On 03/30/2019 03:14 AM, Chris Wilson wrote:
> Quoting Xiaolin Zhang (2019-03-29 13:32:40)
>> + spin_lock(&engine->i915->vgpu.shared_page_lock);
>> + shared_page->ring_id = engine->id;
>> + for (n = 0; n < execlists_num_ports(execlists); n++)
>> + shared_page->descs[n
On 03/29/2019 11:40 PM, Chris Wilson wrote:
> Quoting Xiaolin Zhang (2019-03-29 13:32:40)
>> diff --git a/drivers/gpu/drm/i915/i915_irq.c
>> b/drivers/gpu/drm/i915/i915_irq.c
>> index 2f78829..28e8ee0 100644
>> --- a/drivers/gpu/drm/i915/i915_irq.c
>> +++ b/drivers/gpu/drm/i915/i915_irq.c
>> @@ -3
Quoting Xiaolin Zhang (2019-03-29 13:32:40)
> + spin_lock(&engine->i915->vgpu.shared_page_lock);
> + shared_page->ring_id = engine->id;
> + for (n = 0; n < execlists_num_ports(execlists); n++)
> + shared_page->descs[n] = descs[n];
> +
> + __raw_i915_write32(unc
Quoting Xiaolin Zhang (2019-03-29 13:32:40)
> diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
> index 2f78829..28e8ee0 100644
> --- a/drivers/gpu/drm/i915/i915_irq.c
> +++ b/drivers/gpu/drm/i915/i915_irq.c
> @@ -37,6 +37,7 @@
> #include "i915_drv.h"
> #include "i915
Quoting Xiaolin Zhang (2019-03-29 13:32:40)
> It is performance optimization to override the actual submisison backend
> in order to eliminate execlists csb process and reduce mmio trap numbers
> for workload submission without contextswith interrupt by talking with
> GVT via PV submisison notifica
It is performance optimization to override the actual submisison backend
in order to eliminate execlists csb process and reduce mmio trap numbers
for workload submission without contextswith interrupt by talking with
GVT via PV submisison notification mechanism between guest and GVT.
Use PV_SUBMIS
