: Tidy up cz SMU manager
Tom St Denis [Fri, 26 Aug 2016 16:42:36 +0000 (12:42 -0400)]
drm/amd/powerplay: Tidy up cz SMU manager

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/powerplay: Fix CZ SMU firmware load check (v4)
Tom St Denis [Fri, 26 Aug 2016 14:29:19 +0000 (10:29 -0400)]
drm/amd/powerplay:  Fix CZ SMU firmware load check (v4)

Remove an errant return in the middle of the check
function as well as check for success in the start
function.

(v2) Add return check to smu_load_fw()
(v3) Don't return early if SMU load check fails
(v4) No returns!  :-)

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: throttle buffer migrations at CS using a fixed MBps limit (v2)
Marek Olšák [Wed, 17 Aug 2016 21:49:27 +0000 (23:49 +0200)]
drm/amdgpu: throttle buffer migrations at CS using a fixed MBps limit (v2)

The old mechanism used a per-submission limit that didn't take previous
submissions within the same time frame into account. It also filled VRAM
slowly when VRAM usage dropped due to a big eviction or buffer deallocation.

This new method establishes a configurable MBps limit that is obeyed when
VRAM usage is very high. When VRAM usage is not very high, it gives
the driver the freedom to fill it quickly. The result is more consistent
performance.

It can't keep the BO move rate low if lots of evictions are happening due
to VRAM fragmentation, or if a big buffer is being migrated.

The amdgpu.moverate parameter can be used to set a non-default limit.
Measurements can be done to find out which amdgpu.moverate setting gives
the best results.

Mainly APUs and cards with small VRAM will benefit from this. For F1 2015,
anything with 2 GB VRAM or less will benefit.

Some benchmark results - F1 2015 (Tonga 2GB):

Limit      MinFPS AvgFPS
Old code:  14     32.6
128 MB/s:  28     41
64 MB/s:   15.5   43
32 MB/s:   28.7   43.4
8 MB/s:    27.8   44.4
8 MB/s:    21.9   42.8 (different run)

Random drops in Min FPS can still occur (due to fragmented VRAM?), but
the average FPS is much better. 8 MB/s is probably a good limit for this
game & the current VRAM management. The random FPS drops are still to be
tackled.

v2: use a spinlock

Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/amdgpu: Tidy up cz_dpm.c
Tom St Denis [Thu, 25 Aug 2016 17:02:06 +0000 (13:02 -0400)]
drm/amd/amdgpu: Tidy up cz_dpm.c

Various minor formatting changes.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/amdgpu: Clean up memory leak in cz_dpm_init().
Tom St Denis [Thu, 25 Aug 2016 16:41:15 +0000 (12:41 -0400)]
drm/amd/amdgpu: Clean up memory leak in cz_dpm_init().

If init fails free up any allocated memory.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/amdgpu: Fix memleak in cz_parse_power_table()
Tom St Denis [Thu, 25 Aug 2016 16:16:24 +0000 (12:16 -0400)]
drm/amd/amdgpu: Fix memleak in cz_parse_power_table()

If one of the entries fails to be allocated then free
all of the previous entries before freeing the array which
holds their pointers.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: refine uvd gate logic for CI.
Rex Zhu [Wed, 24 Aug 2016 11:39:06 +0000 (19:39 +0800)]
drm/amdgpu: refine uvd gate logic for CI.

uvd dpm will be controlled by uvd.
dpm just disable uvd dpm in case of suspend when play video.
due to the new logic of uvd_begin_use/end_use,
if disable uvd dpm in late init, will have no chance to
enable uvd dpm after resume until play video again.

Change-Id: I70e3d7efe63edad37f26e6c5ea089da1135c0ab5
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/powerplay: simplify struct amd_pp_init.
Rex Zhu [Mon, 22 Aug 2016 12:47:28 +0000 (20:47 +0800)]
drm/amd/powerplay: simplify struct amd_pp_init.

delete the members not needed when amd_powerplay_init.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/powerplay: get system info by cgs interface.
Rex Zhu [Mon, 22 Aug 2016 12:47:01 +0000 (20:47 +0800)]
drm/amd/powerplay: get system info by cgs interface.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add support for getting sub_device id and
Rex Zhu [Mon, 22 Aug 2016 12:48:13 +0000 (20:48 +0800)]
drm/amdgpu: add support for getting sub_device id and
sub_vendor_id in cgs interface.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add vce bypass mode for tonga.
Rex Zhu [Thu, 18 Aug 2016 10:27:14 +0000 (18:27 +0800)]
drm/amdgpu: add vce bypass mode for tonga.

fix issue that encode test failed on the second time when
vce dpm enabled on tonga.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/powerplay: add bypass mode for vce 2.0.
Rex Zhu [Thu, 18 Aug 2016 05:02:46 +0000 (13:02 +0800)]
drm/amd/powerplay: add bypass mode for vce 2.0.

fix issue after vce encode, the eclk stay high.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: refine function name for consistency
Rex Zhu [Thu, 18 Aug 2016 04:37:02 +0000 (12:37 +0800)]
drm/amdgpu: refine function name for consistency

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu/vce3: add support for third vce ring
Alex Deucher [Wed, 24 Aug 2016 21:15:33 +0000 (17:15 -0400)]
drm/amdgpu/vce3: add support for third vce ring

Not of much use at the moment (we don't really use
the second ring either), but may be useful later.

Reviewed-by: JimQu <Jim.Qu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: track the number of vce rings
Alex Deucher [Wed, 24 Aug 2016 20:56:21 +0000 (16:56 -0400)]
drm/amdgpu: track the number of vce rings

Rather than using a hardcoded value.  This allows
different versions to expose more or less rings.

No functional change.

Reviewed-by: JimQu <Jim.Qu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: rename suspend_kms and resume_kms
Alex Deucher [Tue, 23 Aug 2016 17:25:49 +0000 (13:25 -0400)]
drm/amdgpu: rename suspend_kms and resume_kms

The old names were dragged over from radeon.  The new ones
better match the naming conventions used in the driver.

No functional change.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: use memcpy_toio for VCE firmware upload
Christian König [Tue, 23 Aug 2016 09:18:59 +0000 (11:18 +0200)]
drm/amdgpu: use memcpy_toio for VCE firmware upload

Try to be clean here, even when it's a noop on x86.

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>
8 years agodrm/amdgpu: use memcpy_to/fromio for UVD fw upload
Christian König [Tue, 23 Aug 2016 09:00:17 +0000 (11:00 +0200)]
drm/amdgpu: use memcpy_to/fromio for UVD fw upload

Also use the firmware size from the header instead of calculating it.

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>
8 years agodrm/amd/powerplay: delete useless code in iceland_hwmgr.c.
Rex Zhu [Mon, 22 Aug 2016 12:49:36 +0000 (20:49 +0800)]
drm/amd/powerplay: delete useless code in iceland_hwmgr.c.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/radeon: switch UVD code to use UVD_NO_OP for padding
Alex Deucher [Tue, 23 Aug 2016 14:07:28 +0000 (10:07 -0400)]
drm/radeon: switch UVD code to use UVD_NO_OP for padding

Replace packet2's with packet0 writes to UVD_NO_OP.  The
value written to UVD_NO_OP does not matter.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: switch UVD code to use UVD_NO_OP for padding
Alex Deucher [Tue, 23 Aug 2016 13:12:21 +0000 (09:12 -0400)]
drm/amdgpu: switch UVD code to use UVD_NO_OP for padding

Replace packet2's with packet0 writes to UVD_NO_OP.  The
value written to UVD_NO_OP does not matter.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/radeon: add support for UVD_NO_OP register
Alex Deucher [Mon, 22 Aug 2016 22:03:22 +0000 (18:03 -0400)]
drm/radeon: add support for UVD_NO_OP register

Writes to this register are the preferred way to do NOPs.

Bump the driver version as well.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add support for UVD_NO_OP register
Alex Deucher [Mon, 22 Aug 2016 21:58:14 +0000 (17:58 -0400)]
drm/amdgpu: add support for UVD_NO_OP register

Writes to this register are the preferred way to do NOPs.

Bump the driver version as well.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: fix VCE ib alignment value
Alex Deucher [Tue, 23 Aug 2016 14:44:16 +0000 (10:44 -0400)]
drm/amdgpu: fix VCE ib alignment value

The VCE rings only require single dword alignment.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: fix IB alignment for UVD
Alex Deucher [Mon, 22 Aug 2016 20:31:36 +0000 (16:31 -0400)]
drm/amdgpu: fix IB alignment for UVD

According to the hw team, it should be 16, not 8.

Cc: Peter Fang <peter.fang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
8 years agodrm/amd/amdgpu: Print ring name in amdgpu_ib_schedule()
Tom St Denis [Mon, 22 Aug 2016 14:54:28 +0000 (10:54 -0400)]
drm/amd/amdgpu: Print ring name in amdgpu_ib_schedule()

If the ring isn't ready lets print out which ring name
to help debugging.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/radeon: remove dead code, si_mc_load_microcode (v2)
Heinrich Schuchardt [Sun, 21 Aug 2016 20:36:29 +0000 (22:36 +0200)]
drm/radeon: remove dead code, si_mc_load_microcode (v2)

In an if block for (running == 0) running cannot be non-zero.

v2: agd: remove unused variable

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/radeon/cik: remove dead code (v2)
Heinrich Schuchardt [Sun, 21 Aug 2016 19:41:49 +0000 (21:41 +0200)]
drm/radeon/cik: remove dead code (v2)

In an if block for (running == 0) running cannot be non-zero.

v2: agd: remove unused variable

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/powerplay: avoid NULL dereference, cz_hwmgr.c
Heinrich Schuchardt [Sun, 21 Aug 2016 18:27:02 +0000 (20:27 +0200)]
drm/amd/powerplay: avoid NULL dereference, cz_hwmgr.c

if (a == NULL || a->b == NULL)
leads to a NULL pointer dereference if a == NULL.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/powerplay: avoid NULL pointer dereference
Heinrich Schuchardt [Sun, 21 Aug 2016 18:21:27 +0000 (20:21 +0200)]
drm/amd/powerplay: avoid NULL pointer dereference

if (a == NULL || a->b == NULL)
leads to a NULL pointer dereference if a == NULL.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu/gmc8: remove dead code (v2)
Heinrich Schuchardt [Sun, 21 Aug 2016 18:10:41 +0000 (20:10 +0200)]
drm/amdgpu/gmc8: remove dead code (v2)

In an if block for (running == 0) running cannot be non-zero.

v2: agd: remove unused variable

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu/gmc7: remove dead code (v2)
Heinrich Schuchardt [Sun, 21 Aug 2016 18:06:32 +0000 (20:06 +0200)]
drm/amdgpu/gmc7: remove dead code (v2)

In an if block for (running == 0) running cannot be non-zero.

v2: agd: remove unused variable

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: Fix indentation in dce_v8_0_audio_write_sad_regs()
Alexandre Demers [Mon, 22 Aug 2016 00:38:26 +0000 (20:38 -0400)]
drm/amdgpu: Fix indentation in dce_v8_0_audio_write_sad_regs()

Fixed indentation for readability.

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Signed-off-by: Alexandre Demers <alexandre.f.demers@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: Use correct mask in dce_v8_0_afmt_setmode() and fix comment typos.
Alexandre Demers [Mon, 22 Aug 2016 00:38:41 +0000 (20:38 -0400)]
drm/amdgpu: Use correct mask in dce_v8_0_afmt_setmode() and fix comment typos.

We were using the same mask twice. Looking at radeon, it seems
we should be using HDMI_AVI_INFO_CONT instead as the second mask.

Being there, fix typos in comments and improved readability.

I haven't looked at other DCEs, the mask may also be wrong for them.

Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alexandre Demers <alexandre.f.demers@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: cleanup amdgpu_vm_bo_update params
Christian König [Tue, 16 Aug 2016 12:43:17 +0000 (14:43 +0200)]
drm/amdgpu: cleanup amdgpu_vm_bo_update params

Make it more obvious what we are doing 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>
8 years agodrm/amdgpu: stop adding dummy entry in amdgpu_ttm_placement_init
Christian König [Mon, 15 Aug 2016 12:08:54 +0000 (14:08 +0200)]
drm/amdgpu: stop adding dummy entry in amdgpu_ttm_placement_init

AMDGPU_GEM_CREATE_NO_CPU_ACCESS and AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED are
obviously mutual exclusive. So stop adding a dummy entry without effect when
both are specified.

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>
8 years agodrm/amdgpu: cleanup amdgpu_ttm_placement_init
Christian König [Mon, 15 Aug 2016 12:06:50 +0000 (14:06 +0200)]
drm/amdgpu: cleanup amdgpu_ttm_placement_init

Make it more clear what this function does. No intendet functional change.

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>
8 years agodrm/amdgpu: fix lru size grouping v2
Christian König [Wed, 17 Aug 2016 11:44:20 +0000 (13:44 +0200)]
drm/amdgpu: fix lru size grouping v2

Adding a BO can make it the insertion point for larger sizes as well.

v2: add a comment about the guard structure.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/powerplay: Tidy up cz_hwmgr.c
Tom St Denis [Thu, 18 Aug 2016 14:25:52 +0000 (10:25 -0400)]
drm/amd/powerplay:  Tidy up cz_hwmgr.c

Clean up whitespace and formatting.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add AMDGPU_INFO_NUM_EVICTIONS
Marek Olšák [Wed, 17 Aug 2016 21:58:58 +0000 (23:58 +0200)]
drm/amdgpu: add AMDGPU_INFO_NUM_EVICTIONS

For profiling.

v2: really bump the minor version

Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: recover vram bo from shadow after gpu reset V2
Chunming Zhou [Thu, 21 Jul 2016 09:20:52 +0000 (17:20 +0800)]
drm/amdgpu: recover vram bo from shadow after gpu reset V2

V2:
1. don't directly submit to many jobs at the same time.
2. delete unrelated printk.

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>
8 years agodrm/amdgpu: link all shadow bo V2
Chunming Zhou [Wed, 17 Aug 2016 03:41:30 +0000 (11:41 +0800)]
drm/amdgpu: link all shadow bo V2

V2:
1. use mutex instead of spinlock for shadow list, since its process could
sleep.
2. move list_del to bo destroy phase.

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>
8 years agodrm/amdgpu: update pt shadow while updating pt V2
Chunming Zhou [Mon, 15 Aug 2016 03:46:21 +0000 (11:46 +0800)]
drm/amdgpu: update pt shadow while updating pt V2

V2:
move shadow parameter to amdgpu_pte_update_params.

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Cc: minutemaidpark@hotmail.com
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: update pd shadow while updating pd V2
Chunming Zhou [Mon, 15 Aug 2016 03:36:54 +0000 (11:36 +0800)]
drm/amdgpu: update pd shadow while updating pd V2

V2:
Checking if shadow is valid.

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>
8 years agodrm/amdgpu: sync bo and shadow V3
Chunming Zhou [Thu, 4 Aug 2016 08:51:18 +0000 (16:51 +0800)]
drm/amdgpu: sync bo and shadow V3

Use shadow flag to judge which direction to sync.
V2:
Don't need bo pin, so remove it.

V3:
1. Split to two functions, one is backup_to_shadow, another is
restore_from_shadow.
2. Clean up previous shadow direction difinitions.

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>
8 years agodrm/amdgpu: add direct submision option for copy_buffer
Chunming Zhou [Mon, 15 Aug 2016 02:46:04 +0000 (10:46 +0800)]
drm/amdgpu: add direct submision option for copy_buffer

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>
8 years agodrm/amdgpu: add need backup function V2
Chunming Zhou [Fri, 5 Aug 2016 09:30:17 +0000 (17:30 +0800)]
drm/amdgpu: add need backup function V2

V2:
add checking if need backup in amdgpu_bo_create.

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>
8 years agodrm/amd/amdgpu: S3 resumed failed after 4-5 times loop
jimqu [Fri, 19 Aug 2016 01:31:50 +0000 (09:31 +0800)]
drm/amd/amdgpu: S3 resumed failed after 4-5 times loop

Phenomenon: software hang when device resume back, read UVD fence is 0xffffffff
and read pcie pid is 0xffff.
The issue is caused by VCE reset when update cg setting. according to HW programming
guide, adjust update VCE cg sequence.
The patch apply to VCE2.0.

Signed-off-by: JimQu <Jim.Qu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: use domain's gpu_offset for start addr
Flora Cui [Thu, 18 Aug 2016 05:18:09 +0000 (13:18 +0800)]
drm/amdgpu: use domain's gpu_offset for start addr

Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Ken Wang <Qingqing.Wang@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: update gart_pin_size only if the bo is pined to GTT
Flora Cui [Thu, 18 Aug 2016 05:17:07 +0000 (13:17 +0800)]
drm/amdgpu: update gart_pin_size only if the bo is pined to GTT

Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: check domain sanity in amdgpu_bo_pin_restricted()
Flora Cui [Thu, 18 Aug 2016 04:55:13 +0000 (12:55 +0800)]
drm/amdgpu: check domain sanity in amdgpu_bo_pin_restricted()

abort if the bo is pined to other domain already

Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agoamdgpu: move ttm stuff to amdgpu_ttm.h
Flora Cui [Tue, 2 Aug 2016 03:32:41 +0000 (11:32 +0800)]
amdgpu: move ttm stuff to amdgpu_ttm.h

Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/powerplay: enable power containment features for tonga.
Rex Zhu [Wed, 17 Aug 2016 07:48:30 +0000 (15:48 +0800)]
drm/amd/powerplay: enable power containment features for tonga.

v2: fix build error introduced when fix code style problems.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/amdgpu: Add more debugfs config data
Tom St Denis [Wed, 17 Aug 2016 16:00:51 +0000 (12:00 -0400)]
drm/amd/amdgpu:  Add more debugfs config data

Adds family and external_rev_id to config data

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: Hardcode virtual DCE vblank / scanout position return values
Emily Deng [Wed, 17 Aug 2016 06:59:20 +0000 (14:59 +0800)]
drm/amdgpu: Hardcode virtual DCE vblank / scanout position return values

For virtual display feature, by hardcoding 0 for the vblank counter and
-EINVAL for the scanout position return value, we signal to the core DRM code that there are
no hardware counters we can use for these.

Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amd/amdgpu: Add more config data for debugfs
Tom St Denis [Fri, 12 Aug 2016 19:14:31 +0000 (15:14 -0400)]
drm/amd/amdgpu: Add more config data for debugfs

Adds rev_id as well as cg/pg flags to help debug runtime.

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: fix coding style in amdgpu_object.c
Christian König [Fri, 12 Aug 2016 14:50:12 +0000 (16:50 +0200)]
drm/amdgpu: fix coding style in amdgpu_object.c

Just a few 80 chars problems.

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>
8 years agodrm/amdgpu: add function pointer to the pte_update_params
Christian König [Fri, 12 Aug 2016 11:29:18 +0000 (13:29 +0200)]
drm/amdgpu: add function pointer to the pte_update_params

Remember what function to call while planning the commands instead
of figuring it our later on.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: stop splitting PTE commands into smaller ones
Christian König [Fri, 12 Aug 2016 10:59:59 +0000 (12:59 +0200)]
drm/amdgpu: stop splitting PTE commands into smaller ones

It doesn't make much sense to create bigger commands first which we then need
to split into smaller one again. Just make sure the commands we create aren't
to big in the first place.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: remove AMDGPU_VM_NO_FLUSH define
Christian König [Fri, 12 Aug 2016 09:40:11 +0000 (11:40 +0200)]
drm/amdgpu: remove AMDGPU_VM_NO_FLUSH define

Not used any more.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: cleanup the write_pte implementations
Christian König [Fri, 12 Aug 2016 09:33:30 +0000 (11:33 +0200)]
drm/amdgpu: cleanup the write_pte implementations

We don't need the gart mapping handling here any more.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: remove pages_addr handling from the VM code
Christian König [Thu, 11 Aug 2016 14:44:15 +0000 (16:44 +0200)]
drm/amdgpu: remove pages_addr handling from the VM code

Not needed any more.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: write PTEs directly into the IB.
Christian König [Thu, 11 Aug 2016 12:06:54 +0000 (14:06 +0200)]
drm/amdgpu: write PTEs directly into the IB.

Write the PTEs at the end of the IB instead of directly into the SDMA commands.
This can save quite some CPU cycles building the entries.

This doesn't change the DW estimation because PTEs where embedded into the IB
before as well. It just moves them to the end of the IB.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
8 years agodrm/amdgpu: add shadow flag V2
Chunming Zhou [Thu, 4 Aug 2016 07:47:50 +0000 (15:47 +0800)]
drm/amdgpu: add shadow flag V2

Indicate if need to sync between bo and shadow, where sync to where.
V2:
Rename to backup_shadow

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>
8 years agodrm/amdgpu: allocate shadow for pd/pt bo V2
Chunming Zhou [Thu, 4 Aug 2016 05:59:32 +0000 (13:59 +0800)]
drm/amdgpu: allocate shadow for pd/pt bo V2

The pd/pt shadow bo will be used to backup page table, when gpu reset
happens, we can restore the page table by them.
V2:
Free shadow bo.

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>
8 years agodrm/amdgpu: validate shadow as well when validating bo
Chunming Zhou [Thu, 4 Aug 2016 05:05:46 +0000 (13:05 +0800)]
drm/amdgpu: validate shadow as well when validating bo

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>
8 years agodrm/amdgpu: add shadow bo support V2
Chunming Zhou [Tue, 26 Jul 2016 06:13:21 +0000 (14:13 +0800)]
drm/amdgpu: add shadow bo support V2

shadow bo is the shadow of a bo, which is always in GTT,
which can be used to backup the original bo.
V2:
reference shadow parent, shadow bo will be freed by who allocted him.

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>