For the nouveau and drm core changes
Reviewed-by: Lyude Paul
On Mon, 2018-11-12 at 16:01 +0100, Maarten Lankhorst wrote:
> We already have __drm_atomic_helper_connector_reset() and
> __drm_atomic_helper_plane_reset(), extend this to crtc as well.
>
> Most drivers already have a gpu reset hook,
On Wed, Nov 21, 2018 at 1:11 PM Andrey Grodzovsky
wrote:
>
> It's needed for device reset of entire hive.
>
> Signed-off-by: Andrey Grodzovsky
Reviewed-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 ++
> drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c | 7 ++-
> 2 files
Depends what was the reason for triggering the reset for that node how
do we know ?
If the reason was RAS error that probably not hard to check all errors
are cleared, but
if the reason was job timeout on that specific node I will need to
recheck that no jobs are left in incomplete state
state.
On Wed, Nov 21, 2018 at 5:00 PM Oak Zeng wrote:
>
> v2: Changed file name for consistency
>
> Change-Id: Ib2c570224321bb7002d2ed01f43ac70203e86f88
> Signed-off-by: Oak Zeng
> Suggested-by: Felix Kuehling
> Suggested-by: Alex Deucher
Just drop this patch and use the vega10 definition for
On Wed, Nov 21, 2018 at 1:11 PM Andrey Grodzovsky
wrote:
>
> This is prep work for updating each PSP FW in hive after
> GPU reset.
> Split into build topology SW state and update each PSP FW in the hive.
> Save topology and count of XGMI devices for reuse.
>
> Signed-off-by: Andrey Grodzovsky
>
On 11/21/2018 02:29 PM, Alex Deucher wrote:
> On Wed, Nov 21, 2018 at 1:11 PM Andrey Grodzovsky
> wrote:
>> This is prep work for updating each PSP FW in hive after
>> GPU reset.
>> Split into build topology SW state and update each PSP FW in the hive.
>> Save topology and count of XGMI devices
v2: Use enum definition instead of hardcoded number
Change-Id: Id64eb98f5b1c24b51eb2fd5a083086fc3515813d
Signed-off-by: Oak Zeng
Suggested-by: Felix Kuehling
Suggested-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/Makefile | 2 +-
drivers/gpu/drm/amd/amdgpu/vi.h | 2 +-
v2: Use enum definition instead of hardcoded value
Change-Id: I04d22fb717ac50483c0835f160a2e860e344f358
Signed-off-by: Oak Zeng
Suggested-by: Felix Kuehling
Suggested-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 50
Change-Id: I84475efcfb482c474fccb133010abb5df5f4
Signed-off-by: Oak Zeng
Suggested-by: Felix Kuehling
Suggested-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 27 ---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
Change-Id: I2f004bbbe2565035460686f4fc16e86b77a2a9b5
Signed-off-by: Oak Zeng
Suggested-by: Felix Kuehling
Suggested-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
This is a preparation work to make reserved doorbell index per device,
instead of using a global macro definition. By doing this, we can easily
change doorbell layout for future ASICs while not affecting ASICs in
production.
Change-Id: If08e2bc9d0749748ed4083ba4eb32a4698763085
Signed-off-by: Oak
v2: Changed file name for consistency
Change-Id: Ib2c570224321bb7002d2ed01f43ac70203e86f88
Signed-off-by: Oak Zeng
Suggested-by: Felix Kuehling
Suggested-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/Makefile | 2 +-
drivers/gpu/drm/amd/amdgpu/soc15.h | 1 +
On Wed, Nov 21, 2018 at 2:36 PM Grodzovsky, Andrey
wrote:
>
>
>
> On 11/21/2018 02:29 PM, Alex Deucher wrote:
> > On Wed, Nov 21, 2018 at 1:11 PM Andrey Grodzovsky
> > wrote:
> >> This is prep work for updating each PSP FW in hive after
> >> GPU reset.
> >> Split into build topology SW state and
On Wed, Nov 21, 2018 at 5:00 PM Oak Zeng wrote:
>
> v2: Use enum definition instead of hardcoded value
>
> Change-Id: I04d22fb717ac50483c0835f160a2e860e344f358
> Signed-off-by: Oak Zeng
> Suggested-by: Felix Kuehling
> Suggested-by: Alex Deucher
This patch should come before patch 5 so we
Applied. thanks!
On Wed, Nov 21, 2018 at 3:16 AM Zhang, Jerry(Junwei)
wrote:
>
> On 11/21/18 3:11 PM, Brajeswar Ghosh wrote:
> > Remove gca/gfx_8_0_d.h which is included more than once
> >
> > Signed-off-by: Brajeswar Ghosh
> Reviewed-by: Junwei Zhang
>
> > ---
> >
Applied. thanks!
On Wed, Nov 21, 2018 at 9:25 AM Brajeswar Ghosh
wrote:
>
> Remove dm_services_types.h which is included more than once
>
> Signed-off-by: Brajeswar Ghosh
> ---
> drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git
Applied. thanks!
On Wed, Nov 21, 2018 at 9:24 AM Brajeswar Ghosh
wrote:
>
> Remove gca/gfx_8_0_sh_mask.h which is included more than once
>
> Signed-off-by: Brajeswar Ghosh
> ---
> drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git
On 2018-11-20 10:16 a.m., David Francis wrote:
> ACPI ATIF has a function called query
> backlight transfer characteristics. Among the
> information returned by this function is
> the minimum and maximum input signals for the
> backlight
>
> Call that function on ACPI init. When DM
> backlight
I saw you use the global xgmi_mutex to prevent concurrent reset to be
triggered by different nodes , but after the mutex been released ,
current node may grap the mutex and continue to do another reset .
Maybe we should check the GPU status and skip the reset in this case
since the
On Wed, Nov 21, 2018 at 5:00 PM Oak Zeng wrote:
>
> v2: Use enum definition instead of hardcoded number
>
> Change-Id: Id64eb98f5b1c24b51eb2fd5a083086fc3515813d
> Signed-off-by: Oak Zeng
> Suggested-by: Felix Kuehling
> Suggested-by: Alex Deucher
Since this covers cik and vi, maybe call it
(resending because previous email switched to HTML mode and was filtered out)
Hi Tejun,
On Tue, Nov 20, 2018 at 5:30 PM Tejun Heo wrote:
> On Tue, Nov 20, 2018 at 10:21:14PM +, Ho, Kenny wrote:
> > By this reply, are you suggesting that vendor specific resources
> > will never be acceptable
On Wed, Nov 21, 2018 at 5:00 PM Oak Zeng wrote:
>
> This is a preparation work to make reserved doorbell index per device,
> instead of using a global macro definition. By doing this, we can easily
> change doorbell layout for future ASICs while not affecting ASICs in
> production.
>
> Change-Id:
On Wed, Nov 21, 2018 at 5:00 PM Oak Zeng wrote:
>
> Change-Id: I2f004bbbe2565035460686f4fc16e86b77a2a9b5
> Signed-off-by: Oak Zeng
> Suggested-by: Felix Kuehling
> Suggested-by: Alex Deucher
I think it would be cleaner to call this in the soc files directly
(cik.c, vi.c, soc15.c in the
On 2018-11-20 11:00 a.m., David Francis wrote:
> Adaptive Backlight Management (ABM) is a feature
> that reduces backlight level to save power, while
> increasing pixel contrast and pixel luminance
> to maintain readability and image quality.
>
> ABM will adjust in response to the
> pixel
On 12/11/2018 15:01, Maarten Lankhorst wrote:
> We already have __drm_atomic_helper_connector_reset() and
> __drm_atomic_helper_plane_reset(), extend this to crtc as well.
>
> Most drivers already have a gpu reset hook, correct it.
> Nouveau already implemented its own
On Wed, Nov 21, 2018 at 5:00 PM Oak Zeng wrote:
>
> v2: Use enum definition instead of hardcoded value
>
> Change-Id: Ib72058337f0aa53adfc6c6aae5341a7cd665111a
> Signed-off-by: Oak Zeng
> Suggested-by: Felix Kuehling
> Suggested-by: Alex Deucher
Reviewed-by: Alex Deucher
> ---
>
On Wed, Nov 21, 2018 at 1:10 PM Andrey Grodzovsky
wrote:
>
> Currently just a place holder until support from PSP in place.
>
> Signed-off-by: Andrey Grodzovsky
I think we can drop this patch for now until we figure out what the
proper reset method will be for this. We may end up using baco.
Hi Tejun,
On Tue, Nov 20, 2018 at 5:30 PM Tejun Heo wrote:
> On Tue, Nov 20, 2018 at 10:21:14PM +, Ho, Kenny wrote:
> > By this reply, are you suggesting that vendor specific resources
> > will never be acceptable to be managed under cgroup? Let say a user
>
> I wouldn't say never but
On Wed, Nov 21, 2018 at 01:40:43PM +0200, Jani Nikula wrote:
> On Tue, 20 Nov 2018, Ville Syrjala wrote:
> > From: Ville Syrjälä
> >
> > Make life easier for drivers by simply passing the connector
> > to drm_hdmi_avi_infoframe_from_display_mode() and
> > drm_hdmi_avi_infoframe_quant_range().
On Tue, 20 Nov 2018, Ville Syrjala wrote:
> From: Ville Syrjälä
>
> Make life easier for drivers by simply passing the connector
> to drm_hdmi_avi_infoframe_from_display_mode() and
> drm_hdmi_avi_infoframe_quant_range(). That way drivers don't
> need to worry about is_hdmi2_sink mess.
Overall
Am 20.11.18 um 19:58 schrieb Kenny Ho:
Account for the total size of buffer object requested to amdgpu by
buffer type on a per cgroup basis.
x prefix in the control file name x.bo_requested.amd.stat signify
experimental.
Change-Id: Ifb680c4bcf3652879a7a659510e25680c2465cf6
Signed-off-by: Kenny
Am 20.11.18 um 21:57 schrieb Eric Anholt:
Kenny Ho writes:
Account for the number of command submitted to amdgpu by type on a per
cgroup basis, for the purpose of profiling/monitoring applications.
For profiling other drivers, I've used perf tracepoints, which let you
get useful timelines of
Am 20.11.18 um 19:58 schrieb Kenny Ho:
Since many parts of the DRM subsystem has vendor-specific
implementations, we introduce mechanisms for vendor to register their
specific resources and control files to the DRM cgroup subsystem. A
vendor will register itself with the DRM cgroup subsystem
Am 20.11.18 um 19:58 schrieb Kenny Ho:
Change-Id: Ib66c44ac1b1c367659e362a2fc05b6fbb3805876
Signed-off-by: Kenny Ho
---
drivers/gpu/drm/amd/amdgpu/Makefile | 3 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7
drivers/gpu/drm/amd/amdgpu/amdgpu_drmcgrp.c | 37
Also deleted doorbell index macro definition
Change-Id: I84475efcfb482c474fccb133010abb5df5f4
Signed-off-by: Oak Zeng
Suggested-by: Felix Kuehling
Suggested-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h| 106 -
Change-Id: Id64eb98f5b1c24b51eb2fd5a083086fc3515813d
Signed-off-by: Oak Zeng
Suggested-by: Felix Kuehling
Suggested-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/Makefile| 3 +-
drivers/gpu/drm/amd/amdgpu/vi.h| 2 +-
This is a preparation work to make reserved doorbell index per device,
instead of using a global macro definition. By doing this, we can easily
change doorbell layout for future ASICs while not affecting ASICs in
production.
Change-Id: If08e2bc9d0749748ed4083ba4eb32a4698763085
Signed-off-by: Oak
Change-Id: Ib72058337f0aa53adfc6c6aae5341a7cd665111a
Signed-off-by: Oak Zeng
Suggested-by: Felix Kuehling
Suggested-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/Makefile| 2 +-
drivers/gpu/drm/amd/amdgpu/soc15.h | 1 +
Change-Id: I04d22fb717ac50483c0835f160a2e860e344f358
Signed-off-by: Oak Zeng
Suggested-by: Felix Kuehling
Suggested-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/Makefile| 4 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 +-
drivers/gpu/drm/amd/amdgpu/soc15.h
The doorbell index defines should be compatible with what is used in
windows driver . I don't see the necessary to introduce the new init
file for this instead of use the original MACRO defines. We need to
coordinate with windows driver team for a new user queue SDMA doorbell
range and used
Remove dm_services_types.h which is included more than once
Signed-off-by: Brajeswar Ghosh
---
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
Remove gca/gfx_8_0_sh_mask.h which is included more than once
Signed-off-by: Brajeswar Ghosh
---
drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
index
On 2018-11-16 2:59 p.m., Bhawanpreet Lakha wrote:
> Before:
> We use drm_match_cea_mode() to get the vic for any mode we
> want to set, most of the time vic will be different for the new mode.
>
> DC uses memcmp to check if timing changed, in this case DC will
> say timing changed and we endup
Reviewed-by: Philip Yang
On 2018-11-21 9:54 a.m., Alex Deucher wrote:
> Keep it disabled until we confirm it's ready.
>
> Signed-off-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git
Keep it disabled until we confirm it's ready.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index
Do we need this for vega12 as well?
From: amd-gfx on behalf of Alex Deucher
Sent: Wednesday, November 21, 2018 2:25:06 AM
To: Quan, Evan
Cc: amd-gfx list
Subject: Re: [PATCH] drm/amdgpu: disable page queue support for Vega20
Reviewed-by: Alex Deucher
On Wed,
when returned fence is not valid mostly due to userspace ignored
previous error causes NULL pointer dereference.
Signed-off-by: Deepak Sharma
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
when returned fence is not valid mostly due to userspace ignored
previous error causes NULL pointer dereference
Signed-off-by: Deepak Sharma
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> -Original Message-
> From: amd-gfx On Behalf Of
> Sharma, Deepak
> Sent: Thursday, November 22, 2018 10:37 AM
> To: amd-gfx@lists.freedesktop.org
> Cc: Sharma, Deepak
> Subject: [PATCH] drm/amd: add the checking to avoid NULL pointer
> dereference
>
> when returned fence is not
On 2018年11月22日 14:56, Sharma, Deepak wrote:
when returned fence is not valid mostly due to userspace ignored
previous error causes NULL pointer dereference.
Signed-off-by: Deepak Sharma
Reviewed-by: Chunming Zhou
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 ++
1 file changed, 2
On Mon, Oct 22, 2018 at 10:53:22PM +0530, Arun KS wrote:
> Remove managed_page_count_lock spinlock and instead use atomic
> variables.
>
> Suggested-by: Michal Hocko
> Suggested-by: Vlastimil Babka
> Signed-off-by: Arun KS
>
> ---
> As discussed here,
>
This is prep work for updating each PSP FW in hive after
GPU reset.
Split into build topology SW state and update each PSP FW in the hive.
Save topology and count of XGMI devices for reuse.
Signed-off-by: Andrey Grodzovsky
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 5 +++
This set of patches adds support to reset entire XGMI hive
when reset is required.
Patches 1 and 2 add place holder for PSP mode 0 reset.
The actual implementation is still TBD on PSP side.
Patches 3-4 refactoring a bit the XGMI infrastructure as
preparaton for the actual hive reset change.
It's needed for device reset of entire hive.
Signed-off-by: Andrey Grodzovsky
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c | 7 ++-
2 files changed, 8 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
Currently just a place holder until support from PSP in place.
Signed-off-by: Andrey Grodzovsky
---
drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h | 3 +++
drivers/gpu/drm/amd/amdgpu/psp_v11_0.c | 8
2 files changed, 11 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
In case of active XGMI hive do mode 0 reset as requsted
by design.
Signed-off-by: Andrey Grodzovsky
---
drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
For XGMI hive case do reset in steps where each step iterates over
all devs in hive. This especially important for asic reset
since all PSP FW in hive must come up within a limited time
(around 1 sec) to properply negotiate the link.
Do this by refactoring amdgpu_device_gpu_recover and
Hi Ville,
Thank you for the patch.
On Tuesday, 20 November 2018 18:13:42 EET Ville Syrjala wrote:
> From: Ville Syrjälä
>
> Make life easier for drivers by simply passing the connector
> to drm_hdmi_avi_infoframe_from_display_mode() and
> drm_hdmi_avi_infoframe_quant_range(). That way drivers
On Wed, Nov 21, 2018 at 10:52 AM Oak Zeng wrote:
>
> Change-Id: Ib72058337f0aa53adfc6c6aae5341a7cd665111a
> Signed-off-by: Oak Zeng
> Suggested-by: Felix Kuehling
> Suggested-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/Makefile| 2 +-
>
On Wed, Nov 21, 2018 at 10:52 AM Oak Zeng wrote:
>
> Change-Id: I04d22fb717ac50483c0835f160a2e860e344f358
> Signed-off-by: Oak Zeng
> Suggested-by: Felix Kuehling
> Suggested-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/Makefile| 4 +-
>
The new doorbell layout for vega20 was an agreement b/t Felix/Alex and Windows
team. Windows team want us to try the new layout first. If it works, they will
apply the same layout.
Previously we used a MACRO definitions for doorbell layout. There are 32-bit
and 64-bit doorbell layout.
>No need to add a new file for this. Just add this to vega20_reg_init.c.
Agreed.
>Also, please use the doorbell enums rather than hardcoding the numbers.
Ok. Where should I put the doorbell enums for vega20? In amdgpu.h or in
vega20_reg_init.c?
thanks,
Oak
-Original Message-
From:
Hi Dave,
- OD fixes for powerplay
- Vega20 fixes
- KFD fix for Kaveri
- add missing firmware declaration for hainan (SI chip)
- Fix DC user experience regressions related to panels that support >8 bpc
The following changes since commit 9826b1138e497dfb48f5cc64a82c219b4d0932da:
Merge branch
On Wed, Nov 21, 2018 at 10:52 AM Oak Zeng wrote:
>
> Change-Id: Ib2c570224321bb7002d2ed01f43ac70203e86f88
> Signed-off-by: Oak Zeng
> Suggested-by: Felix Kuehling
> Suggested-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/Makefile| 2 +-
>
On Wed, Nov 21, 2018 at 10:52 AM Oak Zeng wrote:
>
> Change-Id: I2f004bbbe2565035460686f4fc16e86b77a2a9b5
> Signed-off-by: Oak Zeng
> Suggested-by: Felix Kuehling
> Suggested-by: Alex Deucher
I think it would be cleaner to add this to the *_set_ip_blocks()
functions in cik.c, vi.c, and
On Wed, Nov 21, 2018 at 11:37 AM Zeng, Oak wrote:
>
> >No need to add a new file for this. Just add this to vega20_reg_init.c.
> Agreed.
>
> >Also, please use the doorbell enums rather than hardcoding the numbers.
> Ok. Where should I put the doorbell enums for vega20? In amdgpu.h or in
>
66 matches
Mail list logo
