From: Ilija Hadzic Date: Thu, 13 Oct 2011 03:29:34 +0000 (-0400) Subject: drm/radeon/kms: simplify evergreen blit code X-Git-Url: https://git.karo-electronics.de/?a=commitdiff_plain;h=eb32d0c34e64666b5f9d9c040ac85d96ecd1e6ee;p=linux-beck.git drm/radeon/kms: simplify evergreen blit code Covert 4k pages to multiples of 64x64x4 tiles. This is also more efficient than a scanline based approach from the MC's perspective. Signed-off-by: Alex Deucher Signed-off-by: Ilija Hadzic Signed-off-by: Dave Airlie --- diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index 562f9a3a69f1..7cd402412054 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -3180,14 +3180,14 @@ int evergreen_copy_blit(struct radeon_device *rdev, mutex_lock(&rdev->r600_blit.mutex); rdev->r600_blit.vb_ib = NULL; - r = evergreen_blit_prepare_copy(rdev, num_pages * RADEON_GPU_PAGE_SIZE); + r = evergreen_blit_prepare_copy(rdev, num_pages); if (r) { if (rdev->r600_blit.vb_ib) radeon_ib_free(rdev, &rdev->r600_blit.vb_ib); mutex_unlock(&rdev->r600_blit.mutex); return r; } - evergreen_kms_blit_copy(rdev, src_offset, dst_offset, num_pages * RADEON_GPU_PAGE_SIZE); + evergreen_kms_blit_copy(rdev, src_offset, dst_offset, num_pages); evergreen_blit_done_copy(rdev, fence); mutex_unlock(&rdev->r600_blit.mutex); return 0; diff --git a/drivers/gpu/drm/radeon/evergreen_blit_kms.c b/drivers/gpu/drm/radeon/evergreen_blit_kms.c index 7eb78b3b30b7..f073dc9af116 100644 --- a/drivers/gpu/drm/radeon/evergreen_blit_kms.c +++ b/drivers/gpu/drm/radeon/evergreen_blit_kms.c @@ -44,6 +44,10 @@ #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) +#define MAX_RECT_DIM 16384 + /* emits 17 */ static void set_render_target(struct radeon_device *rdev, int format, @@ -56,7 +60,7 @@ set_render_target(struct radeon_device *rdev, int format, if (h < 8) h = 8; - cb_color_info = ((format << 2) | (1 << 24) | (1 << 8)); + cb_color_info = ((format << 2) | (1 << 24) | (2 << 8)); pitch = (w / 8) - 1; slice = ((w * h) / 64) - 1; @@ -67,7 +71,7 @@ set_render_target(struct radeon_device *rdev, int format, radeon_ring_write(rdev, slice); radeon_ring_write(rdev, 0); radeon_ring_write(rdev, cb_color_info); - radeon_ring_write(rdev, (1 << 4)); + radeon_ring_write(rdev, 0); radeon_ring_write(rdev, (w - 1) | ((h - 1) << 16)); radeon_ring_write(rdev, 0); radeon_ring_write(rdev, 0); @@ -179,7 +183,7 @@ set_tex_resource(struct radeon_device *rdev, sq_tex_resource_word0 = (1 << 0); /* 2D */ sq_tex_resource_word0 |= ((((pitch >> 3) - 1) << 6) | ((w - 1) << 18)); - sq_tex_resource_word1 = ((h - 1) << 0) | (1 << 28); + sq_tex_resource_word1 = ((h - 1) << 0) | (2 << 28); /* xyzw swizzles */ sq_tex_resource_word4 = (0 << 16) | (1 << 19) | (2 << 22) | (3 << 25); @@ -751,30 +755,80 @@ static void evergreen_vb_ib_put(struct radeon_device *rdev) radeon_ib_free(rdev, &rdev->r600_blit.vb_ib); } -int evergreen_blit_prepare_copy(struct radeon_device *rdev, int size_bytes) + +/* maps the rectangle to the buffer so that satisfies the following properties: + * - dimensions are less or equal to the hardware limit (MAX_RECT_DIM) + * - rectangle consists of integer number of pages + * - height is an integer multiple of RECT_UNIT_H + * - width is an integer multiple of RECT_UNIT_W + * - (the above three conditions also guarantee tile-aligned size) + * - it is as square as possible (sides ratio never greater than 2:1) + * - uses maximum number of pages that fit the above constraints + * + * input: buffer size, pointers to width/height variables + * return: number of pages that were successfully mapped to the rectangle + * width/height of the rectangle + */ +static unsigned evergreen_blit_create_rect(unsigned num_pages, int *width, int *height) +{ + unsigned max_pages; + unsigned pages = num_pages; + int w, h; + + if (num_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_pages / rect_order) { + h *= 2; + rect_order *= 4; + if (h >= MAX_RECT_DIM) { + h = MAX_RECT_DIM; + break; + } + } + max_pages = (MAX_RECT_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 evergreen_blit_prepare_copy(struct radeon_device *rdev, unsigned num_pages) { int r; - int ring_size, line_size; - int max_size; + int ring_size; /* loops of emits + fence emit possible */ - int dwords_per_loop = 74, num_loops; + int dwords_per_loop = 74, num_loops = 0; r = evergreen_vb_ib_get(rdev); if (r) return r; - /* 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; - - /* 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); + /* num loops */ + while (num_pages) { + num_pages -= evergreen_blit_create_rect(num_pages, NULL, NULL); + num_loops++; + } /* calculate number of loops correctly */ ring_size = num_loops * dwords_per_loop; /* set default + shaders */ @@ -806,183 +860,72 @@ void evergreen_blit_done_copy(struct radeon_device *rdev, struct radeon_fence *f void evergreen_kms_blit_copy(struct radeon_device *rdev, u64 src_gpu_addr, u64 dst_gpu_addr, - int size_bytes) + unsigned num_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); + num_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); - } + while (num_pages) { + int w, h; + unsigned size_in_bytes; + unsigned pages_per_loop = evergreen_blit_create_rect(num_pages, &w, &h); - vb[0] = i2f(dst_x); - vb[1] = 0; - vb[2] = i2f(src_x); - vb[3] = 0; + size_in_bytes = pages_per_loop * RADEON_GPU_PAGE_SIZE; + DRM_DEBUG("rectangle w=%d h=%d\n", w, h); - 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 10 */ - 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); + if ((rdev->r600_blit.vb_used + 48) > rdev->r600_blit.vb_total) { + WARN_ON(1); + } + vb[0] = 0; + vb[1] = 0; + vb[2] = 0; + vb[3] = 0; - /* dst 17 */ - set_render_target(rdev, COLOR_8, - dst_x + cur_size, h, - dst_gpu_addr); + vb[4] = 0; + vb[5] = i2f(h); + vb[6] = 0; + vb[7] = i2f(h); - /* scissors 12 */ - set_scissors(rdev, dst_x, 0, dst_x + cur_size, h); + vb[8] = i2f(w); + vb[9] = i2f(h); + vb[10] = i2f(w); + vb[11] = i2f(h); - /* 15 */ - vb_gpu_addr = rdev->r600_blit.vb_ib->gpu_addr + rdev->r600_blit.vb_used; - set_vtx_resource(rdev, vb_gpu_addr); + /* src 10 */ + set_tex_resource(rdev, FMT_8_8_8_8, w, h, w, src_gpu_addr); - /* draw 10 */ - draw_auto(rdev); + /* 5 */ + cp_set_surface_sync(rdev, + PACKET3_TC_ACTION_ENA, size_in_bytes, src_gpu_addr); - /* 5 */ - cp_set_surface_sync(rdev, - PACKET3_CB_ACTION_ENA | PACKET3_CB0_DEST_BASE_ENA, - cur_size * h, dst_gpu_addr); + /* dst 17 */ + set_render_target(rdev, COLOR_8_8_8_8, w, h, dst_gpu_addr); - vb += 12; - rdev->r600_blit.vb_used += 12 * 4; + /* scissors 12 */ + set_scissors(rdev, 0, 0, w, h); - src_gpu_addr += cur_size * h; - dst_gpu_addr += cur_size * h; - size_bytes -= cur_size * h; - } - } 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); - } + /* Vertex buffer setup 15 */ + vb_gpu_addr = rdev->r600_blit.vb_ib->gpu_addr + rdev->r600_blit.vb_used; + set_vtx_resource(rdev, vb_gpu_addr); - if ((rdev->r600_blit.vb_used + 48) > rdev->r600_blit.vb_total) { - WARN_ON(1); - } + /* draw 10 */ + draw_auto(rdev); - 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 10 */ - 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 17 */ - 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 15 */ - 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); - - /* 74 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; - } + /* 5 */ + cp_set_surface_sync(rdev, + PACKET3_CB_ACTION_ENA | PACKET3_CB0_DEST_BASE_ENA, + size_in_bytes, dst_gpu_addr); + + /* 74 ring dwords per loop */ + vb += 12; + rdev->r600_blit.vb_used += 4*12; + src_gpu_addr += size_in_bytes; + dst_gpu_addr += size_in_bytes; + num_pages -= pages_per_loop; } } - diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h index 3d7a0d7c6a9a..62877ef7e0c6 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.h +++ b/drivers/gpu/drm/radeon/radeon_asic.h @@ -423,11 +423,11 @@ void evergreen_disable_interrupt_state(struct radeon_device *rdev); int evergreen_blit_init(struct radeon_device *rdev); void evergreen_blit_fini(struct radeon_device *rdev); /* evergreen blit */ -int evergreen_blit_prepare_copy(struct radeon_device *rdev, int size_bytes); +int evergreen_blit_prepare_copy(struct radeon_device *rdev, unsigned num_pages); void evergreen_blit_done_copy(struct radeon_device *rdev, struct radeon_fence *fence); void evergreen_kms_blit_copy(struct radeon_device *rdev, u64 src_gpu_addr, u64 dst_gpu_addr, - int size_bytes); + unsigned num_pages); /* * cayman