Re: [PATCH libdrm] intel: Add EXEC_OBJECT_SUPPORTS_48BADDRESS flag.

[Dave Gordon <david.s.gordon@intel.com>]

On 23/06/15 13:21, Michel Thierry wrote:
> Gen8+ supports 48-bit virtual addresses, but some objects must always be
> allocated inside the 32-bit address range.
> 
> In specific, any resource used with flat/heapless (0x00000000-0xfffff000)
> General State Heap (GSH) or Intruction State Heap (ISH) must be in a
> 32-bit range, because the General State Offset and Instruction State Offset
> are limited to 32-bits.
> 
> Provide a flag to set when the 4GB limit is not necessary in a given bo.
> 48-bit range will only be used when explicitly requested.
> 
> Cc: dri-devel@lists.freedesktop.org
> Signed-off-by: Michel Thierry <michel.thierry@intel.com>
> ---
>  include/drm/i915_drm.h    |  3 ++-
>  intel/intel_bufmgr.c      | 12 ++++++++++++
>  intel/intel_bufmgr.h      |  2 ++
>  intel/intel_bufmgr_gem.c  | 48 +++++++++++++++++++++++++++++++++++++++++++----
>  intel/intel_bufmgr_priv.h | 11 +++++++++++
>  5 files changed, 71 insertions(+), 5 deletions(-)
> 
> diff --git a/include/drm/i915_drm.h b/include/drm/i915_drm.h
> index ded43b1..d6fb99d 100644
> --- a/include/drm/i915_drm.h
> +++ b/include/drm/i915_drm.h
> @@ -680,7 +680,8 @@ struct drm_i915_gem_exec_object2 {
>  #define EXEC_OBJECT_NEEDS_FENCE (1<<0)
>  #define EXEC_OBJECT_NEEDS_GTT	(1<<1)
>  #define EXEC_OBJECT_WRITE	(1<<2)
> -#define __EXEC_OBJECT_UNKNOWN_FLAGS -(EXEC_OBJECT_WRITE<<1)
> +#define EXEC_OBJECT_SUPPORTS_48BADDRESS (1<<3)

Please change the name, so you don't concatenate the "B" of "48B" with
the "A" of "ADDRESS", making something which appears to relate to a "BAD
DRESS". .._48B_ADDRESS.. and corresponding lowercase variants would be
fine :)

.Dave.

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx