#include "brw_compiler.h"
#include "intel_aub.h"
+#include "isl/isl.h"
+#include "blorp/blorp.h"
+
#ifdef __cplusplus
extern "C" {
/* Evil hack for using libdrm in a c++ compiler. */
BRW_STATE_URB_SIZE,
BRW_STATE_CC_STATE,
BRW_STATE_BLORP,
+ BRW_STATE_VIEWPORT_COUNT,
BRW_NUM_STATE_BITS
};
#define BRW_NEW_PROGRAM_CACHE (1ull << BRW_STATE_PROGRAM_CACHE)
#define BRW_NEW_STATE_BASE_ADDRESS (1ull << BRW_STATE_STATE_BASE_ADDRESS)
#define BRW_NEW_VUE_MAP_GEOM_OUT (1ull << BRW_STATE_VUE_MAP_GEOM_OUT)
+#define BRW_NEW_VIEWPORT_COUNT (1ull << BRW_STATE_VIEWPORT_COUNT)
#define BRW_NEW_TRANSFORM_FEEDBACK (1ull << BRW_STATE_TRANSFORM_FEEDBACK)
#define BRW_NEW_RASTERIZER_DISCARD (1ull << BRW_STATE_RASTERIZER_DISCARD)
#define BRW_NEW_STATS_WM (1ull << BRW_STATE_STATS_WM)
struct brw_shader {
- struct gl_shader base;
+ struct gl_linked_shader base;
bool compiled_once;
};
static inline bool brw_any_flat_varyings(struct interpolation_mode_map *map)
{
for (int i = 0; i < BRW_VARYING_SLOT_COUNT; i++)
- if (map->mode[i] == INTERP_QUALIFIER_FLAT)
+ if (map->mode[i] == INTERP_MODE_FLAT)
return true;
return false;
static inline bool brw_any_noperspective_varyings(struct interpolation_mode_map *map)
{
for (int i = 0; i < BRW_VARYING_SLOT_COUNT; i++)
- if (map->mode[i] == INTERP_QUALIFIER_NOPERSPECTIVE)
+ if (map->mode[i] == INTERP_MODE_NOPERSPECTIVE)
return true;
return false;
/** Buffer object containing the uploaded vertex data */
drm_intel_bo *bo;
uint32_t offset;
+ uint32_t size;
/** Byte stride between elements in the uploaded array */
GLuint stride;
GLuint step_rate;
uint32_t state_batch_offset;
enum brw_gpu_ring ring;
bool needs_sol_reset;
+ bool state_base_address_emitted;
struct {
uint32_t *map_next;
/**
* Optional scratch buffer used to store spilled register values and
* variably-indexed GRF arrays.
+ *
+ * The contents of this buffer are short-lived so the same memory can be
+ * re-used at will for multiple shader programs (executed by the same fixed
+ * function). However reusing a scratch BO for which shader invocations
+ * are still in flight with a per-thread scratch slot size other than the
+ * original can cause threads with different scratch slot size and FFTID
+ * (which may be executed in parallel depending on the shader stage and
+ * hardware generation) to map to an overlapping region of the scratch
+ * space, which can potentially lead to mutual scratch space corruption.
+ * For that reason if you borrow this scratch buffer you should only be
+ * using the slot size given by the \c per_thread_scratch member below,
+ * unless you're taking additional measures to synchronize thread execution
+ * across slot size changes.
*/
drm_intel_bo *scratch_bo;
+ /**
+ * Scratch slot size allocated for each thread in the buffer object given
+ * by \c scratch_bo.
+ */
+ uint32_t per_thread_scratch;
+
/** Offset in the program cache to the program */
uint32_t prog_offset;
struct shader_times;
-struct brw_l3_config;
+struct gen_l3_config;
/**
* brw_context is derived from gl_context.
struct
{
- void (*update_texture_surface)(struct gl_context *ctx,
- unsigned unit,
- uint32_t *surf_offset,
- bool for_gather);
uint32_t (*update_renderbuffer_surface)(struct brw_context *brw,
struct gl_renderbuffer *rb,
- bool layered, unsigned unit,
+ uint32_t flags, unsigned unit,
uint32_t surf_index);
-
- void (*emit_texture_surface_state)(struct brw_context *brw,
- struct intel_mipmap_tree *mt,
- GLenum target,
- unsigned min_layer,
- unsigned max_layer,
- unsigned min_level,
- unsigned max_level,
- unsigned format,
- unsigned swizzle,
- uint32_t *surf_offset,
- int surf_index,
- bool rw, bool for_gather);
- void (*emit_buffer_surface_state)(struct brw_context *brw,
- uint32_t *out_offset,
- drm_intel_bo *bo,
- unsigned buffer_offset,
- unsigned surface_format,
- unsigned buffer_size,
- unsigned pitch,
- bool rw);
void (*emit_null_surface_state)(struct brw_context *brw,
unsigned width,
unsigned height,
*/
bool perf_debug;
- uint32_t max_gtt_map_object_size;
+ uint64_t max_gtt_map_object_size;
int gen;
int gt;
*/
bool needs_unlit_centroid_workaround;
+ struct isl_device isl_dev;
+
+ struct blorp_context blorp;
+
GLuint NewGLState;
struct {
struct brw_state_flags pipelines[BRW_NUM_PIPELINES];
/* Summary of size and varying of active arrays, so we can check
* for changes to this state:
*/
+ bool index_bounds_valid;
unsigned int min_index, max_index;
/* Offset from start of vertex buffer so we can avoid redefining
/* Updates are signaled by BRW_NEW_INDEX_BUFFER. */
drm_intel_bo *bo;
+ uint32_t size;
GLuint type;
/* Offset to index buffer index to use in CMD_3D_PRIM so that we can
*/
int num_samples;
- /**
- * Platform specific constants containing the maximum number of threads
- * for each pipeline stage.
- */
- unsigned max_vs_threads;
- unsigned max_hs_threads;
- unsigned max_ds_threads;
- unsigned max_gs_threads;
- unsigned max_wm_threads;
- unsigned max_cs_threads;
-
/* BRW_NEW_URB_ALLOCATIONS:
*/
struct {
bool constrained;
- GLuint min_vs_entries; /* Minimum number of VS entries */
- GLuint max_vs_entries; /* Maximum number of VS entries */
- GLuint max_hs_entries; /* Maximum number of HS entries */
- GLuint max_ds_entries; /* Maximum number of DS entries */
- GLuint max_gs_entries; /* Maximum number of GS entries */
-
GLuint nr_vs_entries;
GLuint nr_hs_entries;
GLuint nr_ds_entries;
GLuint cs_start;
/**
* URB size in the current configuration. The units this is expressed
- * in are somewhat inconsistent, see brw_device_info::urb::size.
+ * in are somewhat inconsistent, see gen_device_info::urb::size.
*
* FINISHME: Represent the URB size consistently in KB on all platforms.
*/
* instead of vp_bo.
*/
uint32_t vp_offset;
+
+ /**
+ * The number of viewports to use. If gl_ViewportIndex is written,
+ * we can have up to ctx->Const.MaxViewports viewports. If not,
+ * the viewport index is always 0, so we can only emit one.
+ */
+ uint8_t viewport_count;
} clip;
uint32_t num_instances;
int basevertex;
+ int baseinstance;
struct {
- const struct brw_l3_config *config;
+ const struct gen_l3_config *config;
} l3;
struct {
struct brw_fast_clear_state *fast_clear_state;
+ /* Array of flags telling if auxiliary buffer is disabled for corresponding
+ * renderbuffer. If draw_aux_buffer_disabled[i] is set then use of
+ * auxiliary buffer for gl_framebuffer::_ColorDrawBuffers[i] is
+ * disabled.
+ * This is needed in case the same underlying buffer is also configured
+ * to be sampled but with a format that the sampling engine can't treat
+ * compressed or fast cleared.
+ */
+ bool draw_aux_buffer_disabled[MAX_DRAW_BUFFERS];
+
__DRIcontext *driContext;
- struct intel_screen *intelScreen;
+ struct intel_screen *screen;
};
/*======================================================================
extern const char *const brw_vendor_string;
extern const char *
-brw_get_renderer_string(const struct intel_screen *intelScreen);
+brw_get_renderer_string(const struct intel_screen *screen);
enum {
DRI_CONF_BO_REUSE_DISABLED,
/*======================================================================
* brw_misc_state.c
*/
-struct gl_renderbuffer *brw_get_rb_for_slice(struct brw_context *brw,
- struct intel_mipmap_tree *mt,
- unsigned level, unsigned layer,
- bool flat);
-
-void brw_meta_updownsample(struct brw_context *brw,
- struct intel_mipmap_tree *src,
- struct intel_mipmap_tree *dst);
-
-void brw_meta_fbo_stencil_blit(struct brw_context *brw,
- struct gl_framebuffer *read_fb,
- struct gl_framebuffer *draw_fb,
- GLfloat srcX0, GLfloat srcY0,
- GLfloat srcX1, GLfloat srcY1,
- GLfloat dstX0, GLfloat dstY0,
- GLfloat dstX1, GLfloat dstY1);
-
-void brw_meta_stencil_updownsample(struct brw_context *brw,
- struct intel_mipmap_tree *src,
- struct intel_mipmap_tree *dst);
-
-bool brw_meta_fast_clear(struct brw_context *brw,
- struct gl_framebuffer *fb,
- GLbitfield mask,
- bool partial_clear);
-
void
brw_meta_resolve_color(struct brw_context *brw,
struct intel_mipmap_tree *mt);
-void
-brw_meta_fast_clear_free(struct brw_context *brw);
-
/*======================================================================
* brw_misc_state.c
void brw_annotate_aub(struct brw_context *brw);
/*======================================================================
- * brw_tex.c
+ * intel_tex_validate.c
*/
void brw_validate_textures( struct brw_context *brw );
static inline int
brw_get_scratch_size(int size)
{
- return util_next_power_of_two(size | 1023);
+ return MAX2(1024, util_next_power_of_two(size));
}
void brw_get_scratch_bo(struct brw_context *brw,
drm_intel_bo **scratch_bo, int size);
+void brw_alloc_stage_scratch(struct brw_context *brw,
+ struct brw_stage_state *stage_state,
+ unsigned per_thread_size,
+ unsigned thread_count);
void brw_init_shader_time(struct brw_context *brw);
int brw_get_shader_time_index(struct brw_context *brw,
struct gl_shader_program *shader_prog,
void brw_vec4_alloc_reg_set(struct brw_compiler *compiler);
/* brw_disasm.c */
-int brw_disassemble_inst(FILE *file, const struct brw_device_info *devinfo,
+int brw_disassemble_inst(FILE *file, const struct gen_device_info *devinfo,
struct brw_inst *inst, bool is_compacted);
/* brw_vs.c */
uint32_t *out_offset, unsigned num_vector_components,
unsigned stride_dwords, unsigned offset_dwords);
void brw_upload_ubo_surfaces(struct brw_context *brw,
- struct gl_shader *shader,
+ struct gl_linked_shader *shader,
struct brw_stage_state *stage_state,
struct brw_stage_prog_data *prog_data);
void brw_upload_abo_surfaces(struct brw_context *brw,
- struct gl_shader *shader,
+ struct gl_linked_shader *shader,
struct brw_stage_state *stage_state,
struct brw_stage_prog_data *prog_data);
void brw_upload_image_surfaces(struct brw_context *brw,
- struct gl_shader *shader,
+ struct gl_linked_shader *shader,
struct brw_stage_state *stage_state,
struct brw_stage_prog_data *prog_data);
/* brw_surface_formats.c */
bool brw_render_target_supported(struct brw_context *brw,
struct gl_renderbuffer *rb);
-bool brw_losslessly_compressible_format(const struct brw_context *brw,
- uint32_t brw_format);
uint32_t brw_depth_format(struct brw_context *brw, mesa_format format);
/* brw_performance_monitor.c */
gen7_resume_transform_feedback(struct gl_context *ctx,
struct gl_transform_feedback_object *obj);
+/* hsw_sol.c */
+void
+hsw_begin_transform_feedback(struct gl_context *ctx, GLenum mode,
+ struct gl_transform_feedback_object *obj);
+void
+hsw_end_transform_feedback(struct gl_context *ctx,
+ struct gl_transform_feedback_object *obj);
+void
+hsw_pause_transform_feedback(struct gl_context *ctx,
+ struct gl_transform_feedback_object *obj);
+void
+hsw_resume_transform_feedback(struct gl_context *ctx,
+ struct gl_transform_feedback_object *obj);
+
/* brw_blorp_blit.cpp */
GLbitfield
brw_blorp_framebuffer(struct brw_context *brw,
unsigned gs_size, unsigned fs_size);
void
-gen7_emit_urb_state(struct brw_context *brw,
- unsigned nr_vs_entries,
- unsigned vs_size, unsigned vs_start,
- unsigned nr_hs_entries,
- unsigned hs_size, unsigned hs_start,
- unsigned nr_ds_entries,
- unsigned ds_size, unsigned ds_start,
- unsigned nr_gs_entries,
- unsigned gs_size, unsigned gs_start);
-
+gen6_upload_urb(struct brw_context *brw, unsigned vs_size,
+ bool gs_present, unsigned gs_size);
+void
+gen7_upload_urb(struct brw_context *brw, unsigned vs_size,
+ bool gs_present, bool tess_present);
/* brw_reset.c */
extern GLenum
brw_get_graphics_reset_status(struct gl_context *ctx);
+void
+brw_check_for_reset(struct brw_context *brw);
/* brw_compute.c */
extern void
bool brw_do_cubemap_normalize(struct exec_list *instructions);
bool brw_lower_texture_gradients(struct brw_context *brw,
struct exec_list *instructions);
-bool brw_do_lower_unnormalized_offset(struct exec_list *instructions);
extern const char * const conditional_modifier[16];
extern const char *const pred_ctrl_align16[16];
uint32_t tile_x, uint32_t tile_y);
void gen8_hiz_exec(struct brw_context *brw, struct intel_mipmap_tree *mt,
- unsigned int level, unsigned int layer, enum gen6_hiz_op op);
+ unsigned int level, unsigned int layer, enum blorp_hiz_op op);
uint32_t get_hw_prim_for_gl_prim(int mode);
/* brw_pipe_control.c */
int brw_init_pipe_control(struct brw_context *brw,
- const struct brw_device_info *info);
+ const struct gen_device_info *info);
void brw_fini_pipe_control(struct brw_context *brw);
void brw_emit_pipe_control_flush(struct brw_context *brw, uint32_t flags);