drm/i915: untie opregion init and asle irq/pipestat enable

Stop calling intel_opregion_enable_asle() and consequently
intel_enable_asle() on opregion init. It should not be necessary for
these reasons:

1) On PCH split platforms, it only enables GSE interrupt, which is
   enabled in irq postinstall anyway. Moreover, the irq enable uses the
   wrong bit on IVB+.

2) On gen 2, it would enable a reserved pipestat bit. If there were gen
   2 systems with opregion asle support, that is. And the gen 2 irq
   handler won't handle it anyway.

3) On gen 3-4, the irq postinstall will call
   intel_opregion_enable_asle() to enable the pipestat.

In short, move the asle irq/pipestat enable responsibility to irq
postinstall, which already happens to be in place.

This should not cause any functional changes, but only do the one line
change here for easier bisectability, just in case, and leave all the
cleanups this allows to followup patches.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

1 files changed, 0 insertions, 2 deletions

diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c
index 5b6d202ec76d..4e697993a3f2 100644
--- a/drivers/gpu/drm/i915/intel_opregion.c
+++ b/drivers/gpu/drm/i915/intel_opregion.c
@@ -427,8 +427,6 @@ void intel_opregion_init(struct drm_device *dev)
 	}
 
 	if (opregion->asle) {
-		intel_opregion_enable_asle(dev);
-
 		iowrite32(ASLE_TCHE_BLC_EN, &opregion->asle->tche);
 		iowrite32(ASLE_ARDY_READY, &opregion->asle->ardy);
 	}