提交 27dcfc10 编写于 作者: A Alex Deucher 提交者: Dave Airlie

drm/radeon/kms: use linear aligned for evergreen/ni bo blits

Not only is linear aligned supposedly more performant,
linear general is only supported by the CB in single
slice mode.  The texture hardware doesn't support
linear general, but I think the hw automatically
upgrades it to linear aligned.
Signed-off-by: NAlex Deucher <alexdeucher@gmail.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>
上级 1ea9dbf2
...@@ -55,7 +55,7 @@ set_render_target(struct radeon_device *rdev, int format, ...@@ -55,7 +55,7 @@ set_render_target(struct radeon_device *rdev, int format,
if (h < 8) if (h < 8)
h = 8; h = 8;
cb_color_info = ((format << 2) | (1 << 24)); cb_color_info = ((format << 2) | (1 << 24) | (1 << 8));
pitch = (w / 8) - 1; pitch = (w / 8) - 1;
slice = ((w * h) / 64) - 1; slice = ((w * h) / 64) - 1;
...@@ -176,7 +176,7 @@ set_tex_resource(struct radeon_device *rdev, ...@@ -176,7 +176,7 @@ set_tex_resource(struct radeon_device *rdev,
sq_tex_resource_word0 = (1 << 0); /* 2D */ sq_tex_resource_word0 = (1 << 0); /* 2D */
sq_tex_resource_word0 |= ((((pitch >> 3) - 1) << 6) | sq_tex_resource_word0 |= ((((pitch >> 3) - 1) << 6) |
((w - 1) << 18)); ((w - 1) << 18));
sq_tex_resource_word1 = ((h - 1) << 0); sq_tex_resource_word1 = ((h - 1) << 0) | (1 << 28);
/* xyzw swizzles */ /* xyzw swizzles */
sq_tex_resource_word4 = (0 << 16) | (1 << 19) | (2 << 22) | (3 << 25); sq_tex_resource_word4 = (0 << 16) | (1 << 19) | (2 << 22) | (3 << 25);
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册