Rework the internals of the dma_resv object to allow adding more than one
write fence and remember for each fence what purpose it had.
This allows removing the workaround from amdgpu which used a container for
this instead.
Signed-off-by: Christian König
Reviewed-by: Daniel Vetter
Cc: amd-gfx@l
Smatch reports the following:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_hw_sequencer.c:2174
dcn10_enable_vblanks_synchronization() warn: if statement not indented
Signed-off-by: Haowen Bai
---
drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c | 14 +++---
1 file change
DP/HDMI audio on AMD PRO VII stops working after S3:
[ 149.450391] amdgpu :63:00.0: amdgpu: MODE1 reset
[ 149.450395] amdgpu :63:00.0: amdgpu: GPU mode1 reset
[ 149.450494] amdgpu :63:00.0: amdgpu: GPU psp mode1 reset
[ 149.983693] snd_hda_intel :63:00.1: refused to change power
The pointer dc is dereferencing pointer plane_state before plane_state
is being null checked. Fix this by assigning plane_state->ctx->dc to
dc only if plane_state is not NULL, otherwise just NULL.
Signed-off-by: Haowen Bai
---
drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c | 2 +-
1 f
Dear Arunpravin,
Thank you for your patch.
Am 07.04.22 um 07:46 schrieb Arunpravin Paneer Selvam:
- Switch to drm buddy allocator
- Add resource cursor support for drm buddy
I though after the last long discussion, you would actually act on the
review comments. Daniel wrote a good summary,
On 2022-04-07 01:52, Haowen Bai wrote:
> The pointer dc is dereferencing pointer plane_state before plane_state
> is being null checked. Fix this by assigning plane_state->ctx->dc to
> dc only if plane_state is not NULL, otherwise just NULL.
>
> Signed-off-by: Haowen Bai
> ---
> drivers/gpu/d
On 2022-04-07 02:00, Haowen Bai wrote:
> Smatch reports the following:
> drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_hw_sequencer.c:2174
> dcn10_enable_vblanks_synchronization() warn: if statement not indented
>
> Signed-off-by: Haowen Bai
Reviewed-by: Harry Wentland
Harry
> ---
>
Patch merged to amd-staging-drm-next.
Thanks a lot!
On 2022-04-05 15:32, Simon Ser wrote:
I've tested this patch and it fixes my bug [1]. Thanks!
Tested-by: Simon Ser
[1]: https://gitlab.freedesktop.org/drm/amd/-/issues/1734>
Hi Joshi,
Per your comment, I produced a fix to work with multi-GPU system for hotplug
support for our group’s internal usage.
I have tested on a 4-node MI100 system, which seems to be working. It is pushed
in the github PR.
The details are in:
https://github.com/RadeonOpenCompute/ROCK-Kernel-
Applied. Thanks!
Alex
On Thu, Apr 7, 2022 at 10:18 AM Harry Wentland wrote:
>
>
>
> On 2022-04-07 02:00, Haowen Bai wrote:
> > Smatch reports the following:
> > drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_hw_sequencer.c:2174
> > dcn10_enable_vblanks_synchronization() warn: if statement
Applied. Thanks!
Alex
On Thu, Apr 7, 2022 at 8:21 AM Kai-Heng Feng
wrote:
>
> DP/HDMI audio on AMD PRO VII stops working after S3:
> [ 149.450391] amdgpu :63:00.0: amdgpu: MODE1 reset
> [ 149.450395] amdgpu :63:00.0: amdgpu: GPU mode1 reset
> [ 149.450494] amdgpu :63:00.0: amdgpu
Actually this just causes another warning. Dropped for now. More below.
On Thu, Apr 7, 2022 at 11:52 AM Alex Deucher wrote:
>
> Applied. Thanks!
>
> Alex
>
> On Thu, Apr 7, 2022 at 10:18 AM Harry Wentland wrote:
> >
> >
> >
> > On 2022-04-07 02:00, Haowen Bai wrote:
> > > Smatch reports the f
Currently, the IO-links to the device being removed from topology,
are not cleared. As a result, there would be dangling links left in
the KFD topology. This patch aims to fix the following:
1. Cleanup all IO links to the device being removed.
2. Ensure that node numbering in sysfs and nodes proxim
On Wed, Apr 6, 2022 at 10:03 AM Yongqiang Sun wrote:
>
> driver loading failed on VEGA10 SRIOV VF with linux host due to a wide
> range of stolen reserved vram.
> Since VEGA10 SRIOV VF need to reserve vram for firmware with windows
> Hyper_V host specifically, check high part of FB address
> to on
[AMD Official Use Only]
Hi Shuotao,
Just sent out the patch to cleanup IO links upon KFD device removal to the
public mailing list.
Please try it, review it and let us know how it goes for you.
Thank you.
Regards,
Mukul
From: Shuotao Xu
Sent: Thursday, April 7, 2022 11:09 AM
To: Joshi, Mukul
Applied. Thanks!
Alex
On Wed, Apr 6, 2022 at 3:49 PM Tomasz Moń wrote:
>
> Enabling gfxoff quirk results in perfectly usable graphical user
> interface on MacBook Pro (15-inch, 2019) with Radeon Pro Vega 20 4 GB.
>
> Without the quirk, X server is completely unusable as every few seconds
> ther
Thanks Mukul very much!
The code looks neat, although kfd_locked looks still would cause trouble. I
will try it.
Best,
Shuotao
From: Joshi, Mukul
Date: Friday, April 8, 2022 at 12:23 AM
To: Shuotao Xu , Grodzovsky, Andrey
, amd-gfx@lists.freedesktop.org
Cc: Ziyue Yang , Lei Qu , Peng
Cheng
[AMD Official Use Only]
Thanks Shuotao.
If the IO link cleanup works ok for you, you can use this patch as the base for
adding your changes to add Hot Plug support. You can send a separate patch for
that.
Regards,
Mukul
From: Shuotao Xu
Sent: Thursday, April 7, 2022 12:28 PM
To: Joshi, Mukul
I suggest adding another patch to handle unbalanced decrement of
kfd_lock in kgd2kfd_suspend. This patch alone is not enough to fix
all removal issues.
Andrey
On 2022-04-07 12:15, Mukul Joshi wrote:
Currently, the IO-links to the device being removed from topology,
are not cleared. As a result,
On Thu, Apr 7, 2022 at 8:21 AM Kai-Heng Feng
wrote:
>
> DP/HDMI audio on AMD PRO VII stops working after S3:
> [ 149.450391] amdgpu :63:00.0: amdgpu: MODE1 reset
> [ 149.450395] amdgpu :63:00.0: amdgpu: GPU mode1 reset
> [ 149.450494] amdgpu :63:00.0: amdgpu: GPU psp mode1 reset
> [
Thanks Joshi. I will just do that. Will hold onto my current patch based on one
in the github PR.
Best regards,
Shuotao
From: Joshi, Mukul
Date: Friday, April 8, 2022 at 12:33 AM
To: Shuotao Xu , Grodzovsky, Andrey
, amd-gfx@lists.freedesktop.org
Cc: Ziyue Yang , Lei Qu , Peng
Cheng , Ran S
Applied. Thanks!
Alex
On Mon, Apr 4, 2022 at 11:57 AM Harry Wentland wrote:
>
>
>
> On 2022-04-04 11:43, Tom Rix wrote:
> >
> > On 4/4/22 8:22 AM, Harry Wentland wrote:
> >>
> >> On 2022-04-03 10:21, Tom Rix wrote:
> >>> Smatch reports this issue
> >>> hdcp1_execution.c:500:29: warning: functio
On 2022-04-07 12:07, Alex Deucher wrote:
> Actually this just causes another warning. Dropped for now. More below.
>
> On Thu, Apr 7, 2022 at 11:52 AM Alex Deucher wrote:
>>
>> Applied. Thanks!
>>
>> Alex
>>
>> On Thu, Apr 7, 2022 at 10:18 AM Harry Wentland
>> wrote:
>>>
>>>
>>>
>>> On 20
On Thu, Apr 7, 2022 at 1:32 PM Harry Wentland wrote:
>
>
>
> On 2022-04-07 12:07, Alex Deucher wrote:
> > Actually this just causes another warning. Dropped for now. More below.
> >
> > On Thu, Apr 7, 2022 at 11:52 AM Alex Deucher wrote:
> >>
> >> Applied. Thanks!
> >>
> >> Alex
> >>
> >> On T
On Thu, Apr 7, 2022 at 2:21 AM Christian König
wrote:
>
> Am 06.04.22 um 18:50 schrieb Grigory Vasilyev:
> > Instead of the 'amdgpu_ring_priority_level' type,
> > the 'amdgpu_gfx_pipe_priority' type was used,
> > which is an error when setting ring priority.
> > This is a minor error, but may caus
[Public]
> -Original Message-
> From: Alex Deucher
> Sent: Thursday, April 7, 2022 12:08
> To: Kai-Heng Feng
> Cc: Deucher, Alexander ; Koenig, Christian
> ; Pan, Xinhui ; David
> Airlie ; Maling list - DRI developers de...@lists.freedesktop.org>; LKML ; amd-
> gfx list ; Chiu, Solomon
Fixes: 8011a76ad370
("drm/amdgpu: Add stolen reserved memory for MI25 SRIOV.")
driver loading failed on VEGA10 SRIOV VF with linux host due to a wide
range of stolen reserved vram.
Since VEGA10 SRIOV VF need to reserve vram for firmware with windows
Hyper_V host specifically, check hypervisor type
Alex Deucher I will do it. Initially, I came across an error that
different enum types are used.
Now I looked at the files, and indeed AMDGPU_GFX_PIPE_PRIO_* is used
instead of AMDGPU_RING_PRIO_*.
The question remains whether it is worth increasing the priority to
AMDGPU_RING_PRIO_MAX=3 ?
Regards,
On Thu, Apr 7, 2022 at 2:35 PM Yongqiang Sun wrote:
>
> Fixes: 8011a76ad370
> ("drm/amdgpu: Add stolen reserved memory for MI25 SRIOV.")
Please put this line between the commit message and your signed-off-by.
>
> driver loading failed on VEGA10 SRIOV VF with linux host due to a wide
> range of s
Use do_div() rather than a a 64 bit divide.
Fixes: 3fe5739db48843 ("drm/amd/display: Add flip interval workaround")
Reported-by: kernel test robot
Signed-off-by: Alex Deucher
Cc: Angus Wang
Cc: Anthony Koo
Cc: Aric Cyr
---
drivers/gpu/drm/amd/display/modules/freesync/freesync.c | 5 -
1
cayman_default_state and cayman_default_size are only
used in ni.c. Single file symbols should be static.
So move their definitions to cayman_blit_shaders.h
and change their storage-class-specifier to static.
Remove unneeded cayman_blit_shader.c
cayman_ps/vs definitions were removed with
commit
On Thu, Apr 07, 2022 at 03:50:29PM -0400, Alex Deucher wrote:
> Use do_div() rather than a a 64 bit divide.
>
> Fixes: 3fe5739db48843 ("drm/amd/display: Add flip interval workaround")
> Reported-by: kernel test robot
> Signed-off-by: Alex Deucher
> Cc: Angus Wang
> Cc: Anthony Koo
> Cc: Aric C
Application could change XNACK enabled to disabled while KFD is draining
stall retry fault, the check for whether to drain retry faults must be
before the check for whether xnack_enabled, to avoid report incorrect vm
fault after application changes XNACK mode.
Signed-off-by: Philip Yang
---
driv
Change-Id: I09a2aae85cde3ab2cb6b042b973da6839ad024ec
Signed-off-by: Stanley.Yang
---
drivers/gpu/drm/amd/amdgpu/umc_v6_7.c | 62 ++-
1 file changed, 60 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c
b/drivers/gpu/drm/amd/amdgpu/umc_v6_7.
Contrary to the smu_cmn_send_smc_msg_with_param documentation, two
cases exist where messages are silently dropped with no error returned
to the caller. These cases occur in unusual situations where either:
1. the caller is a virtual GPU, or
2. a PCI recovery is underway and the HW is not ye
The synchronize_rcu call in destroy_events can take several ms, which
noticeably slows down applications destroying many events. Use kfree_rcu
to free the event structure asynchronously and eliminate the
synchronize_rcu call in the user thread.
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/a
MEC firmware sometimes sends signal interrupts without a valid context ID
on end of pipe events that don't intend to signal any HSA signals.
This triggers the slow path in kfd_signal_event_interrupt that scans the
entire event page for signaled events. Detect these signals in the top
half interrupt
Am 2022-04-07 um 21:21 schrieb Philip Yang:
Application could change XNACK enabled to disabled while KFD is draining
stall retry fault, the check for whether to drain retry faults must be
typo: stall -> stale
I think there is also a word missing, linking the two sentences. Maybe
"... stale re
[AMD Official Use Only]
> -Original Message-
> From: Stanley.Yang
> Sent: Friday, April 8, 2022 10:18 AM
> To: amd-gfx@lists.freedesktop.org; Zhang, Hawking
> ; Zhou1, Tao ; Li, Candice
>
> Cc: Yang, Stanley
> Subject: [PATCH Review 1/1] drm/amdgpu: print more correctable error info
>
We shall consider centralize UMC MCA status check in a helper function, at
least, querying IPID, SYND, and MCA_STATUS should be the same for both ue and
ce.
Regards,
Hawking
-Original Message-
From: Zhou1, Tao
Sent: Friday, April 8, 2022 10:59
To: Yang, Stanley ; amd-gfx@lists.freedes
Thanks for your suggestion, it’s better to centralize umc mca status check in a
helper function, will update.
Regards,
Stanley
> -邮件原件-
> 发件人: Zhang, Hawking
> 发送时间: Friday, April 8, 2022 11:11 AM
> 收件人: Zhou1, Tao ; Yang, Stanley
> ; amd-gfx@lists.freedesktop.org; Li, Candice
>
> 抄送: Y
Use flexible array member in ip discovery struct as recommended[1].
[1]
https://www.kernel.org/doc/html/latest/process/deprecated.html#zero-length-and-one-element-arrays
Signed-off-by: Lijo Lazar
---
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 2 +-
drivers/gpu/drm/amd/include/discovery.h
Dear Darren,
Thank you for your patch.
Am 08.04.22 um 04:26 schrieb Darren Powell:
Contrary to the smu_cmn_send_smc_msg_with_param documentation, two
cases exist where messages are silently dropped with no error returned
to the caller. These cases occur in unusual situations where either
It's useful to know the connector's max supported bpc for IGT
testing. Expose it via a debugfs file on the connector "output_bpc".
Example: cat /sys/kernel/debug/dri/0/DP-1/output_bpc
Cc: Jani Nikula
Cc: Ville Syrjälä
Cc: Harry Wentland
Signed-off-by: Bhanuprakash Modem
---
drivers/gpu/drm/d
This new debugfs will expose the currently using bpc by crtc.
It is very useful for verifying whether we enter the correct
output color depth from IGT.
This patch will also add the connector's max supported bpc to
"i915_display_info" debugfs.
Example:
cat /sys/kernel/debug/dri/0/crtc-0/i915_curre
This series will expose the Connector's max supported bpc via connector
debugfs and Crtc's current bpc via crtc debugfs. Also move the existing
vendor specific "output_bpc" logic to drm.
Test-with: 20220408065143.1485069-2-bhanuprakash.mo...@intel.com
Bhanuprakash Modem (3):
drm/debug: Expose c
As drm_connector already have the display_info, instead of creating
"output_bpc" debugfs in vendor specific driver, move the logic to
the drm layer.
This patch will also move "Current" bpc to the crtc debugfs from
connector debugfs, since we are getting this info from crtc_state.
Cc: Harry Wentla
47 matches
Mail list logo
