On 2020-09-09 10:06, Zhenyu Wang wrote:
On 2020.09.09 09:43:21 +0800, Colin Xu wrote:
I tested this patch on the suspend/resume case with vGPU created (no need
really activate), can still observer the system freeze issue as mentioned in
another patch I sent. So I suppose we still need decouple
On 2020.09.09 09:43:21 +0800, Colin Xu wrote:
> I tested this patch on the suspend/resume case with vGPU created (no need
> really activate), can still observer the system freeze issue as mentioned in
> another patch I sent. So I suppose we still need decouple context pin/unpin
> with submission se
I tested this patch on the suspend/resume case with vGPU created (no
need really activate), can still observer the system freeze issue as
mentioned in another patch I sent. So I suppose we still need decouple
context pin/unpin with submission setup/clean, but move to workload
create/destroy?
On 2020.09.07 23:02:03 +0300, Zhi Wang wrote:
> To support ww locking and per-object implemented in i915, GVT scheduler needs
> to be refined. Most of the changes are located in shadow batch buffer, shadow
> wa context in GVT-g, where use quite a lot of i915 gem object APIs.
>
> Cc: Maarten Lankho
To support ww locking and per-object implemented in i915, GVT scheduler needs
to be refined. Most of the changes are located in shadow batch buffer, shadow
wa context in GVT-g, where use quite a lot of i915 gem object APIs.
Cc: Maarten Lankhorst
Cc: Joonas Lahtinen
Cc: Zhenyu Wang
Signed-off-by
