#include "intel_batchbuffer.h"
+#include "intel_fbo.h"
#include "intel_regions.h"
#include "brw_context.h"
struct intel_context *intel = &brw->intel;
BEGIN_BATCH(6);
- OUT_BATCH(CMD_BINDING_TABLE_PTRS << 16 | (6 - 2));
+ OUT_BATCH(_3DSTATE_BINDING_TABLE_POINTERS << 16 | (6 - 2));
OUT_BATCH(brw->vs.bind_bo_offset);
OUT_BATCH(0); /* gs */
OUT_BATCH(0); /* clip */
const struct brw_tracked_state brw_binding_table_pointers = {
.dirty = {
.mesa = 0,
- .brw = BRW_NEW_BATCH | BRW_NEW_BINDING_TABLE,
+ .brw = BRW_NEW_BATCH
+ | BRW_NEW_VS_BINDING_TABLE
+ | BRW_NEW_GS_BINDING_TABLE
+ | BRW_NEW_PS_BINDING_TABLE,
.cache = 0,
},
.emit = upload_binding_table_pointers,
struct intel_context *intel = &brw->intel;
BEGIN_BATCH(4);
- OUT_BATCH(CMD_BINDING_TABLE_PTRS << 16 |
+ OUT_BATCH(_3DSTATE_BINDING_TABLE_POINTERS << 16 |
GEN6_BINDING_TABLE_MODIFY_VS |
GEN6_BINDING_TABLE_MODIFY_GS |
GEN6_BINDING_TABLE_MODIFY_PS |
const struct brw_tracked_state gen6_binding_table_pointers = {
.dirty = {
.mesa = 0,
- .brw = BRW_NEW_BATCH | BRW_NEW_BINDING_TABLE,
+ .brw = BRW_NEW_BATCH
+ | BRW_NEW_VS_BINDING_TABLE
+ | BRW_NEW_GS_BINDING_TABLE
+ | BRW_NEW_PS_BINDING_TABLE,
.cache = 0,
},
.emit = upload_gen6_binding_table_pointers,
}
BEGIN_BATCH(7);
- OUT_BATCH(CMD_PIPELINED_STATE_POINTERS << 16 | (7 - 2));
- OUT_RELOC(brw->vs.state_bo, I915_GEM_DOMAIN_INSTRUCTION, 0, 0);
+ OUT_BATCH(_3DSTATE_PIPELINED_POINTERS << 16 | (7 - 2));
+ OUT_RELOC(intel->batch.bo, I915_GEM_DOMAIN_INSTRUCTION, 0,
+ brw->vs.state_offset);
if (brw->gs.prog_active)
- OUT_RELOC(brw->gs.state_bo, I915_GEM_DOMAIN_INSTRUCTION, 0, 1);
+ OUT_RELOC(brw->intel.batch.bo, I915_GEM_DOMAIN_INSTRUCTION, 0,
+ brw->gs.state_offset | 1);
else
OUT_BATCH(0);
- OUT_RELOC(brw->clip.state_bo, I915_GEM_DOMAIN_INSTRUCTION, 0, 1);
- OUT_RELOC(brw->sf.state_bo, I915_GEM_DOMAIN_INSTRUCTION, 0, 0);
- OUT_RELOC(brw->wm.state_bo, I915_GEM_DOMAIN_INSTRUCTION, 0, 0);
- OUT_RELOC(brw->cc.state_bo, I915_GEM_DOMAIN_INSTRUCTION, 0,
+ OUT_RELOC(brw->intel.batch.bo, I915_GEM_DOMAIN_INSTRUCTION, 0,
+ brw->clip.state_offset | 1);
+ OUT_RELOC(brw->intel.batch.bo, I915_GEM_DOMAIN_INSTRUCTION, 0,
+ brw->sf.state_offset);
+ OUT_RELOC(brw->intel.batch.bo, I915_GEM_DOMAIN_INSTRUCTION, 0,
+ brw->wm.state_offset);
+ OUT_RELOC(brw->intel.batch.bo, I915_GEM_DOMAIN_INSTRUCTION, 0,
brw->cc.state_offset);
ADVANCE_BATCH();
brw->state.dirty.brw |= BRW_NEW_PSP;
}
-
-static void prepare_psp_urb_cbs(struct brw_context *brw)
-{
- brw_add_validated_bo(brw, brw->vs.state_bo);
- brw_add_validated_bo(brw, brw->gs.state_bo);
- brw_add_validated_bo(brw, brw->clip.state_bo);
- brw_add_validated_bo(brw, brw->sf.state_bo);
- brw_add_validated_bo(brw, brw->wm.state_bo);
-}
-
static void upload_psp_urb_cbs(struct brw_context *brw )
{
upload_pipelined_state_pointers(brw);
CACHE_NEW_WM_UNIT |
CACHE_NEW_CC_UNIT)
},
- .prepare = prepare_psp_urb_cbs,
.emit = upload_psp_urb_cbs,
};
static void prepare_depthbuffer(struct brw_context *brw)
{
- struct intel_region *region = brw->state.depth_region;
-
- if (region != NULL)
- brw_add_validated_bo(brw, region->buffer);
+ struct intel_context *intel = &brw->intel;
+ struct gl_context *ctx = &intel->ctx;
+ struct gl_framebuffer *fb = ctx->DrawBuffer;
+ struct intel_renderbuffer *drb = intel_get_renderbuffer(fb, BUFFER_DEPTH);
+ struct intel_renderbuffer *srb = intel_get_renderbuffer(fb, BUFFER_STENCIL);
+
+ if (drb)
+ brw_add_validated_bo(brw, drb->region->buffer);
+ if (drb && drb->hiz_region)
+ brw_add_validated_bo(brw, drb->hiz_region->buffer);
+ if (srb)
+ brw_add_validated_bo(brw, srb->region->buffer);
}
static void emit_depthbuffer(struct brw_context *brw)
{
struct intel_context *intel = &brw->intel;
- struct intel_region *region = brw->state.depth_region;
+ struct gl_context *ctx = &intel->ctx;
+ struct gl_framebuffer *fb = ctx->DrawBuffer;
+ /* _NEW_BUFFERS */
+ struct intel_renderbuffer *depth_irb = intel_get_renderbuffer(fb, BUFFER_DEPTH);
+ struct intel_renderbuffer *stencil_irb = intel_get_renderbuffer(fb, BUFFER_STENCIL);
+ struct intel_region *hiz_region = depth_irb ? depth_irb->hiz_region : NULL;
unsigned int len;
+ /*
+ * If either depth or stencil buffer has packed depth/stencil format,
+ * then don't use separate stencil. Emit only a depth buffer.
+ */
+ if (depth_irb && depth_irb->Base.Format == MESA_FORMAT_S8_Z24) {
+ stencil_irb = NULL;
+ } else if (!depth_irb && stencil_irb
+ && stencil_irb->Base.Format == MESA_FORMAT_S8_Z24) {
+ depth_irb = stencil_irb;
+ stencil_irb = NULL;
+ }
+
if (intel->gen >= 6)
len = 7;
else if (intel->is_g4x || intel->gen == 5)
else
len = 5;
- if (region == NULL) {
+ if (!depth_irb && !stencil_irb) {
BEGIN_BATCH(len);
OUT_BATCH(_3DSTATE_DEPTH_BUFFER << 16 | (len - 2));
OUT_BATCH((BRW_DEPTHFORMAT_D32_FLOAT << 18) |
OUT_BATCH(0);
ADVANCE_BATCH();
+
+ } else if (!depth_irb && stencil_irb) {
+ /*
+ * There exists a separate stencil buffer but no depth buffer.
+ *
+ * The stencil buffer inherits most of its fields from
+ * 3DSTATE_DEPTH_BUFFER: namely the tile walk, surface type, width, and
+ * height.
+ *
+ * Since the stencil buffer has quirky pitch requirements, its region
+ * was allocated with half height and double cpp. So we need
+ * a multiplier of 2 to obtain the surface's real height.
+ *
+ * Enable the hiz bit because it and the separate stencil bit must have
+ * the same value. From Section 2.11.5.6.1.1 3DSTATE_DEPTH_BUFFER, Bit
+ * 1.21 "Separate Stencil Enable":
+ * [DevIL]: If this field is enabled, Hierarchical Depth Buffer
+ * Enable must also be enabled.
+ *
+ * [DevGT]: This field must be set to the same value (enabled or
+ * disabled) as Hierarchical Depth Buffer Enable
+ */
+ assert(intel->has_separate_stencil);
+ assert(stencil_irb->Base.Format == MESA_FORMAT_S8);
+
+ BEGIN_BATCH(len);
+ OUT_BATCH(_3DSTATE_DEPTH_BUFFER << 16 | (len - 2));
+ OUT_BATCH((BRW_DEPTHFORMAT_D32_FLOAT << 18) |
+ (1 << 21) | /* separate stencil enable */
+ (1 << 22) | /* hiz enable */
+ (BRW_TILEWALK_YMAJOR << 26) |
+ (BRW_SURFACE_2D << 29));
+ OUT_BATCH(0);
+ OUT_BATCH(((stencil_irb->region->width - 1) << 6) |
+ (2 * stencil_irb->region->height - 1) << 19);
+ OUT_BATCH(0);
+ OUT_BATCH(0);
+
+ if (intel->gen >= 6)
+ OUT_BATCH(0);
+
+ ADVANCE_BATCH();
+
} else {
+ struct intel_region *region = depth_irb->region;
unsigned int format;
+ uint32_t tile_x, tile_y, offset;
+
+ /* If using separate stencil, hiz must be enabled. */
+ assert(!stencil_irb || hiz_region);
switch (region->cpp) {
case 2:
case 4:
if (intel->depth_buffer_is_float)
format = BRW_DEPTHFORMAT_D32_FLOAT;
+ else if (hiz_region)
+ format = BRW_DEPTHFORMAT_D24_UNORM_X8_UINT;
else
format = BRW_DEPTHFORMAT_D24_UNORM_S8_UINT;
break;
return;
}
- assert(region->tiling != I915_TILING_X);
- if (intel->gen >= 6)
- assert(region->tiling != I915_TILING_NONE);
+ offset = intel_region_tile_offsets(region, &tile_x, &tile_y);
+
+ assert(intel->gen < 6 || region->tiling == I915_TILING_Y);
+ assert(!hiz_region || region->tiling == I915_TILING_Y);
BEGIN_BATCH(len);
OUT_BATCH(_3DSTATE_DEPTH_BUFFER << 16 | (len - 2));
OUT_BATCH(((region->pitch * region->cpp) - 1) |
(format << 18) |
+ ((hiz_region ? 1 : 0) << 21) | /* separate stencil enable */
+ ((hiz_region ? 1 : 0) << 22) | /* hiz enable */
(BRW_TILEWALK_YMAJOR << 26) |
((region->tiling != I915_TILING_NONE) << 27) |
(BRW_SURFACE_2D << 29));
OUT_RELOC(region->buffer,
I915_GEM_DOMAIN_RENDER, I915_GEM_DOMAIN_RENDER,
- 0);
+ offset);
OUT_BATCH((BRW_SURFACE_MIPMAPLAYOUT_BELOW << 1) |
((region->width - 1) << 6) |
((region->height - 1) << 19));
OUT_BATCH(0);
if (intel->is_g4x || intel->gen >= 5)
- OUT_BATCH(0);
+ OUT_BATCH(tile_x | (tile_y << 16));
+ else
+ assert(tile_x == 0 && tile_y == 0);
if (intel->gen >= 6)
OUT_BATCH(0);
ADVANCE_BATCH();
}
- /* Initialize it for safety. */
- if (intel->gen >= 6) {
+ /* Emit hiz buffer. */
+ if (hiz_region || stencil_irb) {
+ BEGIN_BATCH(3);
+ OUT_BATCH((_3DSTATE_HIER_DEPTH_BUFFER << 16) | (3 - 2));
+ OUT_BATCH(hiz_region->pitch * hiz_region->cpp - 1);
+ OUT_RELOC(hiz_region->buffer,
+ I915_GEM_DOMAIN_RENDER, I915_GEM_DOMAIN_RENDER,
+ 0);
+ ADVANCE_BATCH();
+ }
+
+ /* Emit stencil buffer. */
+ if (hiz_region || stencil_irb) {
+ BEGIN_BATCH(3);
+ OUT_BATCH((_3DSTATE_STENCIL_BUFFER << 16) | (3 - 2));
+ OUT_BATCH(stencil_irb->region->pitch * stencil_irb->region->cpp - 1);
+ OUT_RELOC(stencil_irb->region->buffer,
+ I915_GEM_DOMAIN_RENDER, I915_GEM_DOMAIN_RENDER,
+ 0);
+ ADVANCE_BATCH();
+ }
+
+ /*
+ * On Gen >= 6, emit clear params for safety. If using hiz, then clear
+ * params must be emitted.
+ *
+ * From Section 2.11.5.6.4.1 3DSTATE_CLEAR_PARAMS:
+ * 3DSTATE_CLEAR_PARAMS packet must follow the DEPTH_BUFFER_STATE packet
+ * when HiZ is enabled and the DEPTH_BUFFER_STATE changes.
+ */
+ if (intel->gen >= 6 || hiz_region) {
BEGIN_BATCH(2);
OUT_BATCH(_3DSTATE_CLEAR_PARAMS << 16 | (2 - 2));
OUT_BATCH(0);
const struct brw_tracked_state brw_depthbuffer = {
.dirty = {
- .mesa = 0,
- .brw = BRW_NEW_DEPTH_BUFFER | BRW_NEW_BATCH,
+ .mesa = _NEW_BUFFERS,
+ .brw = BRW_NEW_BATCH,
.cache = 0,
},
.prepare = prepare_depthbuffer,
static void upload_polygon_stipple(struct brw_context *brw)
{
+ struct intel_context *intel = &brw->intel;
struct gl_context *ctx = &brw->intel.ctx;
- struct brw_polygon_stipple bps;
GLuint i;
if (!ctx->Polygon.StippleFlag)
return;
- memset(&bps, 0, sizeof(bps));
- bps.header.opcode = _3DSTATE_POLY_STIPPLE_PATTERN;
- bps.header.length = sizeof(bps)/4-2;
+ BEGIN_BATCH(33);
+ OUT_BATCH(_3DSTATE_POLY_STIPPLE_PATTERN << 16 | (33 - 2));
/* Polygon stipple is provided in OpenGL order, i.e. bottom
* row first. If we're rendering to a window (i.e. the
*/
if (ctx->DrawBuffer->Name == 0) {
for (i = 0; i < 32; i++)
- bps.stipple[i] = ctx->PolygonStipple[31 - i]; /* invert */
+ OUT_BATCH(ctx->PolygonStipple[31 - i]); /* invert */
}
else {
for (i = 0; i < 32; i++)
- bps.stipple[i] = ctx->PolygonStipple[i]; /* don't invert */
+ OUT_BATCH(ctx->PolygonStipple[i]);
}
-
- BRW_CACHED_BATCH_STRUCT(brw, &bps);
+ CACHED_BATCH();
}
const struct brw_tracked_state brw_polygon_stipple = {
static void upload_polygon_stipple_offset(struct brw_context *brw)
{
+ struct intel_context *intel = &brw->intel;
struct gl_context *ctx = &brw->intel.ctx;
- struct brw_polygon_stipple_offset bpso;
if (!ctx->Polygon.StippleFlag)
return;
- memset(&bpso, 0, sizeof(bpso));
- bpso.header.opcode = _3DSTATE_POLY_STIPPLE_OFFSET;
- bpso.header.length = sizeof(bpso)/4-2;
+ BEGIN_BATCH(2);
+ OUT_BATCH(_3DSTATE_POLY_STIPPLE_OFFSET << 16 | (2-2));
/* If we're drawing to a system window (ctx->DrawBuffer->Name == 0),
* we have to invert the Y axis in order to match the OpenGL
* system works just fine, and there's no window system to
* worry about.
*/
- if (brw->intel.ctx.DrawBuffer->Name == 0) {
- bpso.bits0.x_offset = 0;
- bpso.bits0.y_offset = (32 - (ctx->DrawBuffer->Height & 31)) & 31;
- }
- else {
- bpso.bits0.y_offset = 0;
- bpso.bits0.x_offset = 0;
- }
-
- BRW_CACHED_BATCH_STRUCT(brw, &bpso);
+ if (brw->intel.ctx.DrawBuffer->Name == 0)
+ OUT_BATCH((32 - (ctx->DrawBuffer->Height & 31)) & 31);
+ else
+ OUT_BATCH(0);
+ CACHED_BATCH();
}
#define _NEW_WINDOW_POS 0x40000000
*/
static void upload_aa_line_parameters(struct brw_context *brw)
{
+ struct intel_context *intel = &brw->intel;
struct gl_context *ctx = &brw->intel.ctx;
- struct brw_aa_line_parameters balp;
if (!ctx->Line.SmoothFlag || !brw->has_aa_line_parameters)
return;
+ OUT_BATCH(_3DSTATE_AA_LINE_PARAMETERS << 16 | (3 - 2));
/* use legacy aa line coverage computation */
- memset(&balp, 0, sizeof(balp));
- balp.header.opcode = _3DSTATE_AA_LINE_PARAMETERS;
- balp.header.length = sizeof(balp) / 4 - 2;
-
- BRW_CACHED_BATCH_STRUCT(brw, &balp);
+ OUT_BATCH(0);
+ OUT_BATCH(0);
+ CACHED_BATCH();
}
const struct brw_tracked_state brw_aa_line_parameters = {
static void upload_line_stipple(struct brw_context *brw)
{
+ struct intel_context *intel = &brw->intel;
struct gl_context *ctx = &brw->intel.ctx;
- struct brw_line_stipple bls;
GLfloat tmp;
GLint tmpi;
if (!ctx->Line.StippleFlag)
return;
- memset(&bls, 0, sizeof(bls));
- bls.header.opcode = _3DSTATE_LINE_STIPPLE_PATTERN;
- bls.header.length = sizeof(bls)/4 - 2;
-
- bls.bits0.pattern = ctx->Line.StipplePattern;
- bls.bits1.repeat_count = ctx->Line.StippleFactor;
-
+ BEGIN_BATCH(3);
+ OUT_BATCH(_3DSTATE_LINE_STIPPLE_PATTERN << 16 | (3 - 2));
+ OUT_BATCH(ctx->Line.StipplePattern);
tmp = 1.0 / (GLfloat) ctx->Line.StippleFactor;
tmpi = tmp * (1<<13);
-
-
- bls.bits1.inverse_repeat_count = tmpi;
-
- BRW_CACHED_BATCH_STRUCT(brw, &bls);
+ OUT_BATCH(tmpi << 16 | ctx->Line.StippleFactor);
+ CACHED_BATCH();
}
const struct brw_tracked_state brw_line_stipple = {
if (intel->gen >= 6) {
int i;
+ int len = intel->gen >= 7 ? 4 : 3;
- BEGIN_BATCH(3);
- OUT_BATCH(_3DSTATE_MULTISAMPLE << 16 | (3 - 2));
+ BEGIN_BATCH(len);
+ OUT_BATCH(_3DSTATE_MULTISAMPLE << 16 | (len - 2));
OUT_BATCH(MS_PIXEL_LOCATION_CENTER |
MS_NUMSAMPLES_1);
OUT_BATCH(0); /* positions for 4/8-sample */
+ if (intel->gen >= 7)
+ OUT_BATCH(0);
ADVANCE_BATCH();
BEGIN_BATCH(2);
OUT_BATCH(1);
ADVANCE_BATCH();
- for (i = 0; i < 4; i++) {
- BEGIN_BATCH(4);
- OUT_BATCH(_3DSTATE_GS_SVB_INDEX << 16 | (4 - 2));
- OUT_BATCH(i << SVB_INDEX_SHIFT);
- OUT_BATCH(0);
- OUT_BATCH(0xffffffff);
- ADVANCE_BATCH();
+ if (intel->gen < 7) {
+ for (i = 0; i < 4; i++) {
+ BEGIN_BATCH(4);
+ OUT_BATCH(_3DSTATE_GS_SVB_INDEX << 16 | (4 - 2));
+ OUT_BATCH(i << SVB_INDEX_SHIFT);
+ OUT_BATCH(0);
+ OUT_BATCH(0xffffffff);
+ ADVANCE_BATCH();
+ }
}
}
if (intel->gen >= 6) {
BEGIN_BATCH(10);
OUT_BATCH(CMD_STATE_BASE_ADDRESS << 16 | (10 - 2));
- OUT_BATCH(1); /* General state base address */
- OUT_RELOC(intel->batch->buf, I915_GEM_DOMAIN_SAMPLER, 0,
- 1); /* Surface state base address */
- OUT_BATCH(1); /* Dynamic state base address */
- OUT_BATCH(1); /* Indirect object base address */
- OUT_BATCH(1); /* Instruction base address */
+ /* General state base address: stateless DP read/write requests */
+ OUT_BATCH(1);
+ /* Surface state base address:
+ * BINDING_TABLE_STATE
+ * SURFACE_STATE
+ */
+ OUT_RELOC(intel->batch.bo, I915_GEM_DOMAIN_SAMPLER, 0, 1);
+ /* Dynamic state base address:
+ * SAMPLER_STATE
+ * SAMPLER_BORDER_COLOR_STATE
+ * CLIP, SF, WM/CC viewport state
+ * COLOR_CALC_STATE
+ * DEPTH_STENCIL_STATE
+ * BLEND_STATE
+ * Push constants (when INSTPM: CONSTANT_BUFFER Address Offset
+ * Disable is clear, which we rely on)
+ */
+ OUT_RELOC(intel->batch.bo, (I915_GEM_DOMAIN_RENDER |
+ I915_GEM_DOMAIN_INSTRUCTION), 0, 1);
+
+ OUT_BATCH(1); /* Indirect object base address: MEDIA_OBJECT data */
+ OUT_BATCH(1); /* Instruction base address: shader kernels (incl. SIP) */
OUT_BATCH(1); /* General state upper bound */
OUT_BATCH(1); /* Dynamic state upper bound */
OUT_BATCH(1); /* Indirect object upper bound */
BEGIN_BATCH(8);
OUT_BATCH(CMD_STATE_BASE_ADDRESS << 16 | (8 - 2));
OUT_BATCH(1); /* General state base address */
- OUT_RELOC(intel->batch->buf, I915_GEM_DOMAIN_SAMPLER, 0,
+ OUT_RELOC(intel->batch.bo, I915_GEM_DOMAIN_SAMPLER, 0,
1); /* Surface state base address */
OUT_BATCH(1); /* Indirect object base address */
OUT_BATCH(1); /* Instruction base address */
BEGIN_BATCH(6);
OUT_BATCH(CMD_STATE_BASE_ADDRESS << 16 | (6 - 2));
OUT_BATCH(1); /* General state base address */
- OUT_RELOC(intel->batch->buf, I915_GEM_DOMAIN_SAMPLER, 0,
+ OUT_RELOC(intel->batch.bo, I915_GEM_DOMAIN_SAMPLER, 0,
1); /* Surface state base address */
OUT_BATCH(1); /* Indirect object base address */
OUT_BATCH(1); /* General state upper bound */
projects / mesa.git / blobdiff