gen7_blorp_emit_urb_config(struct brw_context *brw,
const brw_blorp_params *params)
{
- /* The minimum valid value is 32. See 3DSTATE_URB_VS,
- * Dword 1.15:0 "VS Number of URB Entries".
+ unsigned urb_size = (brw->is_haswell && brw->gt == 3) ? 32 : 16;
+ gen7_emit_push_constant_state(brw,
+ urb_size / 2 /* vs_size */,
+ 0 /* gs_size */,
+ urb_size / 2 /* fs_size */);
+
+ /* The minimum valid number of VS entries is 32. See 3DSTATE_URB_VS, Dword
+ * 1.15:0 "VS Number of URB Entries".
*/
- int num_vs_entries = 32;
- int vs_size = 2;
- int vs_start = 2; /* skip over push constants */
-
- gen7_emit_urb_state(brw, num_vs_entries, vs_size, vs_start);
+ gen7_emit_urb_state(brw,
+ 32 /* num_vs_entries */,
+ 2 /* vs_size */,
+ 2 /* vs_start */,
+ 0 /* num_gs_entries */,
+ 1 /* gs_size */,
+ 2 /* gs_start */);
}
*/
struct intel_region *region = surface->mt->region;
uint32_t tile_x, tile_y;
- uint8_t mocs = brw->is_haswell ? GEN7_MOCS_L3 : 0;
+ const uint8_t mocs = GEN7_MOCS_L3;
uint32_t tiling = surface->map_stencil_as_y_tiled
? I915_TILING_Y : region->tiling;
/* reloc */
surf[1] =
- surface->compute_tile_offsets(&tile_x, &tile_y) + region->bo->offset;
+ surface->compute_tile_offsets(&tile_x, &tile_y) + region->bo->offset64;
/* Note that the low bits of these fields are missing, so
* there's the possibility of getting in trouble.
drm_intel_bo_emit_reloc(brw->batch.bo,
wm_surf_offset + 4,
region->bo,
- surf[1] - region->bo->offset,
+ surf[1] - region->bo->offset64,
read_domains, write_domain);
gen7_check_surface_setup(surf, is_render_target);
// sampler->ss0.min_mag_neq = 1;
- /* Set LOD bias:
+ /* Set LOD bias:
*/
sampler->ss0.lod_bias = 0;
sampler->ss0.lod_preclamp = 1; /* OpenGL mode */
sampler->ss0.default_color_mode = 0; /* OpenGL/DX10 mode */
- /* Set BaseMipLevel, MaxLOD, MinLOD:
+ /* Set BaseMipLevel, MaxLOD, MinLOD:
*
* XXX: I don't think that using firstLevel, lastLevel works,
* because we always setup the surface state as if firstLevel ==
OUT_BATCH(0);
ADVANCE_BATCH();
+ /**
+ * From Graphics BSpec: 3D-Media-GPGPU Engine > 3D Pipeline Stages >
+ * Geometry > Geometry Shader > State:
+ *
+ * "Note: Because of corruption in IVB:GT2, software needs to flush the
+ * whole fixed function pipeline when the GS enable changes value in
+ * the 3DSTATE_GS."
+ *
+ * The hardware architects have clarified that in this context "flush the
+ * whole fixed function pipeline" means to emit a PIPE_CONTROL with the "CS
+ * Stall" bit set.
+ */
+ if (!brw->is_haswell && brw->gt == 2 && brw->gs.enabled)
+ gen7_emit_cs_stall_flush(brw);
+
BEGIN_BATCH(7);
OUT_BATCH(_3DSTATE_GS << 16 | (7 - 2));
OUT_BATCH(0);
OUT_BATCH(0);
OUT_BATCH(0);
ADVANCE_BATCH();
+ brw->gs.enabled = false;
}
/* 3DSTATE_STREAMOUT
OUT_BATCH(_3DSTATE_SF << 16 | (7 - 2));
OUT_BATCH(params->depth_format <<
GEN7_SF_DEPTH_BUFFER_SURFACE_FORMAT_SHIFT);
- OUT_BATCH(params->num_samples > 1 ? GEN6_SF_MSRAST_ON_PATTERN : 0);
+ OUT_BATCH(params->dst.num_samples > 1 ? GEN6_SF_MSRAST_ON_PATTERN : 0);
OUT_BATCH(0);
OUT_BATCH(0);
OUT_BATCH(0);
dw1 |= GEN7_WM_DISPATCH_ENABLE; /* We are rendering */
}
- if (params->num_samples > 1) {
+ if (params->dst.num_samples > 1) {
dw1 |= GEN7_WM_MSRAST_ON_PATTERN;
if (prog_data && prog_data->persample_msaa_dispatch)
dw2 |= GEN7_WM_MSDISPMODE_PERSAMPLE;
const brw_blorp_params *params,
uint32_t wm_push_const_offset)
{
- uint8_t mocs = brw->is_haswell ? GEN7_MOCS_L3 : 0;
+ const uint8_t mocs = GEN7_MOCS_L3;
/* Make sure the push constants fill an exact integer number of
* registers.
gen7_blorp_emit_depth_stencil_config(struct brw_context *brw,
const brw_blorp_params *params)
{
- struct gl_context *ctx = &brw->ctx;
- uint32_t draw_x = params->depth.x_offset;
- uint32_t draw_y = params->depth.y_offset;
- uint32_t tile_mask_x, tile_mask_y;
- uint8_t mocs = brw->is_haswell ? GEN7_MOCS_L3 : 0;
+ const uint8_t mocs = GEN7_MOCS_L3;
uint32_t surfwidth, surfheight;
uint32_t surftype;
unsigned int depth = MAX2(params->depth.mt->logical_depth0, 1);
GLenum gl_target = params->depth.mt->target;
unsigned int lod;
- brw_get_depthstencil_tile_masks(params->depth.mt,
- params->depth.level,
- params->depth.layer,
- NULL,
- &tile_mask_x, &tile_mask_y);
switch (gl_target) {
case GL_TEXTURE_CUBE_MAP_ARRAY:
case GL_TEXTURE_CUBE_MAP:
surfwidth = params->depth.width;
surfheight = params->depth.height;
} else {
- surfwidth = params->depth.mt->physical_width0;
- surfheight = params->depth.mt->physical_height0;
+ surfwidth = params->depth.mt->logical_width0;
+ surfheight = params->depth.mt->logical_height0;
}
/* 3DSTATE_DEPTH_BUFFER */
{
- uint32_t tile_x = draw_x & tile_mask_x;
- uint32_t tile_y = draw_y & tile_mask_y;
- uint32_t offset =
- intel_region_get_aligned_offset(params->depth.mt->region,
- draw_x & ~tile_mask_x,
- draw_y & ~tile_mask_y, false);
-
- /* According to the Sandy Bridge PRM, volume 2 part 1, pp326-327
- * (3DSTATE_DEPTH_BUFFER dw5), in the documentation for "Depth
- * Coordinate Offset X/Y":
- *
- * "The 3 LSBs of both offsets must be zero to ensure correct
- * alignment"
- *
- * We have no guarantee that tile_x and tile_y are correctly aligned,
- * since they are determined by the mipmap layout, which is only aligned
- * to multiples of 4.
- *
- * So, to avoid hanging the GPU, just smash the low order 3 bits of
- * tile_x and tile_y to 0. This is a temporary workaround until we come
- * up with a better solution.
- */
- WARN_ONCE((tile_x & 7) || (tile_y & 7),
- "Depth/stencil buffer needs alignment to 8-pixel boundaries.\n"
- "Truncating offset, bad rendering may occur.\n");
- tile_x &= ~7;
- tile_y &= ~7;
-
intel_emit_depth_stall_flushes(brw);
BEGIN_BATCH(7);
params->depth_format << 18 |
1 << 22 | /* hiz enable */
1 << 28 | /* depth write */
- BRW_SURFACE_2D << 29);
+ surftype << 29);
OUT_RELOC(params->depth.mt->region->bo,
I915_GEM_DOMAIN_RENDER, I915_GEM_DOMAIN_RENDER,
- offset);
- OUT_BATCH((params->depth.width + tile_x - 1) << 4 |
- (params->depth.height + tile_y - 1) << 18);
- OUT_BATCH(mocs);
- OUT_BATCH(tile_x |
- tile_y << 16);
+ 0);
+ OUT_BATCH((surfwidth - 1) << 4 |
+ (surfheight - 1) << 18 |
+ lod);
+ OUT_BATCH(((depth - 1) << 21) |
+ (min_array_element << 10) |
+ mocs);
OUT_BATCH(0);
+ OUT_BATCH((depth - 1) << 21);
ADVANCE_BATCH();
}
/* 3DSTATE_HIER_DEPTH_BUFFER */
{
struct intel_region *hiz_region = params->depth.mt->hiz_mt->region;
- uint32_t hiz_offset =
- intel_region_get_aligned_offset(hiz_region,
- draw_x & ~tile_mask_x,
- (draw_y & ~tile_mask_y) / 2, false);
BEGIN_BATCH(3);
OUT_BATCH((GEN7_3DSTATE_HIER_DEPTH_BUFFER << 16) | (3 - 2));
(hiz_region->pitch - 1));
OUT_RELOC(hiz_region->bo,
I915_GEM_DOMAIN_RENDER, I915_GEM_DOMAIN_RENDER,
- hiz_offset);
+ 0);
ADVANCE_BATCH();
}
gen7_blorp_emit_depth_disable(struct brw_context *brw,
const brw_blorp_params *params)
{
+ intel_emit_depth_stall_flushes(brw);
+
BEGIN_BATCH(7);
OUT_BATCH(GEN7_3DSTATE_DEPTH_BUFFER << 16 | (7 - 2));
OUT_BATCH(BRW_DEPTHFORMAT_D32_FLOAT << 18 | (BRW_SURFACE_NULL << 29));
OUT_BATCH(0);
OUT_BATCH(0);
ADVANCE_BATCH();
+
+ BEGIN_BATCH(3);
+ OUT_BATCH(GEN7_3DSTATE_HIER_DEPTH_BUFFER << 16 | (3 - 2));
+ OUT_BATCH(0);
+ OUT_BATCH(0);
+ ADVANCE_BATCH();
+
+ BEGIN_BATCH(3);
+ OUT_BATCH(GEN7_3DSTATE_STENCIL_BUFFER << 16 | (3 - 2));
+ OUT_BATCH(0);
+ OUT_BATCH(0);
+ ADVANCE_BATCH();
}
gen7_blorp_exec(struct brw_context *brw,
const brw_blorp_params *params)
{
+ if (brw->gen >= 8)
+ return;
+
brw_blorp_prog_data *prog_data = NULL;
uint32_t cc_blend_state_offset = 0;
uint32_t cc_state_offset = 0;
uint32_t sampler_offset = 0;
uint32_t prog_offset = params->get_wm_prog(brw, &prog_data);
- gen6_blorp_emit_batch_head(brw, params);
- gen6_emit_3dstate_multisample(brw, params->num_samples);
- gen6_emit_3dstate_sample_mask(brw, params->num_samples, 1.0, false, ~0u);
+ gen6_emit_3dstate_multisample(brw, params->dst.num_samples);
+ gen6_emit_3dstate_sample_mask(brw,
+ params->dst.num_samples > 1 ?
+ (1 << params->dst.num_samples) - 1 : 1);
gen6_blorp_emit_state_base_address(brw, params);
gen6_blorp_emit_vertices(brw, params);
gen7_blorp_emit_urb_config(brw, params);
projects / mesa.git / blobdiff