aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/radeon/r600_blit_kms.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/radeon/r600_blit_kms.c')
-rw-r--r--drivers/gpu/drm/radeon/r600_blit_kms.c361
1 files changed, 152 insertions, 209 deletions
diff --git a/drivers/gpu/drm/radeon/r600_blit_kms.c b/drivers/gpu/drm/radeon/r600_blit_kms.c
index 9aa74c3..e09d281 100644
--- a/drivers/gpu/drm/radeon/r600_blit_kms.c
+++ b/drivers/gpu/drm/radeon/r600_blit_kms.c
@@ -42,6 +42,9 @@
#define COLOR_5_6_5 0x8
#define COLOR_8_8_8_8 0x1a
+#define RECT_UNIT_H 32
+#define RECT_UNIT_W (RADEON_GPU_PAGE_SIZE / 4 / RECT_UNIT_H)
+
/* emits 21 on rv770+, 23 on r600 */
static void
set_render_target(struct radeon_device *rdev, int format,
@@ -54,7 +57,9 @@ set_render_target(struct radeon_device *rdev, int format,
if (h < 8)
h = 8;
- cb_color_info = ((format << 2) | (1 << 27) | (1 << 8));
+ cb_color_info = CB_FORMAT(format) |
+ CB_SOURCE_FORMAT(CB_SF_EXPORT_NORM) |
+ CB_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
pitch = (w / 8) - 1;
slice = ((w * h) / 64) - 1;
@@ -164,9 +169,10 @@ set_vtx_resource(struct radeon_device *rdev, u64 gpu_addr)
{
u32 sq_vtx_constant_word2;
- sq_vtx_constant_word2 = ((upper_32_bits(gpu_addr) & 0xff) | (16 << 8));
+ sq_vtx_constant_word2 = SQ_VTXC_BASE_ADDR_HI(upper_32_bits(gpu_addr) & 0xff) |
+ SQ_VTXC_STRIDE(16);
#ifdef __BIG_ENDIAN
- sq_vtx_constant_word2 |= (2 << 30);
+ sq_vtx_constant_word2 |= SQ_VTXC_ENDIAN_SWAP(SQ_ENDIAN_8IN32);
#endif
radeon_ring_write(rdev, PACKET3(PACKET3_SET_RESOURCE, 7));
@@ -195,25 +201,29 @@ set_vtx_resource(struct radeon_device *rdev, u64 gpu_addr)
static void
set_tex_resource(struct radeon_device *rdev,
int format, int w, int h, int pitch,
- u64 gpu_addr)
+ u64 gpu_addr, u32 size)
{
uint32_t sq_tex_resource_word0, sq_tex_resource_word1, sq_tex_resource_word4;
if (h < 1)
h = 1;
- sq_tex_resource_word0 = (1 << 0) | (1 << 3);
- sq_tex_resource_word0 |= ((((pitch >> 3) - 1) << 8) |
- ((w - 1) << 19));
+ sq_tex_resource_word0 = S_038000_DIM(V_038000_SQ_TEX_DIM_2D) |
+ S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1);
+ sq_tex_resource_word0 |= S_038000_PITCH((pitch >> 3) - 1) |
+ S_038000_TEX_WIDTH(w - 1);
+
+ sq_tex_resource_word1 = S_038004_DATA_FORMAT(format);
+ sq_tex_resource_word1 |= S_038004_TEX_HEIGHT(h - 1);
- sq_tex_resource_word1 = (format << 26);
- sq_tex_resource_word1 |= ((h - 1) << 0);
+ sq_tex_resource_word4 = S_038010_REQUEST_SIZE(1) |
+ S_038010_DST_SEL_X(SQ_SEL_X) |
+ S_038010_DST_SEL_Y(SQ_SEL_Y) |
+ S_038010_DST_SEL_Z(SQ_SEL_Z) |
+ S_038010_DST_SEL_W(SQ_SEL_W);
- sq_tex_resource_word4 = ((1 << 14) |
- (0 << 16) |
- (1 << 19) |
- (2 << 22) |
- (3 << 25));
+ cp_set_surface_sync(rdev,
+ PACKET3_TC_ACTION_ENA, size, gpu_addr);
radeon_ring_write(rdev, PACKET3(PACKET3_SET_RESOURCE, 7));
radeon_ring_write(rdev, 0);
@@ -450,7 +460,7 @@ set_default_state(struct radeon_device *rdev)
radeon_ring_write(rdev, sq_stack_resource_mgmt_2);
}
-static inline uint32_t i2f(uint32_t input)
+static uint32_t i2f(uint32_t input)
{
u32 result, i, exponent, fraction;
@@ -483,6 +493,27 @@ int r600_blit_init(struct radeon_device *rdev)
u32 packet2s[16];
int num_packet2s = 0;
+ rdev->r600_blit.primitives.set_render_target = set_render_target;
+ rdev->r600_blit.primitives.cp_set_surface_sync = cp_set_surface_sync;
+ rdev->r600_blit.primitives.set_shaders = set_shaders;
+ rdev->r600_blit.primitives.set_vtx_resource = set_vtx_resource;
+ rdev->r600_blit.primitives.set_tex_resource = set_tex_resource;
+ rdev->r600_blit.primitives.set_scissors = set_scissors;
+ rdev->r600_blit.primitives.draw_auto = draw_auto;
+ rdev->r600_blit.primitives.set_default_state = set_default_state;
+
+ rdev->r600_blit.ring_size_common = 40; /* shaders + def state */
+ rdev->r600_blit.ring_size_common += 16; /* fence emit for VB IB */
+ rdev->r600_blit.ring_size_common += 5; /* done copy */
+ rdev->r600_blit.ring_size_common += 16; /* fence emit for done copy */
+
+ rdev->r600_blit.ring_size_per_loop = 76;
+ /* set_render_target emits 2 extra dwords on rv6xx */
+ if (rdev->family > CHIP_R600 && rdev->family < CHIP_RV770)
+ rdev->r600_blit.ring_size_per_loop += 2;
+
+ rdev->r600_blit.max_dim = 8192;
+
/* pin copy shader into vram if already initialized */
if (rdev->r600_blit.shader_obj)
goto done;
@@ -600,47 +631,80 @@ static void r600_vb_ib_put(struct radeon_device *rdev)
radeon_ib_free(rdev, &rdev->r600_blit.vb_ib);
}
-int r600_blit_prepare_copy(struct radeon_device *rdev, int size_bytes)
+static unsigned r600_blit_create_rect(unsigned num_gpu_pages,
+ int *width, int *height, int max_dim)
+{
+ unsigned max_pages;
+ unsigned pages = num_gpu_pages;
+ int w, h;
+
+ if (num_gpu_pages == 0) {
+ /* not supposed to be called with no pages, but just in case */
+ h = 0;
+ w = 0;
+ pages = 0;
+ WARN_ON(1);
+ } else {
+ int rect_order = 2;
+ h = RECT_UNIT_H;
+ while (num_gpu_pages / rect_order) {
+ h *= 2;
+ rect_order *= 4;
+ if (h >= max_dim) {
+ h = max_dim;
+ break;
+ }
+ }
+ max_pages = (max_dim * h) / (RECT_UNIT_W * RECT_UNIT_H);
+ if (pages > max_pages)
+ pages = max_pages;
+ w = (pages * RECT_UNIT_W * RECT_UNIT_H) / h;
+ w = (w / RECT_UNIT_W) * RECT_UNIT_W;
+ pages = (w * h) / (RECT_UNIT_W * RECT_UNIT_H);
+ BUG_ON(pages == 0);
+ }
+
+
+ DRM_DEBUG("blit_rectangle: h=%d, w=%d, pages=%d\n", h, w, pages);
+
+ /* return width and height only of the caller wants it */
+ if (height)
+ *height = h;
+ if (width)
+ *width = w;
+
+ return pages;
+}
+
+
+int r600_blit_prepare_copy(struct radeon_device *rdev, unsigned num_gpu_pages)
{
int r;
- int ring_size, line_size;
- int max_size;
- /* loops of emits 64 + fence emit possible */
- int dwords_per_loop = 76, num_loops;
+ int ring_size;
+ int num_loops = 0;
+ int dwords_per_loop = rdev->r600_blit.ring_size_per_loop;
r = r600_vb_ib_get(rdev);
if (r)
return r;
- /* set_render_target emits 2 extra dwords on rv6xx */
- if (rdev->family > CHIP_R600 && rdev->family < CHIP_RV770)
- dwords_per_loop += 2;
-
- /* 8 bpp vs 32 bpp for xfer unit */
- if (size_bytes & 3)
- line_size = 8192;
- else
- line_size = 8192*4;
-
- max_size = 8192 * line_size;
+ /* num loops */
+ while (num_gpu_pages) {
+ num_gpu_pages -=
+ r600_blit_create_rect(num_gpu_pages, NULL, NULL,
+ rdev->r600_blit.max_dim);
+ num_loops++;
+ }
- /* major loops cover the max size transfer */
- num_loops = ((size_bytes + max_size) / max_size);
- /* minor loops cover the extra non aligned bits */
- num_loops += ((size_bytes % line_size) ? 1 : 0);
/* calculate number of loops correctly */
ring_size = num_loops * dwords_per_loop;
- /* set default + shaders */
- ring_size += 40; /* shaders + def state */
- ring_size += 10; /* fence emit for VB IB */
- ring_size += 5; /* done copy */
- ring_size += 10; /* fence emit for done copy */
+ ring_size += rdev->r600_blit.ring_size_common;
r = radeon_ring_lock(rdev, ring_size);
if (r)
return r;
- set_default_state(rdev); /* 14 */
- set_shaders(rdev); /* 26 */
+ rdev->r600_blit.primitives.set_default_state(rdev);
+ rdev->r600_blit.primitives.set_shaders(rdev);
return 0;
}
@@ -659,182 +723,61 @@ void r600_blit_done_copy(struct radeon_device *rdev, struct radeon_fence *fence)
void r600_kms_blit_copy(struct radeon_device *rdev,
u64 src_gpu_addr, u64 dst_gpu_addr,
- int size_bytes)
+ unsigned num_gpu_pages)
{
- int max_bytes;
u64 vb_gpu_addr;
u32 *vb;
- DRM_DEBUG("emitting copy %16llx %16llx %d %d\n", src_gpu_addr, dst_gpu_addr,
- size_bytes, rdev->r600_blit.vb_used);
+ DRM_DEBUG("emitting copy %16llx %16llx %d %d\n",
+ src_gpu_addr, dst_gpu_addr,
+ num_gpu_pages, rdev->r600_blit.vb_used);
vb = (u32 *)(rdev->r600_blit.vb_ib->ptr + rdev->r600_blit.vb_used);
- if ((size_bytes & 3) || (src_gpu_addr & 3) || (dst_gpu_addr & 3)) {
- max_bytes = 8192;
-
- while (size_bytes) {
- int cur_size = size_bytes;
- int src_x = src_gpu_addr & 255;
- int dst_x = dst_gpu_addr & 255;
- int h = 1;
- src_gpu_addr = src_gpu_addr & ~255ULL;
- dst_gpu_addr = dst_gpu_addr & ~255ULL;
-
- if (!src_x && !dst_x) {
- h = (cur_size / max_bytes);
- if (h > 8192)
- h = 8192;
- if (h == 0)
- h = 1;
- else
- cur_size = max_bytes;
- } else {
- if (cur_size > max_bytes)
- cur_size = max_bytes;
- if (cur_size > (max_bytes - dst_x))
- cur_size = (max_bytes - dst_x);
- if (cur_size > (max_bytes - src_x))
- cur_size = (max_bytes - src_x);
- }
-
- if ((rdev->r600_blit.vb_used + 48) > rdev->r600_blit.vb_total) {
- WARN_ON(1);
- }
-
- vb[0] = i2f(dst_x);
- vb[1] = 0;
- vb[2] = i2f(src_x);
- vb[3] = 0;
-
- vb[4] = i2f(dst_x);
- vb[5] = i2f(h);
- vb[6] = i2f(src_x);
- vb[7] = i2f(h);
-
- vb[8] = i2f(dst_x + cur_size);
- vb[9] = i2f(h);
- vb[10] = i2f(src_x + cur_size);
- vb[11] = i2f(h);
-
- /* src 9 */
- set_tex_resource(rdev, FMT_8,
- src_x + cur_size, h, src_x + cur_size,
- src_gpu_addr);
-
- /* 5 */
- cp_set_surface_sync(rdev,
- PACKET3_TC_ACTION_ENA, (src_x + cur_size * h), src_gpu_addr);
- /* dst 23 */
- set_render_target(rdev, COLOR_8,
- dst_x + cur_size, h,
- dst_gpu_addr);
+ while (num_gpu_pages) {
+ int w, h;
+ unsigned size_in_bytes;
+ unsigned pages_per_loop =
+ r600_blit_create_rect(num_gpu_pages, &w, &h,
+ rdev->r600_blit.max_dim);
- /* scissors 12 */
- set_scissors(rdev, dst_x, 0, dst_x + cur_size, h);
+ size_in_bytes = pages_per_loop * RADEON_GPU_PAGE_SIZE;
+ DRM_DEBUG("rectangle w=%d h=%d\n", w, h);
- /* 14 */
- vb_gpu_addr = rdev->r600_blit.vb_ib->gpu_addr + rdev->r600_blit.vb_used;
- set_vtx_resource(rdev, vb_gpu_addr);
-
- /* draw 10 */
- draw_auto(rdev);
-
- /* 5 */
- cp_set_surface_sync(rdev,
- PACKET3_CB_ACTION_ENA | PACKET3_CB0_DEST_BASE_ENA,
- cur_size * h, dst_gpu_addr);
-
- vb += 12;
- rdev->r600_blit.vb_used += 12 * 4;
-
- src_gpu_addr += cur_size * h;
- dst_gpu_addr += cur_size * h;
- size_bytes -= cur_size * h;
+ if ((rdev->r600_blit.vb_used + 48) > rdev->r600_blit.vb_total) {
+ WARN_ON(1);
}
- } else {
- max_bytes = 8192 * 4;
-
- while (size_bytes) {
- int cur_size = size_bytes;
- int src_x = (src_gpu_addr & 255);
- int dst_x = (dst_gpu_addr & 255);
- int h = 1;
- src_gpu_addr = src_gpu_addr & ~255ULL;
- dst_gpu_addr = dst_gpu_addr & ~255ULL;
-
- if (!src_x && !dst_x) {
- h = (cur_size / max_bytes);
- if (h > 8192)
- h = 8192;
- if (h == 0)
- h = 1;
- else
- cur_size = max_bytes;
- } else {
- if (cur_size > max_bytes)
- cur_size = max_bytes;
- if (cur_size > (max_bytes - dst_x))
- cur_size = (max_bytes - dst_x);
- if (cur_size > (max_bytes - src_x))
- cur_size = (max_bytes - src_x);
- }
-
- if ((rdev->r600_blit.vb_used + 48) > rdev->r600_blit.vb_total) {
- WARN_ON(1);
- }
- vb[0] = i2f(dst_x / 4);
- vb[1] = 0;
- vb[2] = i2f(src_x / 4);
- vb[3] = 0;
-
- vb[4] = i2f(dst_x / 4);
- vb[5] = i2f(h);
- vb[6] = i2f(src_x / 4);
- vb[7] = i2f(h);
-
- vb[8] = i2f((dst_x + cur_size) / 4);
- vb[9] = i2f(h);
- vb[10] = i2f((src_x + cur_size) / 4);
- vb[11] = i2f(h);
-
- /* src 9 */
- set_tex_resource(rdev, FMT_8_8_8_8,
- (src_x + cur_size) / 4,
- h, (src_x + cur_size) / 4,
- src_gpu_addr);
- /* 5 */
- cp_set_surface_sync(rdev,
- PACKET3_TC_ACTION_ENA, (src_x + cur_size * h), src_gpu_addr);
-
- /* dst 23 */
- set_render_target(rdev, COLOR_8_8_8_8,
- (dst_x + cur_size) / 4, h,
- dst_gpu_addr);
-
- /* scissors 12 */
- set_scissors(rdev, (dst_x / 4), 0, (dst_x + cur_size / 4), h);
-
- /* Vertex buffer setup 14 */
- vb_gpu_addr = rdev->r600_blit.vb_ib->gpu_addr + rdev->r600_blit.vb_used;
- set_vtx_resource(rdev, vb_gpu_addr);
-
- /* draw 10 */
- draw_auto(rdev);
-
- /* 5 */
- cp_set_surface_sync(rdev,
- PACKET3_CB_ACTION_ENA | PACKET3_CB0_DEST_BASE_ENA,
- cur_size * h, dst_gpu_addr);
-
- /* 78 ring dwords per loop */
- vb += 12;
- rdev->r600_blit.vb_used += 12 * 4;
-
- src_gpu_addr += cur_size * h;
- dst_gpu_addr += cur_size * h;
- size_bytes -= cur_size * h;
- }
+ vb[0] = 0;
+ vb[1] = 0;
+ vb[2] = 0;
+ vb[3] = 0;
+
+ vb[4] = 0;
+ vb[5] = i2f(h);
+ vb[6] = 0;
+ vb[7] = i2f(h);
+
+ vb[8] = i2f(w);
+ vb[9] = i2f(h);
+ vb[10] = i2f(w);
+ vb[11] = i2f(h);
+
+ rdev->r600_blit.primitives.set_tex_resource(rdev, FMT_8_8_8_8,
+ w, h, w, src_gpu_addr, size_in_bytes);
+ rdev->r600_blit.primitives.set_render_target(rdev, COLOR_8_8_8_8,
+ w, h, dst_gpu_addr);
+ rdev->r600_blit.primitives.set_scissors(rdev, 0, 0, w, h);
+ vb_gpu_addr = rdev->r600_blit.vb_ib->gpu_addr + rdev->r600_blit.vb_used;
+ rdev->r600_blit.primitives.set_vtx_resource(rdev, vb_gpu_addr);
+ rdev->r600_blit.primitives.draw_auto(rdev);
+ rdev->r600_blit.primitives.cp_set_surface_sync(rdev,
+ PACKET3_CB_ACTION_ENA | PACKET3_CB0_DEST_BASE_ENA,
+ size_in_bytes, dst_gpu_addr);
+
+ vb += 12;
+ rdev->r600_blit.vb_used += 4*12;
+ src_gpu_addr += size_in_bytes;
+ dst_gpu_addr += size_in_bytes;
+ num_gpu_pages -= pages_per_loop;
}
}
-
generated by cgit v1.1 at 2024-04-25 01:05:12 +0000