#include "main/bufferobj.h"
#include "main/context.h"
#include "main/enums.h"
+#include "main/macros.h"
#include "brw_draw.h"
#include "brw_defines.h"
BRW_SURFACEFORMAT_R16G16B16A16_FLOAT
};
+static GLuint uint_types_direct[5] = {
+ 0,
+ BRW_SURFACEFORMAT_R32_UINT,
+ BRW_SURFACEFORMAT_R32G32_UINT,
+ BRW_SURFACEFORMAT_R32G32B32_UINT,
+ BRW_SURFACEFORMAT_R32G32B32A32_UINT
+};
+
static GLuint uint_types_norm[5] = {
0,
BRW_SURFACEFORMAT_R32_UNORM,
BRW_SURFACEFORMAT_R32G32B32A32_USCALED
};
+static GLuint int_types_direct[5] = {
+ 0,
+ BRW_SURFACEFORMAT_R32_SINT,
+ BRW_SURFACEFORMAT_R32G32_SINT,
+ BRW_SURFACEFORMAT_R32G32B32_SINT,
+ BRW_SURFACEFORMAT_R32G32B32A32_SINT
+};
+
static GLuint int_types_norm[5] = {
0,
BRW_SURFACEFORMAT_R32_SNORM,
BRW_SURFACEFORMAT_R32G32B32A32_SSCALED
};
+static GLuint ushort_types_direct[5] = {
+ 0,
+ BRW_SURFACEFORMAT_R16_UINT,
+ BRW_SURFACEFORMAT_R16G16_UINT,
+ BRW_SURFACEFORMAT_R16G16B16A16_UINT,
+ BRW_SURFACEFORMAT_R16G16B16A16_UINT
+};
+
static GLuint ushort_types_norm[5] = {
0,
BRW_SURFACEFORMAT_R16_UNORM,
BRW_SURFACEFORMAT_R16G16B16A16_USCALED
};
+static GLuint short_types_direct[5] = {
+ 0,
+ BRW_SURFACEFORMAT_R16_SINT,
+ BRW_SURFACEFORMAT_R16G16_SINT,
+ BRW_SURFACEFORMAT_R16G16B16A16_SINT,
+ BRW_SURFACEFORMAT_R16G16B16A16_SINT
+};
+
static GLuint short_types_norm[5] = {
0,
BRW_SURFACEFORMAT_R16_SNORM,
BRW_SURFACEFORMAT_R16G16B16A16_SSCALED
};
+static GLuint ubyte_types_direct[5] = {
+ 0,
+ BRW_SURFACEFORMAT_R8_UINT,
+ BRW_SURFACEFORMAT_R8G8_UINT,
+ BRW_SURFACEFORMAT_R8G8B8A8_UINT,
+ BRW_SURFACEFORMAT_R8G8B8A8_UINT
+};
+
static GLuint ubyte_types_norm[5] = {
0,
BRW_SURFACEFORMAT_R8_UNORM,
BRW_SURFACEFORMAT_R8G8B8A8_USCALED
};
+static GLuint byte_types_direct[5] = {
+ 0,
+ BRW_SURFACEFORMAT_R8_SINT,
+ BRW_SURFACEFORMAT_R8G8_SINT,
+ BRW_SURFACEFORMAT_R8G8B8A8_SINT,
+ BRW_SURFACEFORMAT_R8G8B8A8_SINT
+};
+
static GLuint byte_types_norm[5] = {
0,
BRW_SURFACEFORMAT_R8_SNORM,
* Format will be GL_RGBA or possibly GL_BGRA for GLubyte[4] color arrays.
*/
static GLuint get_surface_type( GLenum type, GLuint size,
- GLenum format, GLboolean normalized )
+ GLenum format, bool normalized, bool integer )
{
if (unlikely(INTEL_DEBUG & DEBUG_VERTS))
printf("type %s size %d normalized %d\n",
_mesa_lookup_enum_by_nr(type), size, normalized);
- if (normalized) {
+ if (integer) {
+ assert(format == GL_RGBA); /* sanity check */
+ switch (type) {
+ case GL_INT: return int_types_direct[size];
+ case GL_SHORT: return short_types_direct[size];
+ case GL_BYTE: return byte_types_direct[size];
+ case GL_UNSIGNED_INT: return uint_types_direct[size];
+ case GL_UNSIGNED_SHORT: return ushort_types_direct[size];
+ case GL_UNSIGNED_BYTE: return ubyte_types_direct[size];
+ default: assert(0); return 0;
+ }
+ } else if (normalized) {
switch (type) {
case GL_DOUBLE: return double_types[size];
case GL_FLOAT: return float_types[size];
case GL_UNSIGNED_INT: return uint_types_scale[size];
case GL_UNSIGNED_SHORT: return ushort_types_scale[size];
case GL_UNSIGNED_BYTE: return ubyte_types_scale[size];
+ /* This produces GL_FIXED inputs as values between INT32_MIN and
+ * INT32_MAX, which will be scaled down by 1/65536 by the VS.
+ */
+ case GL_FIXED: return int_types_scale[size];
default: assert(0); return 0;
- }
+ }
}
}
case GL_UNSIGNED_INT: return sizeof(GLuint);
case GL_UNSIGNED_SHORT: return sizeof(GLushort);
case GL_UNSIGNED_BYTE: return sizeof(GLubyte);
- default: return 0;
- }
+ case GL_FIXED: return sizeof(GLuint);
+ default: assert(0); return 0;
+ }
}
-static GLuint get_index_type(GLenum type)
+static GLuint get_index_type(GLenum type)
{
switch (type) {
case GL_UNSIGNED_BYTE: return BRW_INDEX_BYTE;
}
static void
-copy_array_to_vbo_array( struct brw_context *brw,
- struct brw_vertex_element *element,
- struct brw_vertex_buffer *buffer,
- GLuint dst_stride)
+copy_array_to_vbo_array(struct brw_context *brw,
+ struct brw_vertex_element *element,
+ int min, int max,
+ struct brw_vertex_buffer *buffer,
+ GLuint dst_stride)
{
- GLuint size = element->count * dst_stride;
+ if (min == -1) {
+ /* If we don't have computed min/max bounds, then this must be a use of
+ * the current attribute, which has a 0 stride. Otherwise, we wouldn't
+ * know what data to upload.
+ */
+ assert(element->glarray->StrideB == 0);
- buffer->stride = dst_stride;
- if (dst_stride == element->glarray->StrideB) {
- intel_upload_data(&brw->intel, element->glarray->Ptr, size, dst_stride,
+ intel_upload_data(&brw->intel, element->glarray->Ptr,
+ element->element_size,
+ element->element_size,
+ &buffer->bo, &buffer->offset);
+
+ buffer->stride = 0;
+ return;
+ }
+
+ int src_stride = element->glarray->StrideB;
+ const unsigned char *src = element->glarray->Ptr + min * src_stride;
+ int count = max - min + 1;
+ GLuint size = count * dst_stride;
+
+ if (dst_stride == src_stride) {
+ intel_upload_data(&brw->intel, src, size, dst_stride,
&buffer->bo, &buffer->offset);
} else {
- const unsigned char *src = element->glarray->Ptr;
- char *dst = intel_upload_map(&brw->intel, size, dst_stride);
- int i;
+ char * const map = intel_upload_map(&brw->intel, size, dst_stride);
+ char *dst = map;
- for (i = 0; i < element->count; i++) {
+ while (count--) {
memcpy(dst, src, dst_stride);
- src += element->glarray->StrideB;
+ src += src_stride;
dst += dst_stride;
}
- intel_upload_unmap(&brw->intel, dst, size, dst_stride,
+ intel_upload_unmap(&brw->intel, map, size, dst_stride,
&buffer->bo, &buffer->offset);
}
+ buffer->stride = dst_stride;
}
static void brw_prepare_vertices(struct brw_context *brw)
{
struct gl_context *ctx = &brw->intel.ctx;
struct intel_context *intel = intel_context(ctx);
- GLbitfield vs_inputs = brw->vs.prog_data->inputs_read;
+ /* CACHE_NEW_VS_PROG */
+ GLbitfield64 vs_inputs = brw->vs.prog_data->inputs_read;
const unsigned char *ptr = NULL;
- GLuint interleaved = 0, total_size = 0, count = -1;
+ GLuint interleaved = 0, total_size = 0;
unsigned int min_index = brw->vb.min_index;
unsigned int max_index = brw->vb.max_index;
- int i, j;
+ int delta, i, j;
struct brw_vertex_element *upload[VERT_ATTRIB_MAX];
GLuint nr_uploads = 0;
/* Accumulate the list of enabled arrays. */
brw->vb.nr_enabled = 0;
while (vs_inputs) {
- GLuint i = _mesa_ffsll(vs_inputs) - 1;
+ GLuint i = ffsll(vs_inputs) - 1;
struct brw_vertex_element *input = &brw->vb.inputs[i];
- vs_inputs &= ~(1 << i);
- brw->vb.enabled[brw->vb.nr_enabled++] = input;
+ vs_inputs &= ~BITFIELD64_BIT(i);
+ if (input->glarray->Size && get_size(input->glarray->Type))
+ brw->vb.enabled[brw->vb.nr_enabled++] = input;
}
if (brw->vb.nr_enabled == 0)
return;
if (brw->vb.nr_buffers)
- goto validate;
-
- /* XXX: In the rare cases where this happens we fallback all
- * the way to software rasterization, although a tnl fallback
- * would be sufficient. I don't know of *any* real world
- * cases with > 17 vertex attributes enabled, so it probably
- * isn't an issue at this point.
- */
- if (brw->vb.nr_enabled >= BRW_VEP_MAX) {
- intel->Fallback = GL_TRUE; /* boolean, not bitfield */
- return;
- }
+ goto prepare;
for (i = j = 0; i < brw->vb.nr_enabled; i++) {
struct brw_vertex_element *input = brw->vb.enabled[i];
}
if (k == i) {
struct brw_vertex_buffer *buffer = &brw->vb.buffers[j];
+
/* Named buffer object: Just reference its contents directly. */
- buffer->bo = intel_bufferobj_source(intel, intel_buffer,
+ buffer->bo = intel_bufferobj_source(intel,
+ intel_buffer, type_size,
&buffer->offset);
drm_intel_bo_reference(buffer->bo);
buffer->offset += (uintptr_t)glarray->Ptr;
input->buffer = j++;
input->offset = 0;
}
- input->count = glarray->_MaxElement;
/* This is a common place to reach if the user mistakenly supplies
* a pointer in place of a VBO offset. If we just let it go through,
*/
assert(input->offset < brw->vb.buffers[input->buffer].bo->size);
} else {
- input->count = glarray->StrideB ? max_index + 1 : 1;
-
/* Queue the buffer object up to be uploaded in the next pass,
* when we've decided if we're doing interleaved or not.
*/
/* Position array not properly enabled:
*/
if (input->attrib == VERT_ATTRIB_POS && glarray->StrideB == 0) {
- intel->Fallback = GL_TRUE; /* boolean, not bitfield */
+ intel->Fallback = true; /* boolean, not bitfield */
return;
}
{
interleaved = 0;
}
- else if (total_size & (type_size -1))
+ else if ((uintptr_t)(glarray->Ptr - ptr) & (type_size -1))
{
/* enforce natural alignment (for doubles) */
interleaved = 0;
}
- if (count > input->count)
- count = input->count;
-
upload[nr_uploads++] = input;
+ total_size = ALIGN(total_size, type_size);
total_size += input->element_size;
}
}
+ /* If we need to upload all the arrays, then we can trim those arrays to
+ * only the used elements [min_index, max_index] so long as we adjust all
+ * the values used in the 3DPRIMITIVE i.e. by setting the vertex bias.
+ */
+ brw->vb.start_vertex_bias = 0;
+ delta = min_index;
+ if (nr_uploads == brw->vb.nr_enabled) {
+ brw->vb.start_vertex_bias = -delta;
+ delta = 0;
+ }
+ if (delta && !brw->intel.intelScreen->relaxed_relocations)
+ min_index = delta = 0;
+
/* Handle any arrays to be uploaded. */
if (nr_uploads > 1) {
if (interleaved && interleaved <= 2*total_size) {
+ struct brw_vertex_buffer *buffer = &brw->vb.buffers[j];
/* All uploads are interleaved, so upload the arrays together as
* interleaved. First, upload the contents and set up upload[0].
*/
- upload[0]->count = count; /* trim the upload over all arrays */
- copy_array_to_vbo_array(brw,
- upload[0], &brw->vb.buffers[j],
- interleaved);
+ copy_array_to_vbo_array(brw, upload[0], min_index, max_index,
+ buffer, interleaved);
+ buffer->offset -= delta * interleaved;
for (i = 0; i < nr_uploads; i++) {
/* Then, just point upload[i] at upload[0]'s buffer. */
}
else if (total_size < 2048) {
/* Upload non-interleaved arrays into a single interleaved array */
- struct brw_vertex_buffer *buffer = &brw->vb.buffers[j];
- int count = upload[0]->count, offset;
+ struct brw_vertex_buffer *buffer;
+ int count = MAX2(max_index - min_index + 1, 1);
+ int offset;
char *map;
map = intel_upload_map(&brw->intel, total_size * count, total_size);
const unsigned char *src = upload[i]->glarray->Ptr;
int size = upload[i]->element_size;
int stride = upload[i]->glarray->StrideB;
- char *dst = map + offset;
+ char *dst;
int n;
+ offset = ALIGN(offset, get_size(upload[i]->glarray->Type));
+ dst = map + offset;
+ src += min_index * stride;
+
for (n = 0; n < count; n++) {
memcpy(dst, src, size);
src += stride;
offset += size;
}
- intel_upload_unmap(&brw->intel, map, total_size * count, total_size,
+ assert(offset == total_size);
+ buffer = &brw->vb.buffers[j++];
+ intel_upload_unmap(&brw->intel, map, offset * count, offset,
&buffer->bo, &buffer->offset);
buffer->stride = offset;
- j++;
+ buffer->offset -= delta * offset;
nr_uploads = 0;
}
}
/* Upload non-interleaved arrays */
for (i = 0; i < nr_uploads; i++) {
- copy_array_to_vbo_array(brw,
- upload[i], &brw->vb.buffers[j],
- upload[i]->element_size);
+ struct brw_vertex_buffer *buffer = &brw->vb.buffers[j];
+ copy_array_to_vbo_array(brw, upload[i], min_index, max_index,
+ buffer, upload[i]->element_size);
+ buffer->offset -= delta * buffer->stride;
upload[i]->buffer = j++;
upload[i]->offset = 0;
}
/* can we simply extend the current vb? */
- brw->vb.start_vertex_bias = 0;
if (j == brw->vb.nr_current_buffers) {
int delta = 0;
for (i = 0; i < j; i++) {
break;
d = brw->vb.buffers[i].offset - brw->vb.current_buffers[i].offset;
- if (delta == 0)
+ if (d < 0)
+ break;
+ if (i == 0)
delta = d / brw->vb.current_buffers[i].stride;
if (delta * brw->vb.current_buffers[i].stride != d)
break;
}
if (i == j) {
- brw->vb.start_vertex_bias = delta;
+ brw->vb.start_vertex_bias += delta;
while (--j >= 0)
drm_intel_bo_unreference(brw->vb.buffers[j].bo);
j = 0;
brw->vb.nr_buffers = j;
-validate:
+prepare:
brw_prepare_query_begin(brw);
- for (i = 0; i < brw->vb.nr_buffers; i++) {
- brw_add_validated_bo(brw, brw->vb.buffers[i].bo);
- }
}
static void brw_emit_vertices(struct brw_context *brw)
{
struct gl_context *ctx = &brw->intel.ctx;
struct intel_context *intel = intel_context(ctx);
- GLuint i;
+ GLuint i, nr_elements;
+
+ brw_prepare_vertices(brw);
brw_emit_query_begin(brw);
*/
if (brw->vb.nr_enabled == 0) {
BEGIN_BATCH(3);
- OUT_BATCH((CMD_VERTEX_ELEMENT << 16) | 1);
+ OUT_BATCH((_3DSTATE_VERTEX_ELEMENTS << 16) | 1);
if (intel->gen >= 6) {
OUT_BATCH((0 << GEN6_VE0_INDEX_SHIFT) |
GEN6_VE0_VALID |
*/
if (brw->vb.nr_buffers) {
+ if (intel->gen >= 6) {
+ assert(brw->vb.nr_buffers <= 33);
+ } else {
+ assert(brw->vb.nr_buffers <= 17);
+ }
+
BEGIN_BATCH(1 + 4*brw->vb.nr_buffers);
- OUT_BATCH((CMD_VERTEX_BUFFER << 16) | (4*brw->vb.nr_buffers - 1));
+ OUT_BATCH((_3DSTATE_VERTEX_BUFFERS << 16) | (4*brw->vb.nr_buffers - 1));
for (i = 0; i < brw->vb.nr_buffers; i++) {
struct brw_vertex_buffer *buffer = &brw->vb.buffers[i];
uint32_t dw0;
dw0 = BRW_VB0_ACCESS_VERTEXDATA | (i << BRW_VB0_INDEX_SHIFT);
}
+ if (intel->gen >= 7)
+ dw0 |= GEN7_VB0_ADDRESS_MODIFYENABLE;
+
OUT_BATCH(dw0 | (buffer->stride << BRW_VB0_PITCH_SHIFT));
OUT_RELOC(buffer->bo, I915_GEM_DOMAIN_VERTEX, 0, buffer->offset);
if (intel->gen >= 5) {
OUT_RELOC(buffer->bo, I915_GEM_DOMAIN_VERTEX, 0, buffer->bo->size - 1);
} else
- OUT_BATCH(buffer->bo->size / buffer->stride);
+ OUT_BATCH(0);
OUT_BATCH(0); /* Instance data step rate */
brw->vb.current_buffers[i].handle = buffer->bo->handle;
ADVANCE_BATCH();
}
- BEGIN_BATCH(1 + brw->vb.nr_enabled * 2);
- OUT_BATCH((CMD_VERTEX_ELEMENT << 16) | (2*brw->vb.nr_enabled - 1));
+ nr_elements = brw->vb.nr_enabled + brw->vs.prog_data->uses_vertexid;
+
+ /* The hardware allows one more VERTEX_ELEMENTS than VERTEX_BUFFERS, presumably
+ * for VertexID/InstanceID.
+ */
+ if (intel->gen >= 6) {
+ assert(nr_elements <= 34);
+ } else {
+ assert(nr_elements <= 18);
+ }
+
+ BEGIN_BATCH(1 + nr_elements * 2);
+ OUT_BATCH((_3DSTATE_VERTEX_ELEMENTS << 16) | (2 * nr_elements - 1));
for (i = 0; i < brw->vb.nr_enabled; i++) {
struct brw_vertex_element *input = brw->vb.enabled[i];
uint32_t format = get_surface_type(input->glarray->Type,
input->glarray->Size,
input->glarray->Format,
- input->glarray->Normalized);
+ input->glarray->Normalized,
+ input->glarray->Integer);
uint32_t comp0 = BRW_VE1_COMPONENT_STORE_SRC;
uint32_t comp1 = BRW_VE1_COMPONENT_STORE_SRC;
uint32_t comp2 = BRW_VE1_COMPONENT_STORE_SRC;
case 0: comp0 = BRW_VE1_COMPONENT_STORE_0;
case 1: comp1 = BRW_VE1_COMPONENT_STORE_0;
case 2: comp2 = BRW_VE1_COMPONENT_STORE_0;
- case 3: comp3 = BRW_VE1_COMPONENT_STORE_1_FLT;
+ case 3: comp3 = input->glarray->Integer ? BRW_VE1_COMPONENT_STORE_1_INT
+ : BRW_VE1_COMPONENT_STORE_1_FLT;
break;
}
(comp3 << BRW_VE1_COMPONENT_3_SHIFT) |
((i * 4) << BRW_VE1_DST_OFFSET_SHIFT));
}
+
+ if (brw->vs.prog_data->uses_vertexid) {
+ uint32_t dw0 = 0, dw1 = 0;
+
+ dw1 = ((BRW_VE1_COMPONENT_STORE_VID << BRW_VE1_COMPONENT_0_SHIFT) |
+ (BRW_VE1_COMPONENT_STORE_IID << BRW_VE1_COMPONENT_1_SHIFT) |
+ (BRW_VE1_COMPONENT_STORE_PID << BRW_VE1_COMPONENT_2_SHIFT) |
+ (BRW_VE1_COMPONENT_STORE_0 << BRW_VE1_COMPONENT_3_SHIFT));
+
+ if (intel->gen >= 6) {
+ dw0 |= GEN6_VE0_VALID;
+ } else {
+ dw0 |= BRW_VE0_VALID;
+ dw1 |= (i * 4) << BRW_VE1_DST_OFFSET_SHIFT;
+ }
+
+ /* Note that for gl_VertexID, gl_InstanceID, and gl_PrimitiveID values,
+ * the format is ignored and the value is always int.
+ */
+
+ OUT_BATCH(dw0);
+ OUT_BATCH(dw1);
+ }
+
CACHED_BATCH();
}
.dirty = {
.mesa = 0,
.brw = BRW_NEW_BATCH | BRW_NEW_VERTICES,
- .cache = 0,
+ .cache = CACHE_NEW_VS_PROG,
},
- .prepare = brw_prepare_vertices,
.emit = brw_emit_vertices,
};
-static void brw_prepare_indices(struct brw_context *brw)
+static void brw_upload_indices(struct brw_context *brw)
{
struct gl_context *ctx = &brw->intel.ctx;
struct intel_context *intel = &brw->intel;
intel_upload_data(&brw->intel, index_buffer->ptr, ib_size, ib_type_size,
&bo, &offset);
brw->ib.start_vertex_offset = offset / ib_type_size;
- offset = 0;
} else {
offset = (GLuint) (unsigned long) index_buffer->ptr;
* rebase it into a temporary.
*/
if ((get_size(index_buffer->type) - 1) & offset) {
- GLubyte *map = ctx->Driver.MapBuffer(ctx,
- GL_ELEMENT_ARRAY_BUFFER_ARB,
- GL_DYNAMIC_DRAW_ARB,
- bufferobj);
- map += offset;
+ GLubyte *map = ctx->Driver.MapBufferRange(ctx,
+ offset,
+ ib_size,
+ GL_MAP_WRITE_BIT,
+ bufferobj);
intel_upload_data(&brw->intel, map, ib_size, ib_type_size,
&bo, &offset);
brw->ib.start_vertex_offset = offset / ib_type_size;
- offset = 0;
- ctx->Driver.UnmapBuffer(ctx, GL_ELEMENT_ARRAY_BUFFER_ARB, bufferobj);
+ ctx->Driver.UnmapBuffer(ctx, bufferobj);
} else {
/* Use CMD_3D_PRIM's start_vertex_offset to avoid re-uploading
* the index buffer state when we're just moving the start index
*/
brw->ib.start_vertex_offset = offset / ib_type_size;
- bo = intel_bufferobj_source(intel, intel_buffer_object(bufferobj),
+ bo = intel_bufferobj_source(intel,
+ intel_buffer_object(bufferobj),
+ ib_type_size,
&offset);
drm_intel_bo_reference(bo);
+
+ brw->ib.start_vertex_offset += offset / ib_type_size;
}
}
- if (brw->ib.bo != bo || brw->ib.offset != offset) {
+ if (brw->ib.bo != bo) {
drm_intel_bo_unreference(brw->ib.bo);
brw->ib.bo = bo;
- brw->ib.offset = offset;
- brw_add_validated_bo(brw, brw->ib.bo);
brw->state.dirty.brw |= BRW_NEW_INDEX_BUFFER;
} else {
drm_intel_bo_unreference(bo);
}
+
+ if (index_buffer->type != brw->ib.type) {
+ brw->ib.type = index_buffer->type;
+ brw->state.dirty.brw |= BRW_NEW_INDEX_BUFFER;
+ }
}
const struct brw_tracked_state brw_indices = {
.brw = BRW_NEW_INDICES,
.cache = 0,
},
- .prepare = brw_prepare_indices,
+ .emit = brw_upload_indices,
};
static void brw_emit_index_buffer(struct brw_context *brw)
1);
OUT_RELOC(brw->ib.bo,
I915_GEM_DOMAIN_VERTEX, 0,
- brw->ib.offset);
+ 0);
OUT_RELOC(brw->ib.bo,
I915_GEM_DOMAIN_VERTEX, 0,
brw->ib.bo->size - 1);