#include "main/mtypes.h"
#include "main/samplerobj.h"
#include "main/texformat.h"
+#include "main/teximage.h"
#include "program/prog_parameter.h"
#include "intel_mipmap_tree.h"
#include "brw_defines.h"
#include "brw_wm.h"
+/**
+ * Convert an swizzle enumeration (i.e. SWIZZLE_X) to one of the Gen7.5+
+ * "Shader Channel Select" enumerations (i.e. HSW_SCS_RED). The mappings are
+ *
+ * SWIZZLE_X, SWIZZLE_Y, SWIZZLE_Z, SWIZZLE_W, SWIZZLE_ZERO, SWIZZLE_ONE
+ * 0 1 2 3 4 5
+ * 4 5 6 7 0 1
+ * SCS_RED, SCS_GREEN, SCS_BLUE, SCS_ALPHA, SCS_ZERO, SCS_ONE
+ *
+ * which is simply adding 4 then modding by 8 (or anding with 7).
+ */
+static unsigned
+swizzle_to_scs(unsigned swizzle)
+{
+ return (swizzle + 4) & 7;
+}
+
static uint32_t
surface_tiling_mode(uint32_t tiling)
{
case 16:
return GEN8_SURFACE_VALIGN_16;
default:
- assert(!"Unsupported vertical surface alignment.");
- return GEN8_SURFACE_VALIGN_4;
+ unreachable("Unsupported vertical surface alignment.");
}
}
case 16:
return GEN8_SURFACE_HALIGN_16;
default:
- assert(!"Unsupported horizontal surface alignment.");
- return GEN8_SURFACE_HALIGN_4;
+ unreachable("Unsupported horizontal surface alignment.");
}
}
+static uint32_t *
+allocate_surface_state(struct brw_context *brw, uint32_t *out_offset)
+{
+ int dwords = brw->gen >= 9 ? 16 : 13;
+ uint32_t *surf = brw_state_batch(brw, AUB_TRACE_SURFACE_STATE,
+ dwords * 4, 64, out_offset);
+ memset(surf, 0, dwords * 4);
+ return surf;
+}
+
static void
gen8_emit_buffer_surface_state(struct brw_context *brw,
uint32_t *out_offset,
unsigned surface_format,
unsigned buffer_size,
unsigned pitch,
- unsigned mocs,
bool rw)
{
- uint32_t *surf = brw_state_batch(brw, AUB_TRACE_SURFACE_STATE,
- 13 * 4, 64, out_offset);
- memset(surf, 0, 13 * 4);
+ const unsigned mocs = brw->gen >= 9 ? SKL_MOCS_WB : BDW_MOCS_WB;
+ uint32_t *surf = allocate_surface_state(brw, out_offset);
surf[0] = BRW_SURFACE_BUFFER << BRW_SURFACE_TYPE_SHIFT |
surface_format << BRW_SURFACE_FORMAT_SHIFT |
surf[2] = SET_FIELD((buffer_size - 1) & 0x7f, GEN7_SURFACE_WIDTH) |
SET_FIELD(((buffer_size - 1) >> 7) & 0x3fff, GEN7_SURFACE_HEIGHT);
- surf[3] = SET_FIELD(((buffer_size - 1) >> 21) & 0x3f, BRW_SURFACE_DEPTH) |
- (pitch - 1);
+ if (surface_format == BRW_SURFACEFORMAT_RAW)
+ surf[3] = SET_FIELD(((buffer_size - 1) >> 21) & 0x3ff, BRW_SURFACE_DEPTH);
+ else
+ surf[3] = SET_FIELD(((buffer_size - 1) >> 21) & 0x3f, BRW_SURFACE_DEPTH);
+ surf[3] |= (pitch - 1);
surf[7] = SET_FIELD(HSW_SCS_RED, GEN7_SURFACE_SCS_R) |
SET_FIELD(HSW_SCS_GREEN, GEN7_SURFACE_SCS_G) |
SET_FIELD(HSW_SCS_BLUE, GEN7_SURFACE_SCS_B) |
struct intel_mipmap_tree *mt = intelObj->mt;
struct gl_texture_image *firstImage = tObj->Image[0][tObj->BaseLevel];
struct gl_sampler_object *sampler = _mesa_get_samplerobj(ctx, unit);
+ struct intel_mipmap_tree *aux_mt = NULL;
+ uint32_t aux_mode = 0;
mesa_format format = intelObj->_Format;
+ uint32_t mocs_wb = brw->gen >= 9 ? SKL_MOCS_WB : BDW_MOCS_WB;
if (tObj->Target == GL_TEXTURE_BUFFER) {
brw_update_buffer_texture_surface(ctx, unit, surf_offset);
pitch = mt->pitch;
}
+ if (mt->mcs_mt) {
+ aux_mt = mt->mcs_mt;
+ aux_mode = GEN8_SURFACE_AUX_MODE_MCS;
+ }
+
/* If this is a view with restricted NumLayers, then our effective depth
* is not just the miptree depth.
*/
uint32_t tex_format = translate_tex_format(brw, format, sampler->sRGBDecode);
- uint32_t *surf = brw_state_batch(brw, AUB_TRACE_SURFACE_STATE,
- 13 * 4, 64, surf_offset);
+ uint32_t *surf = allocate_surface_state(brw, surf_offset);
surf[0] = translate_tex_target(tObj->Target) << BRW_SURFACE_TYPE_SHIFT |
tex_format << BRW_SURFACE_FORMAT_SHIFT |
surf[0] |= BRW_SURFACE_CUBEFACE_ENABLES;
}
- if (mt->logical_depth0 > 1 && tObj->Target != GL_TEXTURE_3D)
+ if (_mesa_is_array_texture(tObj->Target) ||
+ tObj->Target == GL_TEXTURE_CUBE_MAP)
surf[0] |= GEN8_SURFACE_IS_ARRAY;
- surf[1] = SET_FIELD(BDW_MOCS_WB, GEN8_SURFACE_MOCS) | mt->qpitch >> 2;
+ surf[1] = SET_FIELD(mocs_wb, GEN8_SURFACE_MOCS) | mt->qpitch >> 2;
surf[2] = SET_FIELD(mt->logical_width0 - 1, GEN7_SURFACE_WIDTH) |
SET_FIELD(mt->logical_height0 - 1, GEN7_SURFACE_HEIGHT);
GEN7_SURFACE_MIN_LOD) |
(intelObj->_MaxLevel - tObj->BaseLevel); /* mip count */
- surf[6] = 0;
+ if (aux_mt) {
+ surf[6] = SET_FIELD(mt->qpitch / 4, GEN8_SURFACE_AUX_QPITCH) |
+ SET_FIELD((aux_mt->pitch / 128) - 1, GEN8_SURFACE_AUX_PITCH) |
+ aux_mode;
+ } else {
+ surf[6] = 0;
+ }
/* Handling GL_ALPHA as a surface format override breaks 1.30+ style
* texturing functions that return a float, as our code generation always
(firstImage->_BaseFormat == GL_DEPTH_COMPONENT ||
firstImage->_BaseFormat == GL_DEPTH_STENCIL);
+ surf[7] = mt->fast_clear_color_value;
+
const int swizzle =
unlikely(alpha_depth) ? SWIZZLE_XYZW : brw_get_texture_swizzle(ctx, tObj);
- surf[7] =
- SET_FIELD(brw_swizzle_to_scs(GET_SWZ(swizzle, 0), false), GEN7_SURFACE_SCS_R) |
- SET_FIELD(brw_swizzle_to_scs(GET_SWZ(swizzle, 1), false), GEN7_SURFACE_SCS_G) |
- SET_FIELD(brw_swizzle_to_scs(GET_SWZ(swizzle, 2), false), GEN7_SURFACE_SCS_B) |
- SET_FIELD(brw_swizzle_to_scs(GET_SWZ(swizzle, 3), false), GEN7_SURFACE_SCS_A);
+ surf[7] |=
+ SET_FIELD(swizzle_to_scs(GET_SWZ(swizzle, 0)), GEN7_SURFACE_SCS_R) |
+ SET_FIELD(swizzle_to_scs(GET_SWZ(swizzle, 1)), GEN7_SURFACE_SCS_G) |
+ SET_FIELD(swizzle_to_scs(GET_SWZ(swizzle, 2)), GEN7_SURFACE_SCS_B) |
+ SET_FIELD(swizzle_to_scs(GET_SWZ(swizzle, 3)), GEN7_SURFACE_SCS_A);
*((uint64_t *) &surf[8]) = mt->bo->offset64 + mt->offset; /* reloc */
- surf[10] = 0;
- surf[11] = 0;
+ if (aux_mt) {
+ *((uint64_t *) &surf[10]) = aux_mt->bo->offset64;
+ drm_intel_bo_emit_reloc(brw->batch.bo, *surf_offset + 10 * 4,
+ aux_mt->bo, 0,
+ I915_GEM_DOMAIN_SAMPLER, 0);
+ } else {
+ surf[10] = 0;
+ surf[11] = 0;
+ }
surf[12] = 0;
/* Emit relocation to surface contents */
I915_GEM_DOMAIN_SAMPLER, 0);
}
-static void
-gen8_create_raw_surface(struct brw_context *brw, drm_intel_bo *bo,
- uint32_t offset, uint32_t size,
- uint32_t *out_offset, bool rw)
-{
- gen8_emit_buffer_surface_state(brw,
- out_offset,
- bo,
- offset,
- BRW_SURFACEFORMAT_RAW,
- size,
- 1,
- 0 /* mocs */,
- true /* rw */);
-}
-
/**
- * Create the constant buffer surface. Vertex/fragment shader constants will be
- * read from this buffer with Data Port Read instructions/messages.
+ * Creates a null surface.
+ *
+ * This is used when the shader doesn't write to any color output. An FB
+ * write to target 0 will still be emitted, because that's how the thread is
+ * terminated (and computed depth is returned), so we need to have the
+ * hardware discard the target 0 color output..
*/
static void
-gen8_update_null_renderbuffer_surface(struct brw_context *brw, unsigned unit)
+gen8_emit_null_surface_state(struct brw_context *brw,
+ unsigned width,
+ unsigned height,
+ unsigned samples,
+ uint32_t *out_offset)
{
- struct gl_context *ctx = &brw->ctx;
-
- /* _NEW_BUFFERS */
- const struct gl_framebuffer *fb = ctx->DrawBuffer;
- uint32_t surf_index =
- brw->wm.prog_data->binding_table.render_target_start + unit;
-
- uint32_t *surf = brw_state_batch(brw, AUB_TRACE_SURFACE_STATE, 13 * 4, 64,
- &brw->wm.base.surf_offset[surf_index]);
- memset(surf, 0, 13 * 4);
+ uint32_t *surf = allocate_surface_state(brw, out_offset);
surf[0] = BRW_SURFACE_NULL << BRW_SURFACE_TYPE_SHIFT |
BRW_SURFACEFORMAT_B8G8R8A8_UNORM << BRW_SURFACE_FORMAT_SHIFT |
GEN8_SURFACE_TILING_Y;
- surf[2] = SET_FIELD(fb->Width - 1, GEN7_SURFACE_WIDTH) |
- SET_FIELD(fb->Height - 1, GEN7_SURFACE_HEIGHT);
+ surf[2] = SET_FIELD(width - 1, GEN7_SURFACE_WIDTH) |
+ SET_FIELD(height - 1, GEN7_SURFACE_HEIGHT);
}
/**
struct gl_context *ctx = &brw->ctx;
struct intel_renderbuffer *irb = intel_renderbuffer(rb);
struct intel_mipmap_tree *mt = irb->mt;
+ struct intel_mipmap_tree *aux_mt = NULL;
+ uint32_t aux_mode = 0;
unsigned width = mt->logical_width0;
unsigned height = mt->logical_height0;
unsigned pitch = mt->pitch;
uint32_t surf_type;
bool is_array = false;
int depth = MAX2(irb->layer_count, 1);
- int min_array_element = irb->mt_layer / MAX2(mt->num_samples, 1);
-
+ const int min_array_element = (mt->format == MESA_FORMAT_S_UINT8) ?
+ irb->mt_layer : (irb->mt_layer / MAX2(mt->num_samples, 1));
GLenum gl_target =
rb->TexImage ? rb->TexImage->TexObject->Target : GL_TEXTURE_2D;
-
uint32_t surf_index =
brw->wm.prog_data->binding_table.render_target_start + unit;
+ /* FINISHME: Use PTE MOCS on Skylake. */
+ uint32_t mocs = brw->gen >= 9 ? SKL_MOCS_WT : BDW_MOCS_PTE;
intel_miptree_used_for_rendering(mt);
- /* Render targets can't use IMS layout. */
- assert(mt->msaa_layout != INTEL_MSAA_LAYOUT_IMS);
-
switch (gl_target) {
case GL_TEXTURE_CUBE_MAP_ARRAY:
case GL_TEXTURE_CUBE_MAP:
depth *= 6;
break;
case GL_TEXTURE_3D:
- depth = MAX2(rb->Depth, 1);
+ depth = MAX2(irb->mt->logical_depth0, 1);
/* fallthrough */
default:
surf_type = translate_tex_target(gl_target);
}
/* _NEW_BUFFERS */
- mesa_format rb_format = _mesa_get_render_format(ctx, intel_rb_format(irb));
- assert(brw_render_target_supported(brw, rb));
- format = brw->render_target_format[rb_format];
- if (unlikely(!brw->format_supported_as_render_target[rb_format])) {
- _mesa_problem(ctx, "%s: renderbuffer format %s unsupported\n",
- __FUNCTION__, _mesa_get_format_name(rb_format));
+ /* Render targets can't use IMS layout. Stencil in turn gets configured as
+ * single sampled and indexed manually by the program.
+ */
+ if (mt->format == MESA_FORMAT_S_UINT8) {
+ brw_configure_w_tiled(mt, true, &width, &height, &pitch,
+ &tiling, &format);
+ } else {
+ assert(mt->msaa_layout != INTEL_MSAA_LAYOUT_IMS);
+ assert(brw_render_target_supported(brw, rb));
+ mesa_format rb_format = _mesa_get_render_format(ctx,
+ intel_rb_format(irb));
+ format = brw->render_target_format[rb_format];
+ if (unlikely(!brw->format_supported_as_render_target[rb_format]))
+ _mesa_problem(ctx, "%s: renderbuffer format %s unsupported\n",
+ __func__, _mesa_get_format_name(rb_format));
+ }
+
+ if (mt->mcs_mt) {
+ aux_mt = mt->mcs_mt;
+ aux_mode = GEN8_SURFACE_AUX_MODE_MCS;
}
- uint32_t *surf = brw_state_batch(brw, AUB_TRACE_SURFACE_STATE, 13 * 4, 64,
- &brw->wm.base.surf_offset[surf_index]);
+ uint32_t *surf =
+ allocate_surface_state(brw, &brw->wm.base.surf_offset[surf_index]);
surf[0] = (surf_type << BRW_SURFACE_TYPE_SHIFT) |
(is_array ? GEN7_SURFACE_IS_ARRAY : 0) |
horizontal_alignment(mt) |
surface_tiling_mode(tiling);
- surf[1] = SET_FIELD(BDW_MOCS_WT, GEN8_SURFACE_MOCS) | mt->qpitch >> 2;
+ surf[1] = SET_FIELD(mocs, GEN8_SURFACE_MOCS) | mt->qpitch >> 2;
surf[2] = SET_FIELD(width - 1, GEN7_SURFACE_WIDTH) |
SET_FIELD(height - 1, GEN7_SURFACE_HEIGHT);
surf[3] = (depth - 1) << BRW_SURFACE_DEPTH_SHIFT |
(pitch - 1); /* Surface Pitch */
- surf[4] = gen7_surface_msaa_bits(mt->num_samples, mt->msaa_layout) |
- min_array_element << GEN7_SURFACE_MIN_ARRAY_ELEMENT_SHIFT |
+ surf[4] = min_array_element << GEN7_SURFACE_MIN_ARRAY_ELEMENT_SHIFT |
(depth - 1) << GEN7_SURFACE_RENDER_TARGET_VIEW_EXTENT_SHIFT;
+ if (mt->format != MESA_FORMAT_S_UINT8)
+ surf[4] |= gen7_surface_msaa_bits(mt->num_samples, mt->msaa_layout);
+
surf[5] = irb->mt_level - irb->mt->first_level;
- surf[6] = 0; /* Nothing of relevance. */
+ if (aux_mt) {
+ surf[6] = SET_FIELD(mt->qpitch / 4, GEN8_SURFACE_AUX_QPITCH) |
+ SET_FIELD((aux_mt->pitch / 128) - 1, GEN8_SURFACE_AUX_PITCH) |
+ aux_mode;
+ } else {
+ surf[6] = 0;
+ }
surf[7] = mt->fast_clear_color_value |
SET_FIELD(HSW_SCS_RED, GEN7_SURFACE_SCS_R) |
SET_FIELD(HSW_SCS_BLUE, GEN7_SURFACE_SCS_B) |
SET_FIELD(HSW_SCS_ALPHA, GEN7_SURFACE_SCS_A);
- *((uint64_t *) &surf[8]) = mt->bo->offset64; /* reloc */
+ assert(mt->offset % mt->cpp == 0);
+ *((uint64_t *) &surf[8]) = mt->bo->offset64 + mt->offset; /* reloc */
- /* Nothing of relevance. */
- surf[10] = 0;
- surf[11] = 0;
+ if (aux_mt) {
+ *((uint64_t *) &surf[10]) = aux_mt->bo->offset64;
+ drm_intel_bo_emit_reloc(brw->batch.bo,
+ brw->wm.base.surf_offset[surf_index] + 10 * 4,
+ aux_mt->bo, 0,
+ I915_GEM_DOMAIN_RENDER, I915_GEM_DOMAIN_RENDER);
+ } else {
+ surf[10] = 0;
+ surf[11] = 0;
+ }
surf[12] = 0;
drm_intel_bo_emit_reloc(brw->batch.bo,
brw->wm.base.surf_offset[surf_index] + 8 * 4,
mt->bo,
- 0,
+ mt->offset,
I915_GEM_DOMAIN_RENDER,
I915_GEM_DOMAIN_RENDER);
}
{
brw->vtbl.update_texture_surface = gen8_update_texture_surface;
brw->vtbl.update_renderbuffer_surface = gen8_update_renderbuffer_surface;
- brw->vtbl.update_null_renderbuffer_surface =
- gen8_update_null_renderbuffer_surface;
- brw->vtbl.create_raw_surface = gen8_create_raw_surface;
+ brw->vtbl.emit_null_surface_state = gen8_emit_null_surface_state;
brw->vtbl.emit_buffer_surface_state = gen8_emit_buffer_surface_state;
}