[Intel-gfx] [PATCH 1/2] drm/i915: Flush the PTEs after updating them before suspend
As we use WC updates of the PTE, we are responsible for notifying the hardware when to flush its TLBs. Do so after we zap all the PTEs before suspend (and the BIOS tries to read our GTT). Signed-off-by: Chris Wilson ch...@chris-wilson.co.uk --- drivers/gpu/drm/i915/i915_drv.h | 2 +- drivers/gpu/drm/i915/i915_gem_gtt.c | 14 +- 2 files changed, 14 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index d1513e6ea50d..d0222cb9b7a5 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -2082,7 +2082,7 @@ struct drm_i915_cmd_table { /* Note that the (struct drm_i915_private *) cast is just to shut up gcc. */ #define __I915__(p)((sizeof(*(p)) == sizeof(struct drm_i915_private)) ? \ -(struct drm_i915_private *)(p) : to_i915(p)) +(struct drm_i915_private *)(p) : to_i915((struct drm_device *)p)) #define INTEL_INFO(p) (__I915__(p)-info) #define INTEL_DEVID(p) (INTEL_INFO(p)-device_id) diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index f3fd448505f1..b8d5a5d67a73 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -1316,6 +1316,16 @@ void i915_check_and_clear_faults(struct drm_device *dev) POSTING_READ(RING_FAULT_REG(dev_priv-ring[RCS])); } +static void i915_ggtt_flush(struct drm_i915_private *dev_priv) +{ + if (INTEL_INFO(dev_priv)-gen 6) { + intel_gtt_chipset_flush(); + } else { + I915_WRITE(GFX_FLSH_CNTL_GEN6, GFX_FLSH_CNTL_EN); + POSTING_READ(GFX_FLSH_CNTL_GEN6); + } +} + void i915_gem_suspend_gtt_mappings(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev-dev_private; @@ -1332,6 +1342,8 @@ void i915_gem_suspend_gtt_mappings(struct drm_device *dev) dev_priv-gtt.base.start, dev_priv-gtt.base.total, true); + + i915_ggtt_flush(dev_priv); } void i915_gem_restore_gtt_mappings(struct drm_device *dev) @@ -1384,7 +1396,7 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev) gen6_write_pdes(container_of(vm, struct i915_hw_ppgtt, base)); } - i915_gem_chipset_flush(dev); + i915_ggtt_flush(dev_priv); } int i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj) -- 2.1.0.rc1 ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/2] drm/i915: Flush the PTEs after updating them before suspend
On Wed, Aug 13, 2014 at 11:51:54AM +0100, Chris Wilson wrote: As we use WC updates of the PTE, we are responsible for notifying the hardware when to flush its TLBs. Do so after we zap all the PTEs before suspend (and the BIOS tries to read our GTT). Signed-off-by: Chris Wilson ch...@chris-wilson.co.uk Imo would make more sense to add them to the clear_range functions. -Daniel --- drivers/gpu/drm/i915/i915_drv.h | 2 +- drivers/gpu/drm/i915/i915_gem_gtt.c | 14 +- 2 files changed, 14 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index d1513e6ea50d..d0222cb9b7a5 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -2082,7 +2082,7 @@ struct drm_i915_cmd_table { /* Note that the (struct drm_i915_private *) cast is just to shut up gcc. */ #define __I915__(p) ((sizeof(*(p)) == sizeof(struct drm_i915_private)) ? \ - (struct drm_i915_private *)(p) : to_i915(p)) + (struct drm_i915_private *)(p) : to_i915((struct drm_device *)p)) #define INTEL_INFO(p)(__I915__(p)-info) #define INTEL_DEVID(p) (INTEL_INFO(p)-device_id) diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index f3fd448505f1..b8d5a5d67a73 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -1316,6 +1316,16 @@ void i915_check_and_clear_faults(struct drm_device *dev) POSTING_READ(RING_FAULT_REG(dev_priv-ring[RCS])); } +static void i915_ggtt_flush(struct drm_i915_private *dev_priv) +{ + if (INTEL_INFO(dev_priv)-gen 6) { + intel_gtt_chipset_flush(); + } else { + I915_WRITE(GFX_FLSH_CNTL_GEN6, GFX_FLSH_CNTL_EN); + POSTING_READ(GFX_FLSH_CNTL_GEN6); + } +} + void i915_gem_suspend_gtt_mappings(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev-dev_private; @@ -1332,6 +1342,8 @@ void i915_gem_suspend_gtt_mappings(struct drm_device *dev) dev_priv-gtt.base.start, dev_priv-gtt.base.total, true); + + i915_ggtt_flush(dev_priv); } void i915_gem_restore_gtt_mappings(struct drm_device *dev) @@ -1384,7 +1396,7 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev) gen6_write_pdes(container_of(vm, struct i915_hw_ppgtt, base)); } - i915_gem_chipset_flush(dev); + i915_ggtt_flush(dev_priv); } int i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj) -- 2.1.0.rc1 ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/intel-gfx -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/2] drm/i915: Flush the PTEs after updating them before suspend
On Wed, Aug 13, 2014 at 01:32:08PM +0200, Daniel Vetter wrote: On Wed, Aug 13, 2014 at 11:51:54AM +0100, Chris Wilson wrote: As we use WC updates of the PTE, we are responsible for notifying the hardware when to flush its TLBs. Do so after we zap all the PTEs before suspend (and the BIOS tries to read our GTT). Signed-off-by: Chris Wilson ch...@chris-wilson.co.uk Imo would make more sense to add them to the clear_range functions. They are not required in the clear_range function. Here it is only required in the i915_gem_suspend_gtt_mapping() because a third party is involved. -Chris -- Chris Wilson, Intel Open Source Technology Centre ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/intel-gfx