[Intel-gfx] [PATCH 1/3] drm/i915: Create dumb buffer from LMEM

[Ramalingam C]

On 2019-09-26 at 09:55:16 +0100, Tvrtko Ursulin wrote:
> 
> On 26/09/2019 06:21, Ramalingam C wrote:
> > When LMEM is supported, dumb buffer preferred to be created from LMEM.
> > 
> > This is developed on top of v3 LMEM series
> > https://patchwork.freedesktop.org/series/56683/.
> > 
> > v2:
> >    Parameters are reshuffled. [Chris]
> > 
> > Signed-off-by: Ramalingam C <ramalingam.c at intel.com>
> > cc: Matthew Auld <matthew.auld at intel.com>
> > ---
> >   drivers/gpu/drm/i915/i915_gem.c | 18 +++++++++++++++---
> >   1 file changed, 15 insertions(+), 3 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> > index e458507b1558..6810a549ee98 100644
> > --- a/drivers/gpu/drm/i915/i915_gem.c
> > +++ b/drivers/gpu/drm/i915/i915_gem.c
> > @@ -44,6 +44,7 @@
> >   #include "gem/i915_gem_clflush.h"
> >   #include "gem/i915_gem_context.h"
> >   #include "gem/i915_gem_ioctls.h"
> > +#include "gem/i915_gem_lmem.h"
> >   #include "gem/i915_gem_pm.h"
> >   #include "gt/intel_engine_user.h"
> >   #include "gt/intel_gt.h"
> > @@ -160,6 +161,7 @@ i915_gem_phys_pwrite(struct drm_i915_gem_object *obj,
> >   static int
> >   i915_gem_create(struct drm_file *file,
> >   		struct drm_i915_private *dev_priv,
> > +		enum intel_region_id mem_region,
> >   		u64 *size_p,
> >   		u32 *handle_p)
> >   {
> > @@ -173,7 +175,12 @@ i915_gem_create(struct drm_file *file,
> >   		return -EINVAL;
> >   	/* Allocate the new object */
> > -	obj = i915_gem_object_create_shmem(dev_priv, size);
> > +	if (mem_region == INTEL_MEMORY_LMEM)
> > +		obj = i915_gem_object_create_lmem(dev_priv, size, 0);
> > +	else if (mem_region == INTEL_MEMORY_STOLEN)
> > +		obj = i915_gem_object_create_stolen(dev_priv, size);
> > +	else
> > +		obj = i915_gem_object_create_shmem(dev_priv, size);
> >   	if (IS_ERR(obj))
> >   		return PTR_ERR(obj);
> > @@ -193,6 +200,7 @@ i915_gem_dumb_create(struct drm_file *file,
> >   		     struct drm_device *dev,
> >   		     struct drm_mode_create_dumb *args)
> >   {
> > +	enum intel_region_id mem_region = INTEL_MEMORY_UKNOWN;
> >   	int cpp = DIV_ROUND_UP(args->bpp, 8);
> >   	u32 format;
> > @@ -219,7 +227,11 @@ i915_gem_dumb_create(struct drm_file *file,
> >   		args->pitch = ALIGN(args->pitch, 4096);
> >   	args->size = args->pitch * args->height;
> > -	return i915_gem_create(file, to_i915(dev),
> > +
> > +	if (HAS_LMEM(to_i915(dev)))
> > +		mem_region = INTEL_MEMORY_LMEM;
> > +
> > +	return i915_gem_create(file, to_i915(dev), mem_region,
> >   			       &args->size, &args->handle);
> >   }
> > @@ -238,7 +250,7 @@ i915_gem_create_ioctl(struct drm_device *dev, void *data,
> >   	i915_gem_flush_free_objects(dev_priv);
> > -	return i915_gem_create(file, dev_priv,
> > +	return i915_gem_create(file, dev_priv, INTEL_MEMORY_UKNOWN,
> >   			       &args->size, &args->handle);
> 
> We don't have shmem memory region? Or default? Or is unknown supposed to
> mean default for a given platform?
I take this as default. Not sure if we need to pass anything (default
region) specifically.

-Ram

> 
> Regards,
> 
> Tvrtko
> 
> >   }
> >