[Intel-gfx] [PULL] gvt-fixes

[Jani Nikula]

On Tue, 23 Jun 2020, Zhenyu Wang <zhenyuw at linux.intel.com> wrote:
> On 2020.06.18 15:48:01 +0300, Joonas Lahtinen wrote:
>> Quoting Zhenyu Wang (2020-06-17 07:34:18)
>> > 
>> > Hi,
>> > 
>> > This contains misc fixes for gvt. Two MMIO handler fixes on SKL/CFL,
>> > one mask register bit checking fix exposed in suspend/resume path and
>> > one lockdep error fix for debugfs entry access.
>> 
>> Could not pull this one due to the extra hassle with CI this week.
>> 
>> Jani, can you please pull this next week.
>> 
>
> Got it. Please help to pull then.

Pulled, thanks.

BR,
Jani.


>
> One thing I forgot to mention that change in "drm/i915/gvt: Fix incorrect check of enabled bits in mask registers"
> would cause a minor conflict if backmerging from linux master to dinq, which
> is because of new IS_COMETLAKE. Change like below could resolve that.
>
> diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
> index 9f48db0bf9d5..78ba2857144e 100644
> --- a/drivers/gpu/drm/i915/gvt/handlers.c
> +++ b/drivers/gpu/drm/i915/gvt/handlers.c
> @@ -1734,14 +1734,9 @@ static int ring_mode_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
>  		return 0;
>  	}
>  
> -<<<<<<< HEAD
>  	if ((IS_COFFEELAKE(vgpu->gvt->gt->i915) ||
>  	     IS_COMETLAKE(vgpu->gvt->gt->i915)) &&
> -	    data & _MASKED_BIT_ENABLE(2)) {
> -=======
> -	if (IS_COFFEELAKE(vgpu->gvt->gt->i915) &&
>  	    IS_MASKED_BITS_ENABLED(data, 2)) {
> ->>>>>>> origin/gvt-next-fixes
>  		enter_failsafe_mode(vgpu, GVT_FAILSAFE_UNSUPPORTED_GUEST);
>  		return 0;
>  	}
>
>> 
>> > Thanks.
>> > --
>> > The following changes since commit 8e68c6340d5833077b3753eabedab40755571383:
>> > 
>> >   drm/i915/display: Fix the encoder type check (2020-06-16 11:34:24 +0300)
>> > 
>> > are available in the Git repository at:
>> > 
>> >   https://github.com/intel/gvt-linux tags/gvt-fixes-2020-06-17
>> > 
>> > for you to fetch changes up to a291e4fba259a56a6a274c1989997acb6f0bb03a:
>> > 
>> >   drm/i915/gvt: Use GFP_ATOMIC instead of GFP_KERNEL in atomic context (2020-06-17 12:36:19 +0800)
>> > 
>> > ----------------------------------------------------------------
>> > gvt-fixes-2020-06-17
>> > 
>> > - Two missed MMIO handler fixes for SKL/CFL (Colin)
>> > - Fix mask register bits check (Colin)
>> > - Fix one lockdep error for debugfs entry access (Colin)
>> > 
>> > ----------------------------------------------------------------
>> > Colin Xu (4):
>> >       drm/i915/gvt: Add one missing MMIO handler for D_SKL_PLUS
>> >       drm/i915/gvt: Fix two CFL MMIO handling caused by regression.
>> >       drm/i915/gvt: Fix incorrect check of enabled bits in mask registers
>> >       drm/i915/gvt: Use GFP_ATOMIC instead of GFP_KERNEL in atomic context
>> > 
>> >  drivers/gpu/drm/i915/gvt/debugfs.c      |  2 +-
>> >  drivers/gpu/drm/i915/gvt/handlers.c     | 24 +++++++++++++-----------
>> >  drivers/gpu/drm/i915/gvt/mmio_context.h |  6 +++---
>> >  drivers/gpu/drm/i915/gvt/reg.h          |  5 +++++
>> >  4 files changed, 22 insertions(+), 15 deletions(-)
>> _______________________________________________
>> intel-gvt-dev mailing list
>> intel-gvt-dev at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/intel-gvt-dev

-- 
Jani Nikula, Intel Open Source Graphics Center