X-Git-Url: https://git.libre-soc.org/?a=blobdiff_plain;f=src%2Fmesa%2Fdrivers%2Fdri%2Fi965%2Fbrw_context.h;h=1a4efa3cb0cbd9385ba664e0c8e6274ef074b0f7;hb=9c9f45b82410646d2f7a8576d03de9916118bf07;hp=948b08207ab8b3aa77cca720fea69f4042701b07;hpb=c54e5c2fb22795bcce1ff18f2a0a70d6a4c647a4;p=mesa.git diff --git a/src/mesa/drivers/dri/i965/brw_context.h b/src/mesa/drivers/dri/i965/brw_context.h index 948b08207ab..1a4efa3cb0c 100644 --- a/src/mesa/drivers/dri/i965/brw_context.h +++ b/src/mesa/drivers/dri/i965/brw_context.h @@ -40,6 +40,8 @@ #include "brw_compiler.h" #include "intel_aub.h" +#include "isl/isl.h" + #ifdef __cplusplus extern "C" { /* Evil hack for using libdrm in a c++ compiler. */ @@ -366,7 +368,7 @@ struct brw_compute_program { struct brw_shader { - struct gl_shader base; + struct gl_linked_shader base; bool compiled_once; }; @@ -390,7 +392,7 @@ struct interpolation_mode_map { 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; @@ -399,7 +401,7 @@ static inline bool brw_any_flat_varyings(struct interpolation_mode_map *map) 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; @@ -569,6 +571,7 @@ struct brw_vertex_buffer { /** 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; @@ -619,6 +622,7 @@ struct intel_batchbuffer { uint32_t state_batch_offset; enum brw_gpu_ring ring; bool needs_sol_reset; + bool state_base_address_emitted; struct { uint32_t *map_next; @@ -672,9 +676,28 @@ struct brw_stage_state /** * 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; @@ -721,35 +744,10 @@ struct brw_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 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, @@ -855,7 +853,7 @@ struct brw_context */ bool perf_debug; - uint32_t max_gtt_map_object_size; + uint64_t max_gtt_map_object_size; int gen; int gt; @@ -893,6 +891,8 @@ struct brw_context */ bool needs_unlit_centroid_workaround; + struct isl_device isl_dev; + GLuint NewGLState; struct { struct brw_state_flags pipelines[BRW_NUM_PIPELINES]; @@ -962,6 +962,7 @@ struct brw_context /* 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 @@ -988,6 +989,7 @@ struct brw_context /* 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 @@ -1309,6 +1311,7 @@ struct brw_context uint32_t num_instances; int basevertex; + int baseinstance; struct { const struct brw_l3_config *config; @@ -1372,38 +1375,9 @@ GLboolean brwCreateContext(gl_api api, /*====================================================================== * 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 @@ -1427,12 +1401,17 @@ void brw_init_common_queryobj_functions(struct dd_function_table *functions); void gen4_init_queryobj_functions(struct dd_function_table *functions); void brw_emit_query_begin(struct brw_context *brw); void brw_emit_query_end(struct brw_context *brw); +void brw_query_counter(struct gl_context *ctx, struct gl_query_object *q); +bool brw_is_query_pipelined(struct brw_query_object *query); /** gen6_queryobj.c */ void gen6_init_queryobj_functions(struct dd_function_table *functions); void brw_write_timestamp(struct brw_context *brw, drm_intel_bo *bo, int idx); void brw_write_depth_count(struct brw_context *brw, drm_intel_bo *bo, int idx); +/** hsw_queryobj.c */ +void hsw_init_queryobj_functions(struct dd_function_table *functions); + /** brw_conditional_render.c */ void brw_init_conditional_render_functions(struct dd_function_table *functions); bool brw_check_conditional_render(struct brw_context *brw); @@ -1448,8 +1427,22 @@ void brw_load_register_mem64(struct brw_context *brw, drm_intel_bo *bo, uint32_t read_domains, uint32_t write_domain, uint32_t offset); +void brw_store_register_mem32(struct brw_context *brw, + drm_intel_bo *bo, uint32_t reg, uint32_t offset); void brw_store_register_mem64(struct brw_context *brw, drm_intel_bo *bo, uint32_t reg, uint32_t offset); +void brw_load_register_imm32(struct brw_context *brw, + uint32_t reg, uint32_t imm); +void brw_load_register_imm64(struct brw_context *brw, + uint32_t reg, uint64_t imm); +void brw_load_register_reg(struct brw_context *brw, uint32_t src, + uint32_t dest); +void brw_load_register_reg64(struct brw_context *brw, uint32_t src, + uint32_t dest); +void brw_store_data_imm32(struct brw_context *brw, drm_intel_bo *bo, + uint32_t offset, uint32_t imm); +void brw_store_data_imm64(struct brw_context *brw, drm_intel_bo *bo, + uint32_t offset, uint64_t imm); /*====================================================================== * brw_state_dump.c @@ -1458,7 +1451,7 @@ void brw_debug_batch(struct brw_context *brw); void brw_annotate_aub(struct brw_context *brw); /*====================================================================== - * brw_tex.c + * intel_tex_validate.c */ void brw_validate_textures( struct brw_context *brw ); @@ -1482,10 +1475,14 @@ void brwInitFragProgFuncs( struct dd_function_table *functions ); 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, @@ -1563,23 +1560,21 @@ brw_update_sol_surface(struct brw_context *brw, 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 */ @@ -1637,6 +1632,20 @@ void 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, @@ -1683,20 +1692,14 @@ gen7_emit_push_constant_state(struct brw_context *brw, unsigned vs_size, 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); - +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 @@ -1792,7 +1795,6 @@ brw_program_reloc(struct brw_context *brw, uint32_t state_offset, 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];