[Intel-gfx] [PATCH] [RFC] drm/i915: Handle HPD when it has actually occurred

[Ville Syrjälä]

On Mon, Jul 06, 2015 at 02:31:56PM +0530, Jindal, Sonika wrote:
> 
> 
> On 7/6/2015 2:19 PM, Ville Syrjälä wrote:
> > On Mon, Jul 06, 2015 at 11:23:53AM +0530, Sonika Jindal wrote:
> >> Writing to PCH_PORT_HOTPLUG for each interrupt is not required.
> >> Handle it only if hpd has actually occurred like we handle other
> >> interrupts.
> >>
> >> Signed-off-by: Sonika Jindal <sonika.jindal at intel.com>
> >> ---
> >> Hi,
> >>
> >> I see we don't check for hotplug_trigger before processing the HPD for any of the platform. Is there any reason for this?
> >> For SKL, if I let write to PCH_PORT_HOTPLUG happen for all interrupts, somehow this register gets an invalid value at one point and it zeroes it out.
> >> If I put this check before handling HPD, hotplug behaves fine.
> >> Please let me know if you see any issue with this approach.
> >>
> >> Thanks,
> >> Sonika
> >>
> >>   drivers/gpu/drm/i915/i915_irq.c |   11 ++++++-----
> >>   1 file changed, 6 insertions(+), 5 deletions(-)
> >>
> >> diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
> >> index a6fbe64..2d47372 100644
> >> --- a/drivers/gpu/drm/i915/i915_irq.c
> >> +++ b/drivers/gpu/drm/i915/i915_irq.c
> >> @@ -1760,11 +1760,12 @@ static void cpt_irq_handler(struct drm_device *dev, u32 pch_iir)
> >>   	u32 dig_hotplug_reg;
> >>   	u32 pin_mask, long_mask;
> >>
> >> -	dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG);
> >> -	I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg);
> >> -
> >> -	pch_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, dig_hotplug_reg, hpd_cpt);
> >> -	intel_hpd_irq_handler(dev, pin_mask, long_mask);
> >> +	if (hotplug_trigger) {
> >> +		dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG);
> >> +		I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg);
> >> +		pch_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, dig_hotplug_reg, hpd_cpt);
> >> +		intel_hpd_irq_handler(dev, pin_mask, long_mask);
> >> +	}
> >
> > Deja vu. I think I have the same patch (also for IBX) in my stalled
> > (and never posted) port A HPD branch. So yeah, I think this
> > makes sense.
> >
> > You can also move the dig_hotplug_reg, pin_mask, and long_mask declarations
> > into the if block since they're not needed elsewhere.
> Sure. Do you think I should add this for ibx as well?

Yes.

-- 
Ville Syrjälä
Intel OTC