aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
-rw-r--r--src/mesa/main/glthread_draw.c190
-rw-r--r--src/mesa/main/varray.c18
-rw-r--r--src/mesa/main/varray.h4
3 files changed, 106 insertions, 106 deletions
diff --git a/src/mesa/main/glthread_draw.c b/src/mesa/main/glthread_draw.c
index 7c41496e4e0..ed27cb14bff 100644
--- a/src/mesa/main/glthread_draw.c
+++ b/src/mesa/main/glthread_draw.c
@@ -107,17 +107,17 @@ upload_multi_indices(struct gl_context *ctx, unsigned total_count,
}
static ALWAYS_INLINE bool
-upload_vertices(struct gl_context *ctx, unsigned attrib_mask,
+upload_vertices(struct gl_context *ctx, unsigned user_buffer_mask,
unsigned start_vertex, unsigned num_vertices,
unsigned start_instance, unsigned num_instances,
- struct glthread_attrib_binding *attribs)
+ struct glthread_attrib_binding *buffers)
{
struct glthread_vao *vao = ctx->GLThread.CurrentVAO;
- unsigned attrib_mask_iter = attrib_mask;
- unsigned num_attribs = 0;
+ unsigned attrib_mask_iter = user_buffer_mask;
+ unsigned num_buffers = 0;
- assert((num_vertices || !(attrib_mask & ~vao->NonZeroDivisorMask)) &&
- (num_instances || !(attrib_mask & vao->NonZeroDivisorMask)));
+ assert((num_vertices || !(user_buffer_mask & ~vao->NonZeroDivisorMask)) &&
+ (num_instances || !(user_buffer_mask & vao->NonZeroDivisorMask)));
while (attrib_mask_iter) {
unsigned i = u_bit_scan(&attrib_mask_iter);
@@ -153,10 +153,10 @@ upload_vertices(struct gl_context *ctx, unsigned attrib_mask,
size, &upload_offset, &upload_buffer, NULL);
assert(upload_buffer);
- attribs[num_attribs].buffer = upload_buffer;
- attribs[num_attribs].offset = upload_offset - offset;
- attribs[num_attribs].original_pointer = ptr;
- num_attribs++;
+ buffers[num_buffers].buffer = upload_buffer;
+ buffers[num_buffers].offset = upload_offset - offset;
+ buffers[num_buffers].original_pointer = ptr;
+ num_buffers++;
}
return true;
}
@@ -169,7 +169,7 @@ struct marshal_cmd_DrawArraysInstancedBaseInstance
GLsizei count;
GLsizei instance_count;
GLuint baseinstance;
- GLuint non_vbo_attrib_mask;
+ GLuint user_buffer_mask;
};
void
@@ -181,13 +181,13 @@ _mesa_unmarshal_DrawArraysInstancedBaseInstance(struct gl_context *ctx,
const GLsizei count = cmd->count;
const GLsizei instance_count = cmd->instance_count;
const GLuint baseinstance = cmd->baseinstance;
- const GLuint non_vbo_attrib_mask = cmd->non_vbo_attrib_mask;
- const struct glthread_attrib_binding *attribs =
+ const GLuint user_buffer_mask = cmd->user_buffer_mask;
+ const struct glthread_attrib_binding *buffers =
(const struct glthread_attrib_binding *)(cmd + 1);
/* Bind uploaded buffers if needed. */
- if (non_vbo_attrib_mask) {
- _mesa_InternalBindVertexBuffers(ctx, attribs, non_vbo_attrib_mask,
+ if (user_buffer_mask) {
+ _mesa_InternalBindVertexBuffers(ctx, buffers, user_buffer_mask,
false);
}
@@ -196,8 +196,8 @@ _mesa_unmarshal_DrawArraysInstancedBaseInstance(struct gl_context *ctx,
baseinstance));
/* Restore states. */
- if (non_vbo_attrib_mask) {
- _mesa_InternalBindVertexBuffers(ctx, attribs, non_vbo_attrib_mask,
+ if (user_buffer_mask) {
+ _mesa_InternalBindVertexBuffers(ctx, buffers, user_buffer_mask,
true);
}
}
@@ -205,12 +205,12 @@ _mesa_unmarshal_DrawArraysInstancedBaseInstance(struct gl_context *ctx,
static ALWAYS_INLINE void
draw_arrays_async(struct gl_context *ctx, GLenum mode, GLint first,
GLsizei count, GLsizei instance_count, GLuint baseinstance,
- unsigned non_vbo_attrib_mask,
- const struct glthread_attrib_binding *attribs)
+ unsigned user_buffer_mask,
+ const struct glthread_attrib_binding *buffers)
{
- int attribs_size = util_bitcount(non_vbo_attrib_mask) * sizeof(attribs[0]);
+ int buffers_size = util_bitcount(user_buffer_mask) * sizeof(buffers[0]);
int cmd_size = sizeof(struct marshal_cmd_DrawArraysInstancedBaseInstance) +
- attribs_size;
+ buffers_size;
struct marshal_cmd_DrawArraysInstancedBaseInstance *cmd;
cmd = _mesa_glthread_allocate_command(ctx, DISPATCH_CMD_DrawArraysInstancedBaseInstance,
@@ -220,10 +220,10 @@ draw_arrays_async(struct gl_context *ctx, GLenum mode, GLint first,
cmd->count = count;
cmd->instance_count = instance_count;
cmd->baseinstance = baseinstance;
- cmd->non_vbo_attrib_mask = non_vbo_attrib_mask;
+ cmd->user_buffer_mask = user_buffer_mask;
- if (non_vbo_attrib_mask)
- memcpy(cmd + 1, attribs, attribs_size);
+ if (user_buffer_mask)
+ memcpy(cmd + 1, buffers, buffers_size);
}
static ALWAYS_INLINE void
@@ -233,7 +233,7 @@ draw_arrays(GLenum mode, GLint first, GLsizei count, GLsizei instance_count,
GET_CURRENT_CONTEXT(ctx);
struct glthread_vao *vao = ctx->GLThread.CurrentVAO;
- unsigned non_vbo_attrib_mask = vao->UserPointerMask & vao->Enabled;
+ unsigned user_buffer_mask = vao->UserPointerMask & vao->Enabled;
if (compiled_into_dlist && ctx->GLThread.inside_dlist) {
_mesa_glthread_finish_before(ctx, "DrawArrays");
@@ -247,7 +247,7 @@ draw_arrays(GLenum mode, GLint first, GLsizei count, GLsizei instance_count,
* This is also an error path. Zero counts should still call the driver
* for possible GL errors.
*/
- if (ctx->API == API_OPENGL_CORE || !non_vbo_attrib_mask ||
+ if (ctx->API == API_OPENGL_CORE || !user_buffer_mask ||
count <= 0 || instance_count <= 0) {
draw_arrays_async(ctx, mode, first, count, instance_count, baseinstance,
0, NULL);
@@ -255,10 +255,10 @@ draw_arrays(GLenum mode, GLint first, GLsizei count, GLsizei instance_count,
}
/* Upload and draw. */
- struct glthread_attrib_binding attribs[VERT_ATTRIB_MAX];
+ struct glthread_attrib_binding buffers[VERT_ATTRIB_MAX];
if (!ctx->GLThread.SupportsNonVBOUploads ||
- !upload_vertices(ctx, non_vbo_attrib_mask, first, count, baseinstance,
- instance_count, attribs)) {
+ !upload_vertices(ctx, user_buffer_mask, first, count, baseinstance,
+ instance_count, buffers)) {
_mesa_glthread_finish_before(ctx, "DrawArrays");
CALL_DrawArraysInstancedBaseInstance(ctx->CurrentServerDispatch,
(mode, first, count, instance_count,
@@ -267,7 +267,7 @@ draw_arrays(GLenum mode, GLint first, GLsizei count, GLsizei instance_count,
}
draw_arrays_async(ctx, mode, first, count, instance_count, baseinstance,
- non_vbo_attrib_mask, attribs);
+ user_buffer_mask, buffers);
}
struct marshal_cmd_MultiDrawArrays
@@ -275,7 +275,7 @@ struct marshal_cmd_MultiDrawArrays
struct marshal_cmd_base cmd_base;
GLenum mode;
GLsizei draw_count;
- GLuint non_vbo_attrib_mask;
+ GLuint user_buffer_mask;
};
void
@@ -284,19 +284,19 @@ _mesa_unmarshal_MultiDrawArrays(struct gl_context *ctx,
{
const GLenum mode = cmd->mode;
const GLsizei draw_count = cmd->draw_count;
- const GLuint non_vbo_attrib_mask = cmd->non_vbo_attrib_mask;
+ const GLuint user_buffer_mask = cmd->user_buffer_mask;
const char *variable_data = (const char *)(cmd + 1);
const GLint *first = (GLint *)variable_data;
variable_data += sizeof(GLint) * draw_count;
const GLsizei *count = (GLsizei *)variable_data;
variable_data += sizeof(GLsizei) * draw_count;
- const struct glthread_attrib_binding *attribs =
+ const struct glthread_attrib_binding *buffers =
(const struct glthread_attrib_binding *)variable_data;
/* Bind uploaded buffers if needed. */
- if (non_vbo_attrib_mask) {
- _mesa_InternalBindVertexBuffers(ctx, attribs, non_vbo_attrib_mask,
+ if (user_buffer_mask) {
+ _mesa_InternalBindVertexBuffers(ctx, buffers, user_buffer_mask,
false);
}
@@ -304,8 +304,8 @@ _mesa_unmarshal_MultiDrawArrays(struct gl_context *ctx,
(mode, first, count, draw_count));
/* Restore states. */
- if (non_vbo_attrib_mask) {
- _mesa_InternalBindVertexBuffers(ctx, attribs, non_vbo_attrib_mask,
+ if (user_buffer_mask) {
+ _mesa_InternalBindVertexBuffers(ctx, buffers, user_buffer_mask,
true);
}
}
@@ -313,30 +313,30 @@ _mesa_unmarshal_MultiDrawArrays(struct gl_context *ctx,
static ALWAYS_INLINE void
multi_draw_arrays_async(struct gl_context *ctx, GLenum mode,
const GLint *first, const GLsizei *count,
- GLsizei draw_count, unsigned non_vbo_attrib_mask,
- const struct glthread_attrib_binding *attribs)
+ GLsizei draw_count, unsigned user_buffer_mask,
+ const struct glthread_attrib_binding *buffers)
{
int first_size = sizeof(GLint) * draw_count;
int count_size = sizeof(GLsizei) * draw_count;
- int attribs_size = util_bitcount(non_vbo_attrib_mask) * sizeof(attribs[0]);
+ int buffers_size = util_bitcount(user_buffer_mask) * sizeof(buffers[0]);
int cmd_size = sizeof(struct marshal_cmd_MultiDrawArrays) +
- first_size + count_size + attribs_size;
+ first_size + count_size + buffers_size;
struct marshal_cmd_MultiDrawArrays *cmd;
cmd = _mesa_glthread_allocate_command(ctx, DISPATCH_CMD_MultiDrawArrays,
cmd_size);
cmd->mode = mode;
cmd->draw_count = draw_count;
- cmd->non_vbo_attrib_mask = non_vbo_attrib_mask;
+ cmd->user_buffer_mask = user_buffer_mask;
char *variable_data = (char*)(cmd + 1);
memcpy(variable_data, first, first_size);
variable_data += first_size;
memcpy(variable_data, count, count_size);
- if (non_vbo_attrib_mask) {
+ if (user_buffer_mask) {
variable_data += count_size;
- memcpy(variable_data, attribs, attribs_size);
+ memcpy(variable_data, buffers, buffers_size);
}
}
@@ -347,13 +347,13 @@ _mesa_marshal_MultiDrawArrays(GLenum mode, const GLint *first,
GET_CURRENT_CONTEXT(ctx);
struct glthread_vao *vao = ctx->GLThread.CurrentVAO;
- unsigned non_vbo_attrib_mask = vao->UserPointerMask & vao->Enabled;
+ unsigned user_buffer_mask = vao->UserPointerMask & vao->Enabled;
if (ctx->GLThread.inside_dlist)
goto sync;
if (draw_count >= 0 &&
- (ctx->API == API_OPENGL_CORE || !non_vbo_attrib_mask)) {
+ (ctx->API == API_OPENGL_CORE || !user_buffer_mask)) {
multi_draw_arrays_async(ctx, mode, first, count, draw_count, 0, NULL);
return;
}
@@ -389,13 +389,13 @@ _mesa_marshal_MultiDrawArrays(GLenum mode, const GLint *first,
}
/* Upload and draw. */
- struct glthread_attrib_binding attribs[VERT_ATTRIB_MAX];
- if (!upload_vertices(ctx, non_vbo_attrib_mask, min_index, num_vertices,
- 0, 1, attribs))
+ struct glthread_attrib_binding buffers[VERT_ATTRIB_MAX];
+ if (!upload_vertices(ctx, user_buffer_mask, min_index, num_vertices,
+ 0, 1, buffers))
goto sync;
multi_draw_arrays_async(ctx, mode, first, count, draw_count,
- non_vbo_attrib_mask, attribs);
+ user_buffer_mask, buffers);
return;
sync:
@@ -416,7 +416,7 @@ struct marshal_cmd_DrawElementsInstancedBaseVertexBaseInstance
GLuint baseinstance;
GLuint min_index;
GLuint max_index;
- GLuint non_vbo_attrib_mask;
+ GLuint user_buffer_mask;
const GLvoid *indices;
struct gl_buffer_object *index_buffer;
};
@@ -434,14 +434,14 @@ _mesa_unmarshal_DrawElementsInstancedBaseVertexBaseInstance(struct gl_context *c
const GLuint baseinstance = cmd->baseinstance;
const GLuint min_index = cmd->min_index;
const GLuint max_index = cmd->max_index;
- const GLuint non_vbo_attrib_mask = cmd->non_vbo_attrib_mask;
+ const GLuint user_buffer_mask = cmd->user_buffer_mask;
struct gl_buffer_object *index_buffer = cmd->index_buffer;
- const struct glthread_attrib_binding *attribs =
+ const struct glthread_attrib_binding *buffers =
(const struct glthread_attrib_binding *)(cmd + 1);
/* Bind uploaded buffers if needed. */
- if (non_vbo_attrib_mask) {
- _mesa_InternalBindVertexBuffers(ctx, attribs, non_vbo_attrib_mask,
+ if (user_buffer_mask) {
+ _mesa_InternalBindVertexBuffers(ctx, buffers, user_buffer_mask,
false);
}
if (index_buffer) {
@@ -464,8 +464,8 @@ _mesa_unmarshal_DrawElementsInstancedBaseVertexBaseInstance(struct gl_context *c
if (index_buffer) {
_mesa_InternalBindElementBuffer(ctx, NULL);
}
- if (non_vbo_attrib_mask) {
- _mesa_InternalBindVertexBuffers(ctx, attribs, non_vbo_attrib_mask,
+ if (user_buffer_mask) {
+ _mesa_InternalBindVertexBuffers(ctx, buffers, user_buffer_mask,
true);
}
}
@@ -476,12 +476,12 @@ draw_elements_async(struct gl_context *ctx, GLenum mode, GLsizei count,
GLint basevertex, GLuint baseinstance,
bool index_bounds_valid, GLuint min_index, GLuint max_index,
struct gl_buffer_object *index_buffer,
- unsigned non_vbo_attrib_mask,
- const struct glthread_attrib_binding *attribs)
+ unsigned user_buffer_mask,
+ const struct glthread_attrib_binding *buffers)
{
- int attribs_size = util_bitcount(non_vbo_attrib_mask) * sizeof(attribs[0]);
+ int buffers_size = util_bitcount(user_buffer_mask) * sizeof(buffers[0]);
int cmd_size = sizeof(struct marshal_cmd_DrawElementsInstancedBaseVertexBaseInstance) +
- attribs_size;
+ buffers_size;
struct marshal_cmd_DrawElementsInstancedBaseVertexBaseInstance *cmd;
cmd = _mesa_glthread_allocate_command(ctx, DISPATCH_CMD_DrawElementsInstancedBaseVertexBaseInstance, cmd_size);
@@ -494,12 +494,12 @@ draw_elements_async(struct gl_context *ctx, GLenum mode, GLsizei count,
cmd->baseinstance = baseinstance;
cmd->min_index = min_index;
cmd->max_index = max_index;
- cmd->non_vbo_attrib_mask = non_vbo_attrib_mask;
+ cmd->user_buffer_mask = user_buffer_mask;
cmd->index_bounds_valid = index_bounds_valid;
cmd->index_buffer = index_buffer;
- if (non_vbo_attrib_mask)
- memcpy(cmd + 1, attribs, attribs_size);
+ if (user_buffer_mask)
+ memcpy(cmd + 1, buffers, buffers_size);
}
static void
@@ -511,7 +511,7 @@ draw_elements(GLenum mode, GLsizei count, GLenum type, const GLvoid *indices,
GET_CURRENT_CONTEXT(ctx);
struct glthread_vao *vao = ctx->GLThread.CurrentVAO;
- unsigned non_vbo_attrib_mask = vao->UserPointerMask & vao->Enabled;
+ unsigned user_buffer_mask = vao->UserPointerMask & vao->Enabled;
bool has_user_indices = vao->CurrentElementBufferName == 0;
if (compiled_into_dlist && ctx->GLThread.inside_dlist)
@@ -525,7 +525,7 @@ draw_elements(GLenum mode, GLsizei count, GLenum type, const GLvoid *indices,
if (ctx->API == API_OPENGL_CORE ||
count <= 0 || instance_count <= 0 || max_index < min_index ||
!is_index_type_valid(type) ||
- (!non_vbo_attrib_mask && !has_user_indices)) {
+ (!user_buffer_mask && !has_user_indices)) {
draw_elements_async(ctx, mode, count, type, indices, instance_count,
basevertex, baseinstance, index_bounds_valid,
min_index, max_index, 0, 0, NULL);
@@ -535,7 +535,7 @@ draw_elements(GLenum mode, GLsizei count, GLenum type, const GLvoid *indices,
if (!ctx->GLThread.SupportsNonVBOUploads)
goto sync;
- bool need_index_bounds = non_vbo_attrib_mask & ~vao->NonZeroDivisorMask;
+ bool need_index_bounds = user_buffer_mask & ~vao->NonZeroDivisorMask;
unsigned index_size = get_index_size(type);
if (need_index_bounds && !index_bounds_valid) {
@@ -566,10 +566,10 @@ draw_elements(GLenum mode, GLsizei count, GLenum type, const GLvoid *indices,
if (util_is_vbo_upload_ratio_too_large(count, num_vertices))
goto sync;
- struct glthread_attrib_binding attribs[VERT_ATTRIB_MAX];
- if (non_vbo_attrib_mask &&
- !upload_vertices(ctx, non_vbo_attrib_mask, start_vertex, num_vertices,
- baseinstance, instance_count, attribs))
+ struct glthread_attrib_binding buffers[VERT_ATTRIB_MAX];
+ if (user_buffer_mask &&
+ !upload_vertices(ctx, user_buffer_mask, start_vertex, num_vertices,
+ baseinstance, instance_count, buffers))
goto sync;
/* Upload indices. */
@@ -581,7 +581,7 @@ draw_elements(GLenum mode, GLsizei count, GLenum type, const GLvoid *indices,
draw_elements_async(ctx, mode, count, type, indices, instance_count,
basevertex, baseinstance, index_bounds_valid,
min_index, max_index, index_buffer,
- non_vbo_attrib_mask, attribs);
+ user_buffer_mask, buffers);
return;
sync:
@@ -617,7 +617,7 @@ struct marshal_cmd_MultiDrawElementsBaseVertex
GLenum mode;
GLenum type;
GLsizei draw_count;
- GLuint non_vbo_attrib_mask;
+ GLuint user_buffer_mask;
struct gl_buffer_object *index_buffer;
};
@@ -628,7 +628,7 @@ _mesa_unmarshal_MultiDrawElementsBaseVertex(struct gl_context *ctx,
const GLenum mode = cmd->mode;
const GLenum type = cmd->type;
const GLsizei draw_count = cmd->draw_count;
- const GLuint non_vbo_attrib_mask = cmd->non_vbo_attrib_mask;
+ const GLuint user_buffer_mask = cmd->user_buffer_mask;
struct gl_buffer_object *index_buffer = cmd->index_buffer;
const bool has_base_vertex = cmd->has_base_vertex;
@@ -642,12 +642,12 @@ _mesa_unmarshal_MultiDrawElementsBaseVertex(struct gl_context *ctx,
basevertex = (GLsizei *)variable_data;
variable_data += sizeof(GLsizei) * draw_count;
}
- const struct glthread_attrib_binding *attribs =
+ const struct glthread_attrib_binding *buffers =
(const struct glthread_attrib_binding *)variable_data;
/* Bind uploaded buffers if needed. */
- if (non_vbo_attrib_mask) {
- _mesa_InternalBindVertexBuffers(ctx, attribs, non_vbo_attrib_mask,
+ if (user_buffer_mask) {
+ _mesa_InternalBindVertexBuffers(ctx, buffers, user_buffer_mask,
false);
}
if (index_buffer) {
@@ -668,8 +668,8 @@ _mesa_unmarshal_MultiDrawElementsBaseVertex(struct gl_context *ctx,
if (index_buffer) {
_mesa_InternalBindElementBuffer(ctx, NULL);
}
- if (non_vbo_attrib_mask) {
- _mesa_InternalBindVertexBuffers(ctx, attribs, non_vbo_attrib_mask,
+ if (user_buffer_mask) {
+ _mesa_InternalBindVertexBuffers(ctx, buffers, user_buffer_mask,
true);
}
}
@@ -680,22 +680,22 @@ multi_draw_elements_async(struct gl_context *ctx, GLenum mode,
const GLvoid *const *indices, GLsizei draw_count,
const GLsizei *basevertex,
struct gl_buffer_object *index_buffer,
- unsigned non_vbo_attrib_mask,
- const struct glthread_attrib_binding *attribs)
+ unsigned user_buffer_mask,
+ const struct glthread_attrib_binding *buffers)
{
int count_size = sizeof(GLsizei) * draw_count;
int indices_size = sizeof(indices[0]) * draw_count;
int basevertex_size = basevertex ? sizeof(GLsizei) * draw_count : 0;
- int attribs_size = util_bitcount(non_vbo_attrib_mask) * sizeof(attribs[0]);
+ int buffers_size = util_bitcount(user_buffer_mask) * sizeof(buffers[0]);
int cmd_size = sizeof(struct marshal_cmd_MultiDrawElementsBaseVertex) +
- count_size + indices_size + basevertex_size + attribs_size;
+ count_size + indices_size + basevertex_size + buffers_size;
struct marshal_cmd_MultiDrawElementsBaseVertex *cmd;
cmd = _mesa_glthread_allocate_command(ctx, DISPATCH_CMD_MultiDrawElementsBaseVertex, cmd_size);
cmd->mode = mode;
cmd->type = type;
cmd->draw_count = draw_count;
- cmd->non_vbo_attrib_mask = non_vbo_attrib_mask;
+ cmd->user_buffer_mask = user_buffer_mask;
cmd->index_buffer = index_buffer;
cmd->has_base_vertex = basevertex != NULL;
@@ -710,8 +710,8 @@ multi_draw_elements_async(struct gl_context *ctx, GLenum mode,
variable_data += basevertex_size;
}
- if (non_vbo_attrib_mask)
- memcpy(variable_data, attribs, attribs_size);
+ if (user_buffer_mask)
+ memcpy(variable_data, buffers, buffers_size);
}
void GLAPIENTRY
@@ -724,7 +724,7 @@ _mesa_marshal_MultiDrawElementsBaseVertex(GLenum mode, const GLsizei *count,
GET_CURRENT_CONTEXT(ctx);
struct glthread_vao *vao = ctx->GLThread.CurrentVAO;
- unsigned non_vbo_attrib_mask = vao->UserPointerMask & vao->Enabled;
+ unsigned user_buffer_mask = vao->UserPointerMask & vao->Enabled;
bool has_user_indices = vao->CurrentElementBufferName == 0;
if (ctx->GLThread.inside_dlist)
@@ -734,13 +734,13 @@ _mesa_marshal_MultiDrawElementsBaseVertex(GLenum mode, const GLsizei *count,
if (draw_count >= 0 &&
(ctx->API == API_OPENGL_CORE ||
!is_index_type_valid(type) ||
- (!non_vbo_attrib_mask && !has_user_indices))) {
+ (!user_buffer_mask && !has_user_indices))) {
multi_draw_elements_async(ctx, mode, count, type, indices, draw_count,
basevertex, 0, 0, NULL);
return;
}
- bool need_index_bounds = non_vbo_attrib_mask & ~vao->NonZeroDivisorMask;
+ bool need_index_bounds = user_buffer_mask & ~vao->NonZeroDivisorMask;
/* If the draw count is too high or negative, the queue can't be used.
*
@@ -829,10 +829,10 @@ _mesa_marshal_MultiDrawElementsBaseVertex(GLenum mode, const GLsizei *count,
}
/* Upload vertices. */
- struct glthread_attrib_binding attribs[VERT_ATTRIB_MAX];
- if (non_vbo_attrib_mask &&
- !upload_vertices(ctx, non_vbo_attrib_mask, min_index, num_vertices,
- 0, 1, attribs))
+ struct glthread_attrib_binding buffers[VERT_ATTRIB_MAX];
+ if (user_buffer_mask &&
+ !upload_vertices(ctx, user_buffer_mask, min_index, num_vertices,
+ 0, 1, buffers))
goto sync;
/* Upload indices. */
@@ -848,8 +848,8 @@ _mesa_marshal_MultiDrawElementsBaseVertex(GLenum mode, const GLsizei *count,
/* Draw asynchronously. */
multi_draw_elements_async(ctx, mode, count, type, indices, draw_count,
- basevertex, index_buffer, non_vbo_attrib_mask,
- attribs);
+ basevertex, index_buffer, user_buffer_mask,
+ buffers);
return;
sync:
diff --git a/src/mesa/main/varray.c b/src/mesa/main/varray.c
index 81068d00d0b..c50fd8da646 100644
--- a/src/mesa/main/varray.c
+++ b/src/mesa/main/varray.c
@@ -3261,31 +3261,31 @@ _mesa_BindVertexBuffers(GLuint first, GLsizei count, const GLuint *buffers,
void
_mesa_InternalBindVertexBuffers(struct gl_context *ctx,
- const struct glthread_attrib_binding *attribs,
- GLbitfield attrib_mask,
+ const struct glthread_attrib_binding *buffers,
+ GLbitfield buffer_mask,
GLboolean restore_pointers)
{
struct gl_vertex_array_object *vao = ctx->Array.VAO;
unsigned param_index = 0;
if (restore_pointers) {
- while (attrib_mask) {
- unsigned i = u_bit_scan(&attrib_mask);
+ while (buffer_mask) {
+ unsigned i = u_bit_scan(&buffer_mask);
_mesa_bind_vertex_buffer(ctx, vao, i, NULL,
- (GLintptr)attribs[param_index].original_pointer,
+ (GLintptr)buffers[param_index].original_pointer,
vao->BufferBinding[i].Stride, false, false);
param_index++;
}
return;
}
- while (attrib_mask) {
- unsigned i = u_bit_scan(&attrib_mask);
- struct gl_buffer_object *buf = attribs[param_index].buffer;
+ while (buffer_mask) {
+ unsigned i = u_bit_scan(&buffer_mask);
+ struct gl_buffer_object *buf = buffers[param_index].buffer;
/* The buffer reference is passed to _mesa_bind_vertex_buffer. */
- _mesa_bind_vertex_buffer(ctx, vao, i, buf, attribs[param_index].offset,
+ _mesa_bind_vertex_buffer(ctx, vao, i, buf, buffers[param_index].offset,
vao->BufferBinding[i].Stride, true, true);
param_index++;
}
diff --git a/src/mesa/main/varray.h b/src/mesa/main/varray.h
index 48c257993eb..4508f9cd6b3 100644
--- a/src/mesa/main/varray.h
+++ b/src/mesa/main/varray.h
@@ -379,8 +379,8 @@ _mesa_BindVertexBuffers(GLuint first, GLsizei count, const GLuint *buffers,
void
_mesa_InternalBindVertexBuffers(struct gl_context *ctx,
- const struct glthread_attrib_binding *attribs,
- GLbitfield attrib_mask,
+ const struct glthread_attrib_binding *buffers,
+ GLbitfield buffer_mask,
GLboolean restore_pointers);
void GLAPIENTRY