RE - Maybe this would be better in psp_gfx_if.h since it's sort of part of the
firmware interface, but I guess it's not really part of the psp ring interface.
Yes, these cmds are not submitted to the psp ring and actually issued before
the psp ring setup. They are not in the internal
On Thu, 2019-07-04 at 15:05 -0400, sunpeng...@amd.com wrote:
> From: Ville Syrjälä
>
> All available downstream ports - physical and logical - are exposed for
> each MST device. They are listed in /dev/, following the same naming
> scheme as SST devices by appending an incremental ID.
>
>
gah. So, I was originally going to ask "why didn't we add the connector name
into this?", but then I realized we're doing the right thing already and just
doing card%d-%s % (card_number, path_prop). Which means we probably really don't
want to add a property called mstpath, since it's hardly
[adding dri-devel]
On 2019-07-09 11:59 p.m., Kuehling, Felix wrote:
> This notifies the driver that a BO is about to be released.
>
> Releasing a BO also invokes the move_notify callback from
> ttm_bo_cleanup_memtype_use, but that happens too late for anything
> that would add fences to the BO
Some small nitpicks
On Thu, 2019-07-04 at 15:05 -0400, sunpeng...@amd.com wrote:
> From: Ville Syrjälä
>
> All available downstream ports - physical and logical - are exposed for
> each MST device. They are listed in /dev/, following the same naming
> scheme as SST devices by appending an
On 2019-07-10 3:22 a.m., Michel Dänzer wrote:
> On 2019-07-09 9:00 p.m., Kuehling, Felix wrote:
>> On 2019-07-09 6:34 a.m., Michel Dänzer wrote:
>>> On 2019-07-09 7:32 a.m., Kuehling, Felix wrote:
This memory allocation flag will be used to indicate BOs containing
sensitive data that
gentle ping
On 2019/6/26 19:35, maowenan wrote:
>
>
> On 2019/6/25 1:42, Kim, Jonathan wrote:
>> Immediate return should be ok since perf registration isn't dependent on gpu
>> hw.
>>
>> Reviewed-by: Jonathan Kim
>
> thanks for review.
>
>>
>> -Original Message-
>> From: Mao Wenan
On Tue, Jul 9, 2019 at 11:59 PM Kuehling, Felix wrote:
>
> With mm_nodes larger than 4GB, byte_count in amdgpu_fill_buffer would
> overflow.
>
> Signed-off-by: Felix Kuehling
Good catch.
Reviewed-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 9 +
> 1 file
On Tue, Jul 9, 2019 at 9:17 PM Quan, Evan wrote:
>
> Hi Alex,
>
>
>
> There are actually two MAX_USEC_TIMEOUT related macros.
>
> AMD_MAX_USEC_TIMEOUT is used for SMU and AMDGPU_MAX_USEC_TIMEOUT is used
> widely in other parts.
>
> Maybe you mistreated it as AMDGPU_MAX_USEC_TIMEOUT. I was
On Wed, Jul 10, 2019 at 11:59 AM Hawking Zhang wrote:
>
Please add a patch description.
> Change-Id: Ief4c1e5ca01df0a028a784c0faf37544939733a3
> Signed-off-by: Hawking Zhang
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h | 9 +
> drivers/gpu/drm/amd/amdgpu/psp_v11_0.c | 4 ++--
>
Starting from navi10, driver should send Key Database Load command
to Bootloader before loading sys_drv and sos
Change-Id: Ib82d21840fb77da2217dd8b8f013177e61d72990
Signed-off-by: John Clements
Signed-off-by: Hawking Zhang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 10 +++
Change-Id: Ief4c1e5ca01df0a028a784c0faf37544939733a3
Signed-off-by: Hawking Zhang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h | 9 +
drivers/gpu/drm/amd/amdgpu/psp_v11_0.c | 4 ++--
drivers/gpu/drm/amd/amdgpu/psp_v3_1.c | 4 ++--
3 files changed, 13 insertions(+), 4 deletions(-)
diff
Huang, JinHuiEric would like to recall the message, "[PATCH] drm/amdkfd: fix cp
hang in eviction".
___
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
The cp hang occurs in OCL conformance test only on supermicro
platform which has 40 cores and the test generates 40 threads.
The root cause is race condition in non-protected flags.
The fix is to add flags of is_evicted and is_active(init_mqd())
into protected area.
Signed-off-by: Eric Huang
The cp hang occurs in OCL conformance test only on supermicro
platform which has 40 cores and the test generates 40 threads.
The root cause is race condition in non-protected flags.
The fix is to add flags of is_evicted and is_active(init_mqd())
into protected area.
Signed-off-by: Eric Huang
On Wed, 10 Jul 2019 16:31:40 +0200,
Kazlauskas, Nicholas wrote:
>
> On 7/10/19 9:48 AM, Takashi Iwai wrote:
> > On Tue, 09 Jul 2019 18:30:19 +0200,
> > Nicholas Kazlauskas wrote:
> >>
> >> [Why]
> >> The drm_audio_component can be used to give pin ELD notifications
> >> directly to the sound
On 7/10/19 9:48 AM, Takashi Iwai wrote:
> On Tue, 09 Jul 2019 18:30:19 +0200,
> Nicholas Kazlauskas wrote:
>>
>> [Why]
>> The drm_audio_component can be used to give pin ELD notifications
>> directly to the sound driver. This fixes audio endpoints disappearing
>> due to missing unsolicited
On Tue, 09 Jul 2019 18:30:19 +0200,
Nicholas Kazlauskas wrote:
>
> [Why]
> The drm_audio_component can be used to give pin ELD notifications
> directly to the sound driver. This fixes audio endpoints disappearing
> due to missing unsolicited notifications.
>
> [How]
> Send the notification via
在 2019/7/10 3:26, Kuehling, Felix 写道:
> On 2019-07-09 8:58 a.m., Zhou, David(ChunMing) wrote:
>> I've raised it up when Christian make page fault, at that patch,
>> amdgpu_job_submit_direct uses exclusive page fault ring for that.
>>
>> But if you use amdgpu_job_submit_direct for gerneral rings
It doesn't make sense that freeing BO still uses per-vm resv.
I remember when BO is in release list, its resv will be from per-vm resv
copy. Could you check it?
-David
在 2019/7/10 17:29, Emily Deng 写道:
> For vulkan cts allocation test cases, they will create a series of bos, and
> then free
>
On Fri, Jul 05, 2019 at 10:32:20AM -0400, sunpeng...@amd.com wrote:
> From: Leo Li
>
> This can be used to create more descriptive symlinks for MST aux
> devices. Consider the following udev rule:
>
> SUBSYSTEM=="drm_dp_aux_dev", SUBSYSTEMS=="drm", ATTRS{mstpath}=="?*",
>
On Thu, Jul 04, 2019 at 03:05:16PM -0400, sunpeng...@amd.com wrote:
> From: Leo Li
>
> Implement late_register and early_unregister hooks for MST connectors.
> Call drm helpers for MST connector registration, which registers the
> AUX devices.
>
> Signed-off-by: Leo Li
> ---
>
For vulkan cts allocation test cases, they will create a series of bos, and
then free
them. As it has lots of alloction test cases with the same vm, as per vm
bo feature enable, all of those bos' resv are the same. But the bo free is
quite slow,
as they use the same resv object, for every time,
On 2019-07-09 9:00 p.m., Kuehling, Felix wrote:
> On 2019-07-09 6:34 a.m., Michel Dänzer wrote:
>> On 2019-07-09 7:32 a.m., Kuehling, Felix wrote:
>>> This memory allocation flag will be used to indicate BOs containing
>>> sensitive data that should not be leaked to other processes.
>>>
>>>
24 matches
Mail list logo