Re: [PATCH 4/4] drm/amdgpu: further optimize amdgpu_vm_handle_moved

I failed to git am the patch to test, the patch format has problem? From the patch self, is it possible that the vm->moved list is updated when you use local one? Regards, David Zhou On 2018年04月19日 17:29, Christian König wrote: Splice the moved list to a local one to avoid taking the

Re: [PATCH 1/4] drm/amdgpu: remove unused member

On 2018年04月19日 17:29, Christian König wrote: This lock isn't used any more. Signed-off-by: Christian König Reviewed-by: Chunming Zhou --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h | 3 --- 1 file changed, 3 deletions(-) diff --git

Re: [PATCH libdrm] amdgpu: dynamically detect number of drm devices

On 04/20/2018 09:50 AM, Zhang, Jerry (Junwei) wrote: On 04/19/2018 07:06 PM, Christian König wrote: Wouldn't it be simpler to just set MAX_CARDS_SUPPORTED to 128? Perhaps it's 64 for card number. Although CONTROL node is not used now, but only 64 slots are reserved for each type. otherwise,

Re: [PATCH libdrm] amdgpu: dynamically detect number of drm devices

On 04/19/2018 07:06 PM, Christian König wrote: Wouldn't it be simpler to just set MAX_CARDS_SUPPORTED to 128? Perhaps it's 64 for card number. Although CONTROL node is not used now, but only 64 slots are reserved for each type. otherwise, we may prepare a patch to update to 128. How do you

Re: [PATCH 00/20] drm/amdgpu: gfx off support

On Fri, Apr 20, 2018 at 01:58:33AM +0800, Kuehling, Felix wrote: > On 2018-04-19 09:51 AM, Alex Deucher wrote: > > On Wed, Apr 18, 2018 at 8:22 PM, Huang Rui wrote: > >> GFXOFF is the new GPU feature that save power consumption. It used RLC to > >> poweroff the gfx engine

Re: [PATCH 00/20] drm/amdgpu: gfx off support

On Thu, Apr 19, 2018 at 09:51:29PM +0800, Alex Deucher wrote: > On Wed, Apr 18, 2018 at 8:22 PM, Huang Rui wrote: > > GFXOFF is the new GPU feature that save power consumption. It used RLC to > > poweroff the gfx engine dynamicly when there is no workload on gfx pipe and > >

Re: amdgpu hangs on boot or shutdown on AMD Raven Ridge CPU (Engineer Sample)

On Thu, Apr 19, 2018 at 6:08 PM, drake wrote: > On Tue, Apr 3, 2018 at 11:31 AM, Daniel Drake wrote: >> On Thu, Mar 22, 2018 at 3:09 AM, Daniel Drake wrote: >>> On Tue, Feb 20, 2018 at 10:18 PM, Alex Deucher

Re: amdgpu hangs on boot or shutdown on AMD Raven Ridge CPU (Engineer Sample)

On Tue, Apr 3, 2018 at 11:31 AM, Daniel Drake wrote: > On Thu, Mar 22, 2018 at 3:09 AM, Daniel Drake wrote: >> On Tue, Feb 20, 2018 at 10:18 PM, Alex Deucher wrote: It seems that we are not alone seeing amdgpu-induced stability

Re: [PATCH 0/6] Add documentation power related sysfs interfaces

Thanks. It's great to see this documented. I haven't confirmed that the information is 100% accurate. Acked-by: Felix Kuehling On 2018-04-19 04:05 PM, Alex Deucher wrote: > Provide some initial documenation on how to use the power sysfs > APIs. > > Alex Deucher (6): >

[PATCH] drm/amdgpu: change pp_dpm clk/mclk/pcie input format

1. support more than 8 values when setting get_pp_dpm_mclk/ sclk/pcie, the former design just parse command format like "echo >　pp_dpm_sclk" and current can parse "echo xx xxx > pp_dpm_sclk" whose operation is more user-friendly and convinent and can offer more values; 2. be compatible

[PATCH] drm/amdgpu: support more than 8 values when setting sclk,

mclk and pcie of pp_dpm. 1. support more than 8 values when setting get_pp_dpm_mclk/ sclk/pcie, the former design just parse command format like "echo >　pp_dpm_sclk" and current can parse "echo xx xxx > pp_dpm_sclk" whose operation is more user-friendly and convinent and can offer more

[PATCH] mdgpu: support more than 8 values when setting sclk,

mclk and pcie of pp_dpm. 1. support more than 8 values when setting get_pp_dpm_mclk/ sclk/pcie, the former design just parse command format like "echo >　pp_dpm_sclk" and current can parse "echo xx xxx > pp_dpm_sclk" whose operation is more user-friendly and convinent and can offer more

[PATCH 6/6] drm/amdgpu/pm: document pp_od_clk_voltage

sysfs interface for fine grained clock and voltage control. Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 23 +++ 1 file changed, 23 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c

[PATCH 1/6] drm/amdgpu/pm: document power_dpm_force_performance_level

Provide documentation for power_dpm_force_performance_level which is used to adjust things related to GPU power states. Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 53 ++ 1 file changed, 53 insertions(+)

[PATCH 3/6] drm/amdgpu/pm: document pp_table

This file is for uploading new powerplay tables. Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 11 +++ 1 file changed, 11 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c

[PATCH 4/6] drm/amdgpu/pm: document pp_dpm_sclk pp_dpm_mclk pp_dpm_pcie

Used for manually masking dpm states. Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 18 ++ 1 file changed, 18 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c index

[PATCH 5/6] drm/amdgpu/pm: document pp_power_profile_mode

sysfs file for adjusting power level heuristics. Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 20 1 file changed, 20 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c

[PATCH 2/6] drm/amdgpu/pm: document power_dpm_state

This is a legacy file and is only provided for backwards compatibility. Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 31 +++ 1 file changed, 31 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c

[PATCH 0/6] Add documentation power related sysfs interfaces

Provide some initial documenation on how to use the power sysfs APIs. Alex Deucher (6): drm/amdgpu/pm: document power_dpm_force_performance_level drm/amdgpu/pm: document power_dpm_state drm/amdgpu/pm: document pp_table drm/amdgpu/pm: document pp_dpm_sclk pp_dpm_mclk pp_dpm_pcie

Re: [PATCH 00/20] drm/amdgpu: gfx off support

On Thu, Apr 19, 2018 at 1:58 PM, Felix Kuehling wrote: > On 2018-04-19 09:51 AM, Alex Deucher wrote: >> On Wed, Apr 18, 2018 at 8:22 PM, Huang Rui wrote: >>> GFXOFF is the new GPU feature that save power consumption. It used RLC to >>> poweroff the gfx

Re: [PATCH 00/20] drm/amdgpu: gfx off support

On 2018-04-19 09:51 AM, Alex Deucher wrote: > On Wed, Apr 18, 2018 at 8:22 PM, Huang Rui wrote: >> GFXOFF is the new GPU feature that save power consumption. It used RLC to >> poweroff the gfx engine dynamicly when there is no workload on gfx pipe and >> make >> gfx into

Re: [PATCH] amdkfd: always select MMU_NOTIFIER

On 2018-04-19 06:56 AM, Anders Roxell wrote: > On 28 March 2018 at 18:04, Christian König wrote: >> Am 28.03.2018 um 17:53 schrieb Arnd Bergmann: >>> Building amdkfd without MMU notifiers is broken: >>> >>> In file included from

Re: [PATCH] drm/amd/display: Disallow enabling CRTC without primary plane with FB

On 2018-04-19 04:38 PM, Harry Wentland wrote: > On 2018-04-19 03:43 AM, Michel Dänzer wrote: >> On 2018-04-18 10:26 PM, Harry Wentland wrote: >>> The below commit >>> >>> "drm/atomic: Try to preserve the crtc enabled state in >>> drm_atomic_remove_fb, v2" >>> >>> introduces a slight

[pull] amdgpu drm-next-4.17

Hi Dave, A few fixes for 4.17: - Fix a dark screen issue in DC - Fix clk/voltage dependency tracking for wattman - Update SMU interface for vega12 The following changes since commit a10beabba213924d876f2d10ca9351aeab93f58a: Merge branch 'drm-next-4.17' of

Re: [PATCH libdrm] amdgpu: enlarge the maximum number of cards supported

On 04/19/2018 10:24 PM, Christian König wrote: Am 19.04.2018 um 15:50 schrieb Xiaojie Yuan: 128 is the maximum number of cards that the kernel can support at the moment. Change-Id: I155b7b21635306d8ecc440b85fb8954501ab5599 Signed-off-by: Xiaojie Yuan Reviewed-by:

Re: [PATCH] drm/amd/display: Disallow enabling CRTC without primary plane with FB

On 2018-04-19 03:43 AM, Michel Dänzer wrote: > > [ Dropping stable@ (fixes with Cc: stable are picked up for stable > branches once they land in Linus' tree, there's no point sending them to > stable@ during review), adding dri-devel ] > > On 2018-04-18 10:26 PM, Harry Wentland wrote: >> The

Re: [PATCH libdrm] amdgpu: enlarge the maximum number of cards supported

Am 19.04.2018 um 15:50 schrieb Xiaojie Yuan: 128 is the maximum number of cards that the kernel can support at the moment. Change-Id: I155b7b21635306d8ecc440b85fb8954501ab5599 Signed-off-by: Xiaojie Yuan Reviewed-by: Christian König I assume

Re: [PATCH] drm/amd/pp: Fix output erroneous clock info on RV

On Thu, Apr 19, 2018 at 8:06 AM, Rex Zhu wrote: > Signed-off-by: Rex Zhu Please add a proper patch description explaining the change in units and the switch to using dynamic state rather than hardcoded values. > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c

Re: [PATCH 00/20] drm/amdgpu: gfx off support

On Wed, Apr 18, 2018 at 8:22 PM, Huang Rui wrote: > GFXOFF is the new GPU feature that save power consumption. It used RLC to > poweroff the gfx engine dynamicly when there is no workload on gfx pipe and > make > gfx into "idle" state. > 1. Add three additional RLC ucodes, and

[PATCH libdrm] amdgpu: enlarge the maximum number of cards supported

128 is the maximum number of cards that the kernel can support at the moment. Change-Id: I155b7b21635306d8ecc440b85fb8954501ab5599 Signed-off-by: Xiaojie Yuan --- tests/amdgpu/amdgpu_test.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git

Re: [PATCH libdrm] amdgpu: dynamically detect number of drm devices

On 04/19/2018 09:11 PM, Christian König wrote: Am 19.04.2018 um 15:06 schrieb xiyuan: Hi Christian, The problem I encountered is that amdgpu_test throws segment fault on a platform with 1 iGPU + 4 dGPU. The root cause is that 'drm_count' returned by drmGetDevices() is 5 which exceeds

Re: [PATCH libdrm] amdgpu: dynamically detect number of drm devices

Am 19.04.2018 um 15:06 schrieb xiyuan: Hi Christian, The problem I encountered is that amdgpu_test throws segment fault on a platform with 1 iGPU + 4 dGPU. The root cause is that 'drm_count' returned by drmGetDevices() is 5 which exceeds MAX_CARDS_SUPPORTED, so I decided to make it more

Re: [PATCH 4/8] dma-buf: add peer2peer flag

On Mon, Apr 16, 2018 at 03:38:56PM +0200, Daniel Vetter wrote: > We've broken that assumption in i915 years ago. Not struct page backed > gpu memory is very real. > > Of course we'll never feed such a strange sg table to a driver which > doesn't understand it, but allowing sg_page == NULL works

Re: [PATCH] amdkfd: always select MMU_NOTIFIER

On 28 March 2018 at 18:04, Christian König wrote: > Am 28.03.2018 um 17:53 schrieb Arnd Bergmann: >> >> Building amdkfd without MMU notifiers is broken: >> >> In file included from drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c:28: >>

Re: [PATCH libdrm] amdgpu: dynamically detect number of drm devices

Hi Christian, The problem I encountered is that amdgpu_test throws segment fault on a platform with 1 iGPU + 4 dGPU. The root cause is that 'drm_count' returned by drmGetDevices() is 5 which exceeds MAX_CARDS_SUPPORTED, so I decided to make it more flexible to get the actual number of drm

[PATCH] drm/amd/pp: Fix output erroneous clock info on RV

Signed-off-by: Rex Zhu --- drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c| 4 +-- drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c | 42 --- drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.h | 2 -- 3 files changed, 24 insertions(+), 24 deletions(-)

Re: [PATCH 3/4] drm/amdgpu: further optimize amdgpu_vm_validate_pt_bos

Please ignore this one, it doesn't work correctly. Validating a BO can put it back on the moved list. Christian. Am 19.04.2018 um 11:29 schrieb Christian König: Put all moved BOs on a temporary list and splice that one to the moved list when we are done. Saves us taking and releasing the

Re: [PATCH libdrm] amdgpu: dynamically detect number of drm devices

Wouldn't it be simpler to just set MAX_CARDS_SUPPORTED to 128? Regards, Christian. Am 19.04.2018 um 12:12 schrieb Xiaojie Yuan: Change-Id: I36764951bebbcbf06cf84dd43ee946a34ec7b100 Signed-off-by: Xiaojie Yuan --- tests/amdgpu/amdgpu_test.c | 44

[PATCH libdrm] amdgpu: dynamically detect number of drm devices

Change-Id: I36764951bebbcbf06cf84dd43ee946a34ec7b100 Signed-off-by: Xiaojie Yuan --- tests/amdgpu/amdgpu_test.c | 44 -- tests/amdgpu/amdgpu_test.h | 7 +- 2 files changed, 34 insertions(+), 17 deletions(-) diff --git

[PATCH 4/4] drm/amdgpu: further optimize amdgpu_vm_handle_moved

Splice the moved list to a local one to avoid taking the lock over and over again. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 27 ++- 1 file changed, 14 insertions(+), 13 deletions(-) diff --git

[PATCH 3/4] drm/amdgpu: further optimize amdgpu_vm_validate_pt_bos

Put all moved BOs on a temporary list and splice that one to the moved list when we are done. Saves us taking and releasing the moved lock multiple times. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 34 --

[PATCH 2/4] drm/amdgpu: rework VM state machine lock handling

Only the moved state needs a separate spin lock protection. All other states are protected by reserving the VM anyway. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 70 --

[PATCH 1/4] drm/amdgpu: remove unused member

This lock isn't used any more. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index

Re: [PATCH v2] drm/amdgpu: simplify bo_va list when vm bo update (v2)

On 04/19/2018 03:42 PM, Christian König wrote: Am 19.04.2018 um 09:40 schrieb Zhang, Jerry (Junwei): On 04/19/2018 03:06 PM, Christian König wrote: Am 19.04.2018 um 08:53 schrieb Junwei Zhang: v2: fix compiling warning Signed-off-by: Junwei Zhang ---

Re: [PATCH] drm/amd/display: Disallow enabling CRTC without primary plane with FB

[ Dropping stable@ (fixes with Cc: stable are picked up for stable branches once they land in Linus' tree, there's no point sending them to stable@ during review), adding dri-devel ] On 2018-04-18 10:26 PM, Harry Wentland wrote: > The below commit > > "drm/atomic: Try to preserve the crtc

Re: [PATCH 2/2] drm/amd/powerplay: add registry key to disable ACG

On Thu, Apr 19, 2018 at 03:11:36PM +0800, Kenneth Feng wrote: > For the dummy ACG fuses,need to disable ACG, otherwise > corruption will be caused. > > Change-Id: I03db372bb6e934e9230f0745aa19c683adfdeccb > Signed-off-by: Kenneth Feng > --- >

Re: [PATCH v2] drm/amdgpu: simplify bo_va list when vm bo update (v2)

Am 19.04.2018 um 09:40 schrieb Zhang, Jerry (Junwei): On 04/19/2018 03:06 PM, Christian König wrote: Am 19.04.2018 um 08:53 schrieb Junwei Zhang: v2: fix compiling warning Signed-off-by: Junwei Zhang ---   drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 22

Re: [PATCH v2] drm/amdgpu: simplify bo_va list when vm bo update (v2)

On 04/19/2018 03:06 PM, Christian König wrote: Am 19.04.2018 um 08:53 schrieb Junwei Zhang: v2: fix compiling warning Signed-off-by: Junwei Zhang --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 22 ++ 1 file changed, 10 insertions(+), 12 deletions(-)

Re: [PATCH 1/2] drm/amd/powerplay: header file interface to SMU update

On Thu, Apr 19, 2018 at 03:09:58PM +0800, Kenneth Feng wrote: > Change-Id: I6d309f651dff5f657c1aa424efec2048b9b64a0a > Signed-off-by: Kenneth Feng Reviewed-by: Huang Rui > --- > drivers/gpu/drm/amd/powerplay/inc/vega12/smu9_driver_if.h | 4 +++- > 1

Re: [PATCH 00/20] drm/amdgpu: gfx off support

I can't judge technical correctness, but coding style, commit messages etc..  look good on first glance. Series is Acked-by: Christian König . Regards, Christian. Am 19.04.2018 um 02:22 schrieb Huang Rui: GFXOFF is the new GPU feature that save power consumption. It

Re: [PATCH 1/3] drm/amdgpu: Rename amdgpu_display_framebuffer_domains()

With Alex note addressed patch #1 & #2 are Reviewed-by: Christian König Regards, Christian. Am 18.04.2018 um 23:50 schrieb Samuel Li: It returns supported domains, and domains actually used are to be decided later. Signed-off-by: Samuel Li ---

[PATCH 2/2] drm/amd/powerplay: add registry key to disable ACG

For the dummy ACG fuses,need to disable ACG, otherwise corruption will be caused. Change-Id: I03db372bb6e934e9230f0745aa19c683adfdeccb Signed-off-by: Kenneth Feng --- drivers/gpu/drm/amd/powerplay/hwmgr/vega12_processpptables.c | 5 +

[PATCH 1/2] drm/amd/powerplay: header file interface to SMU update

Change-Id: I6d309f651dff5f657c1aa424efec2048b9b64a0a Signed-off-by: Kenneth Feng --- drivers/gpu/drm/amd/powerplay/inc/vega12/smu9_driver_if.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/powerplay/inc/vega12/smu9_driver_if.h

RE: [PATCH] drm/amdgpu: init gfx9 aperture settings

Reviewed-by: Hawking Zhang Regards, Hawking -Original Message- From: amd-gfx On Behalf Of Flora Cui Sent: 2018年4月19日 13:35 To: amd-gfx@lists.freedesktop.org Cc: Cui, Flora Subject: [PATCH] drm/amdgpu: init

Re: [PATCH v2] drm/amdgpu: simplify bo_va list when vm bo update (v2)

Am 19.04.2018 um 08:53 schrieb Junwei Zhang: v2: fix compiling warning Signed-off-by: Junwei Zhang --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 22 ++ 1 file changed, 10 insertions(+), 12 deletions(-) diff --git

Re: [PATCH 1/1] drm/amdgpu: Enable scatter gather display support

Am 18.04.2018 um 18:21 schrieb Samuel Li: On 2018-04-18 12:16 PM, Christian König wrote: Am 18.04.2018 um 17:29 schrieb Samuel Li: On 2018-04-18 12:14 AM, Alex Deucher wrote: On Tue, Apr 17, 2018 at 8:40 PM, Samuel Li wrote: It's auto by default. For CZ/ST, auto setting

[PATCH v2] drm/amdgpu: simplify bo_va list when vm bo update (v2)

v2: fix compiling warning Signed-off-by: Junwei Zhang --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 22 ++ 1 file changed, 10 insertions(+), 12 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c

Re: [PATCH] drm/amdgpu: simplify bo_va list when vm bo update

On 04/19/2018 02:26 PM, zhoucm1 wrote: On 2018年04月19日 13:31, Junwei Zhang wrote: Signed-off-by: Junwei Zhang --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 21 + 1 file changed, 9 insertions(+), 12 deletions(-) diff --git

Re: [PATCH] drm/amdgpu: simplify bo_va list when vm bo update

On 2018年04月19日 13:31, Junwei Zhang wrote: Signed-off-by: Junwei Zhang --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 21 + 1 file changed, 9 insertions(+), 12 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c