mesa/src/compiler/nir/nir_gather_info.c

/*
 * Copyright © 2015 Intel Corporation
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
 * to deal in the Software without restriction, including without limitation
 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice (including the next
 * paragraph) shall be included in all copies or substantial portions of the
 * Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
 * IN THE SOFTWARE.
 */

#include "nir.h"
#include "nir_deref.h"
#include "main/menums.h"

static bool
src_is_invocation_id(const nir_src *src)
{
   assert(src->is_ssa);
   if (src->ssa->parent_instr->type != nir_instr_type_intrinsic)
      return false;

   return nir_instr_as_intrinsic(src->ssa->parent_instr)->intrinsic ==
             nir_intrinsic_load_invocation_id;
}

static void
get_deref_info(nir_shader *shader, nir_variable *var, nir_deref_instr *deref,
               bool *cross_invocation, bool *indirect)
{
   *cross_invocation = false;
   *indirect = false;

   const bool is_arrayed = nir_is_arrayed_io(var, shader->info.stage);

   nir_deref_path path;
   nir_deref_path_init(&path, deref, NULL);
   assert(path.path[0]->deref_type == nir_deref_type_var);
   nir_deref_instr **p = &path.path[1];

   /* Vertex index is the outermost array index. */
   if (is_arrayed) {
      assert((*p)->deref_type == nir_deref_type_array);
      *cross_invocation = !src_is_invocation_id(&(*p)->arr.index);
      p++;
   }

   /* We always lower indirect dereferences for "compact" array vars. */
   if (!path.path[0]->var->data.compact) {
      /* Non-compact array vars: find out if they are indirect. */
      for (; *p; p++) {
         if ((*p)->deref_type == nir_deref_type_array) {
            *indirect |= !nir_src_is_const((*p)->arr.index);
         } else if ((*p)->deref_type == nir_deref_type_struct) {
            /* Struct indices are always constant. */
         } else {
            unreachable("Unsupported deref type");
         }
      }
   }

   nir_deref_path_finish(&path);
}

static void
set_io_mask(nir_shader *shader, nir_variable *var, int offset, int len,
            nir_deref_instr *deref, bool is_output_read)
{
   for (int i = 0; i < len; i++) {
      assert(var->data.location != -1);

      int idx = var->data.location + offset + i;
      bool is_patch_generic = var->data.patch &&
                              idx != VARYING_SLOT_TESS_LEVEL_INNER &&
                              idx != VARYING_SLOT_TESS_LEVEL_OUTER &&
                              idx != VARYING_SLOT_BOUNDING_BOX0 &&
                              idx != VARYING_SLOT_BOUNDING_BOX1;
      uint64_t bitfield;

      if (is_patch_generic) {
         assert(idx >= VARYING_SLOT_PATCH0 && idx < VARYING_SLOT_TESS_MAX);
         bitfield = BITFIELD64_BIT(idx - VARYING_SLOT_PATCH0);
      }
      else {
         assert(idx < VARYING_SLOT_MAX);
         bitfield = BITFIELD64_BIT(idx);
      }

      bool cross_invocation;
      bool indirect;
      get_deref_info(shader, var, deref, &cross_invocation, &indirect);

      if (var->data.mode == nir_var_shader_in) {
         if (is_patch_generic) {
            shader->info.patch_inputs_read |= bitfield;
            if (indirect)
               shader->info.patch_inputs_read_indirectly |= bitfield;
         } else {
            shader->info.inputs_read |= bitfield;
            if (indirect)
               shader->info.inputs_read_indirectly |= bitfield;
         }

         if (cross_invocation && shader->info.stage == MESA_SHADER_TESS_CTRL)
            shader->info.tess.tcs_cross_invocation_inputs_read |= bitfield;

         if (shader->info.stage == MESA_SHADER_FRAGMENT) {
            shader->info.fs.uses_sample_qualifier |= var->data.sample;
         }
      } else {
         assert(var->data.mode == nir_var_shader_out);
         if (is_output_read) {
            if (is_patch_generic) {
               shader->info.patch_outputs_read |= bitfield;
               if (indirect)
                  shader->info.patch_outputs_accessed_indirectly |= bitfield;
            } else {
               shader->info.outputs_read |= bitfield;
               if (indirect)
                  shader->info.outputs_accessed_indirectly |= bitfield;
            }

            if (cross_invocation && shader->info.stage == MESA_SHADER_TESS_CTRL)
               shader->info.tess.tcs_cross_invocation_outputs_read |= bitfield;
         } else {
            if (is_patch_generic) {
               shader->info.patch_outputs_written |= bitfield;
               if (indirect)
                  shader->info.patch_outputs_accessed_indirectly |= bitfield;
            } else if (!var->data.read_only) {
               shader->info.outputs_written |= bitfield;
               if (indirect)
                  shader->info.outputs_accessed_indirectly |= bitfield;
            }
         }


         if (var->data.fb_fetch_output) {
            shader->info.outputs_read |= bitfield;
            if (shader->info.stage == MESA_SHADER_FRAGMENT)
               shader->info.fs.uses_fbfetch_output = true;
         }

         if (shader->info.stage == MESA_SHADER_FRAGMENT &&
             !is_output_read && var->data.index == 1)
            shader->info.fs.color_is_dual_source = true;
      }
   }
}

/**
 * Mark an entire variable as used.  Caller must ensure that the variable
 * represents a shader input or output.
 */
static void
mark_whole_variable(nir_shader *shader, nir_variable *var,
                    nir_deref_instr *deref, bool is_output_read)
{
   const struct glsl_type *type = var->type;

   if (nir_is_arrayed_io(var, shader->info.stage)) {
      assert(glsl_type_is_array(type));
      type = glsl_get_array_element(type);
   }

   if (var->data.per_view) {
      /* TODO: Per view and Per Vertex are not currently used together.  When
       * they start to be used (e.g. when adding Primitive Replication for GS
       * on Intel), verify that "peeling" the type twice is correct.  This
       * assert ensures we remember it.
       */
      assert(!nir_is_arrayed_io(var, shader->info.stage));
      assert(glsl_type_is_array(type));
      type = glsl_get_array_element(type);
   }

   const unsigned slots =
      var->data.compact ? DIV_ROUND_UP(glsl_get_length(type), 4)
                        : glsl_count_attribute_slots(type, false);

   set_io_mask(shader, var, 0, slots, deref, is_output_read);
}

static unsigned
get_io_offset(nir_deref_instr *deref, nir_variable *var, bool is_arrayed)
{
   if (var->data.compact) {
      assert(deref->deref_type == nir_deref_type_array);
      return nir_src_is_const(deref->arr.index) ?
             (nir_src_as_uint(deref->arr.index) + var->data.location_frac) / 4u :
             (unsigned)-1;
   }

   unsigned offset = 0;

   for (nir_deref_instr *d = deref; d; d = nir_deref_instr_parent(d)) {
      if (d->deref_type == nir_deref_type_array) {
         if (is_arrayed && nir_deref_instr_parent(d)->deref_type == nir_deref_type_var)
            break;

         if (!nir_src_is_const(d->arr.index))
            return -1;

         offset += glsl_count_attribute_slots(d->type, false) *
                   nir_src_as_uint(d->arr.index);
      } else if (d->deref_type == nir_deref_type_struct) {
         const struct glsl_type *parent_type = nir_deref_instr_parent(d)->type;
         for (unsigned i = 0; i < d->strct.index; i++) {
            const struct glsl_type *field_type = glsl_get_struct_field(parent_type, i);
            offset += glsl_count_attribute_slots(field_type, false);
         }
      }
   }

   return offset;
}

/**
 * Try to mark a portion of the given varying as used.  Caller must ensure
 * that the variable represents a shader input or output.
 *
 * If the index can't be interpreted as a constant, or some other problem
 * occurs, then nothing will be marked and false will be returned.
 */
static bool
try_mask_partial_io(nir_shader *shader, nir_variable *var,
                    nir_deref_instr *deref, bool is_output_read)
{
   const struct glsl_type *type = var->type;
   bool is_arrayed = nir_is_arrayed_io(var, shader->info.stage);

   if (is_arrayed) {
      assert(glsl_type_is_array(type));
      type = glsl_get_array_element(type);
   }

   /* Per view variables will be considered as a whole. */
   if (var->data.per_view)
      return false;

   unsigned offset = get_io_offset(deref, var, is_arrayed);
   if (offset == -1)
      return false;

   const unsigned slots =
      var->data.compact ? DIV_ROUND_UP(glsl_get_length(type), 4)
                        : glsl_count_attribute_slots(type, false);

   if (offset >= slots) {
      /* Constant index outside the bounds of the matrix/array.  This could
       * arise as a result of constant folding of a legal GLSL program.
       *
       * Even though the spec says that indexing outside the bounds of a
       * matrix/array results in undefined behaviour, we don't want to pass
       * out-of-range values to set_io_mask() (since this could result in
       * slots that don't exist being marked as used), so just let the caller
       * mark the whole variable as used.
       */
      return false;
   }

   unsigned len = glsl_count_attribute_slots(deref->type, false);
   set_io_mask(shader, var, offset, len, deref, is_output_read);
   return true;
}

/** Returns true if the given intrinsic writes external memory
 *
 * Only returns true for writes to globally visible memory, not scratch and
 * not shared.
 */
bool
nir_intrinsic_writes_external_memory(const nir_intrinsic_instr *instr)
{
   switch (instr->intrinsic) {
   case nir_intrinsic_atomic_counter_inc:
   case nir_intrinsic_atomic_counter_inc_deref:
   case nir_intrinsic_atomic_counter_add:
   case nir_intrinsic_atomic_counter_add_deref:
   case nir_intrinsic_atomic_counter_pre_dec:
   case nir_intrinsic_atomic_counter_pre_dec_deref:
   case nir_intrinsic_atomic_counter_post_dec:
   case nir_intrinsic_atomic_counter_post_dec_deref:
   case nir_intrinsic_atomic_counter_min:
   case nir_intrinsic_atomic_counter_min_deref:
   case nir_intrinsic_atomic_counter_max:
   case nir_intrinsic_atomic_counter_max_deref:
   case nir_intrinsic_atomic_counter_and:
   case nir_intrinsic_atomic_counter_and_deref:
   case nir_intrinsic_atomic_counter_or:
   case nir_intrinsic_atomic_counter_or_deref:
   case nir_intrinsic_atomic_counter_xor:
   case nir_intrinsic_atomic_counter_xor_deref:
   case nir_intrinsic_atomic_counter_exchange:
   case nir_intrinsic_atomic_counter_exchange_deref:
   case nir_intrinsic_atomic_counter_comp_swap:
   case nir_intrinsic_atomic_counter_comp_swap_deref:
   case nir_intrinsic_bindless_image_atomic_add:
   case nir_intrinsic_bindless_image_atomic_and:
   case nir_intrinsic_bindless_image_atomic_comp_swap:
   case nir_intrinsic_bindless_image_atomic_dec_wrap:
   case nir_intrinsic_bindless_image_atomic_exchange:
   case nir_intrinsic_bindless_image_atomic_fadd:
   case nir_intrinsic_bindless_image_atomic_imax:
   case nir_intrinsic_bindless_image_atomic_imin:
   case nir_intrinsic_bindless_image_atomic_inc_wrap:
   case nir_intrinsic_bindless_image_atomic_or:
   case nir_intrinsic_bindless_image_atomic_umax:
   case nir_intrinsic_bindless_image_atomic_umin:
   case nir_intrinsic_bindless_image_atomic_xor:
   case nir_intrinsic_bindless_image_store:
   case nir_intrinsic_bindless_image_store_raw_intel:
   case nir_intrinsic_global_atomic_add:
   case nir_intrinsic_global_atomic_and:
   case nir_intrinsic_global_atomic_comp_swap:
   case nir_intrinsic_global_atomic_exchange:
   case nir_intrinsic_global_atomic_fadd:
   case nir_intrinsic_global_atomic_fcomp_swap:
   case nir_intrinsic_global_atomic_fmax:
   case nir_intrinsic_global_atomic_fmin:
   case nir_intrinsic_global_atomic_imax:
   case nir_intrinsic_global_atomic_imin:
   case nir_intrinsic_global_atomic_or:
   case nir_intrinsic_global_atomic_umax:
   case nir_intrinsic_global_atomic_umin:
   case nir_intrinsic_global_atomic_xor:
   case nir_intrinsic_image_atomic_add:
   case nir_intrinsic_image_atomic_and:
   case nir_intrinsic_image_atomic_comp_swap:
   case nir_intrinsic_image_atomic_dec_wrap:
   case nir_intrinsic_image_atomic_exchange:
   case nir_intrinsic_image_atomic_fadd:
   case nir_intrinsic_image_atomic_imax:
   case nir_intrinsic_image_atomic_imin:
   case nir_intrinsic_image_atomic_inc_wrap:
   case nir_intrinsic_image_atomic_or:
   case nir_intrinsic_image_atomic_umax:
   case nir_intrinsic_image_atomic_umin:
   case nir_intrinsic_image_atomic_xor:
   case nir_intrinsic_image_deref_atomic_add:
   case nir_intrinsic_image_deref_atomic_and:
   case nir_intrinsic_image_deref_atomic_comp_swap:
   case nir_intrinsic_image_deref_atomic_dec_wrap:
   case nir_intrinsic_image_deref_atomic_exchange:
   case nir_intrinsic_image_deref_atomic_fadd:
   case nir_intrinsic_image_deref_atomic_imax:
   case nir_intrinsic_image_deref_atomic_imin:
   case nir_intrinsic_image_deref_atomic_inc_wrap:
   case nir_intrinsic_image_deref_atomic_or:
   case nir_intrinsic_image_deref_atomic_umax:
   case nir_intrinsic_image_deref_atomic_umin:
   case nir_intrinsic_image_deref_atomic_xor:
   case nir_intrinsic_image_deref_store:
   case nir_intrinsic_image_deref_store_raw_intel:
   case nir_intrinsic_image_store:
   case nir_intrinsic_image_store_raw_intel:
   case nir_intrinsic_ssbo_atomic_add:
   case nir_intrinsic_ssbo_atomic_add_ir3:
   case nir_intrinsic_ssbo_atomic_and:
   case nir_intrinsic_ssbo_atomic_and_ir3:
   case nir_intrinsic_ssbo_atomic_comp_swap:
   case nir_intrinsic_ssbo_atomic_comp_swap_ir3:
   case nir_intrinsic_ssbo_atomic_exchange:
   case nir_intrinsic_ssbo_atomic_exchange_ir3:
   case nir_intrinsic_ssbo_atomic_fadd:
   case nir_intrinsic_ssbo_atomic_fcomp_swap:
   case nir_intrinsic_ssbo_atomic_fmax:
   case nir_intrinsic_ssbo_atomic_fmin:
   case nir_intrinsic_ssbo_atomic_imax:
   case nir_intrinsic_ssbo_atomic_imax_ir3:
   case nir_intrinsic_ssbo_atomic_imin:
   case nir_intrinsic_ssbo_atomic_imin_ir3:
   case nir_intrinsic_ssbo_atomic_or:
   case nir_intrinsic_ssbo_atomic_or_ir3:
   case nir_intrinsic_ssbo_atomic_umax:
   case nir_intrinsic_ssbo_atomic_umax_ir3:
   case nir_intrinsic_ssbo_atomic_umin:
   case nir_intrinsic_ssbo_atomic_umin_ir3:
   case nir_intrinsic_ssbo_atomic_xor:
   case nir_intrinsic_ssbo_atomic_xor_ir3:
   case nir_intrinsic_store_global:
   case nir_intrinsic_store_global_ir3:
   case nir_intrinsic_store_ssbo:
   case nir_intrinsic_store_ssbo_ir3:
      return true;

   case nir_intrinsic_store_deref:
   case nir_intrinsic_deref_atomic_add:
   case nir_intrinsic_deref_atomic_imin:
   case nir_intrinsic_deref_atomic_umin:
   case nir_intrinsic_deref_atomic_imax:
   case nir_intrinsic_deref_atomic_umax:
   case nir_intrinsic_deref_atomic_and:
   case nir_intrinsic_deref_atomic_or:
   case nir_intrinsic_deref_atomic_xor:
   case nir_intrinsic_deref_atomic_exchange:
   case nir_intrinsic_deref_atomic_comp_swap:
   case nir_intrinsic_deref_atomic_fadd:
   case nir_intrinsic_deref_atomic_fmin:
   case nir_intrinsic_deref_atomic_fmax:
   case nir_intrinsic_deref_atomic_fcomp_swap:
      return nir_deref_mode_may_be(nir_src_as_deref(instr->src[0]),
                                   nir_var_mem_ssbo | nir_var_mem_global);

   default:
      return false;
   }
}

static void
gather_intrinsic_info(nir_intrinsic_instr *instr, nir_shader *shader,
                      void *dead_ctx)
{
   uint64_t slot_mask = 0;
   uint16_t slot_mask_16bit = 0;

   if (nir_intrinsic_infos[instr->intrinsic].index_map[NIR_INTRINSIC_IO_SEMANTICS] > 0) {
      nir_io_semantics semantics = nir_intrinsic_io_semantics(instr);

      if (semantics.location >= VARYING_SLOT_PATCH0 &&
          semantics.location <= VARYING_SLOT_PATCH31) {
         /* Generic per-patch I/O. */
         assert((shader->info.stage == MESA_SHADER_TESS_EVAL &&
                 instr->intrinsic == nir_intrinsic_load_input) ||
                (shader->info.stage == MESA_SHADER_TESS_CTRL &&
                 (instr->intrinsic == nir_intrinsic_load_output ||
                  instr->intrinsic == nir_intrinsic_store_output)));

         semantics.location -= VARYING_SLOT_PATCH0;
      }

      if (semantics.location >= VARYING_SLOT_VAR0_16BIT &&
          semantics.location <= VARYING_SLOT_VAR15_16BIT) {
         /* Convert num_slots from the units of half vectors to full vectors. */
         unsigned num_slots = (semantics.num_slots + semantics.high_16bits + 1) / 2;
         slot_mask_16bit =
            BITFIELD_RANGE(semantics.location - VARYING_SLOT_VAR0_16BIT, num_slots);
      } else {
         slot_mask = BITFIELD64_RANGE(semantics.location, semantics.num_slots);
         assert(util_bitcount64(slot_mask) == semantics.num_slots);
      }
   }

   switch (instr->intrinsic) {
   case nir_intrinsic_demote:
   case nir_intrinsic_demote_if:
      shader->info.fs.uses_demote = true;
      FALLTHROUGH; /* quads with helper lanes only might be discarded entirely */
   case nir_intrinsic_discard:
   case nir_intrinsic_discard_if:
      /* Freedreno uses the discard_if intrinsic to end GS invocations that
       * don't produce a vertex, so we only set uses_discard if executing on
       * a fragment shader. */
      if (shader->info.stage == MESA_SHADER_FRAGMENT)
         shader->info.fs.uses_discard = true;
      break;

   case nir_intrinsic_terminate:
   case nir_intrinsic_terminate_if:
      assert(shader->info.stage == MESA_SHADER_FRAGMENT);
      shader->info.fs.uses_discard = true;
      break;

   case nir_intrinsic_interp_deref_at_centroid:
   case nir_intrinsic_interp_deref_at_sample:
   case nir_intrinsic_interp_deref_at_offset:
   case nir_intrinsic_interp_deref_at_vertex:
   case nir_intrinsic_load_deref:
   case nir_intrinsic_store_deref:{
      nir_deref_instr *deref = nir_src_as_deref(instr->src[0]);
      if (nir_deref_mode_is_one_of(deref, nir_var_shader_in |
                                          nir_var_shader_out)) {
         nir_variable *var = nir_deref_instr_get_variable(deref);
         bool is_output_read = false;
         if (var->data.mode == nir_var_shader_out &&
             instr->intrinsic == nir_intrinsic_load_deref)
            is_output_read = true;

         if (!try_mask_partial_io(shader, var, deref, is_output_read))
            mark_whole_variable(shader, var, deref, is_output_read);

         /* We need to track which input_reads bits correspond to a
          * dvec3/dvec4 input attribute */
         if (shader->info.stage == MESA_SHADER_VERTEX &&
             var->data.mode == nir_var_shader_in &&
             glsl_type_is_dual_slot(glsl_without_array(var->type))) {
            for (unsigned i = 0; i < glsl_count_attribute_slots(var->type, false); i++) {
               int idx = var->data.location + i;
               shader->info.vs.double_inputs |= BITFIELD64_BIT(idx);
            }
         }
      }
      if (nir_intrinsic_writes_external_memory(instr))
         shader->info.writes_memory = true;
      break;
   }

   case nir_intrinsic_load_input:
   case nir_intrinsic_load_per_vertex_input:
   case nir_intrinsic_load_input_vertex:
   case nir_intrinsic_load_interpolated_input:
      if (shader->info.stage == MESA_SHADER_TESS_EVAL &&
          instr->intrinsic == nir_intrinsic_load_input) {
         shader->info.patch_inputs_read |= slot_mask;
         if (!nir_src_is_const(*nir_get_io_offset_src(instr)))
            shader->info.patch_inputs_read_indirectly |= slot_mask;
      } else {
         shader->info.inputs_read |= slot_mask;
         shader->info.inputs_read_16bit |= slot_mask_16bit;
         if (!nir_src_is_const(*nir_get_io_offset_src(instr))) {
            shader->info.inputs_read_indirectly |= slot_mask;
            shader->info.inputs_read_indirectly_16bit |= slot_mask_16bit;
         }
      }

      if (shader->info.stage == MESA_SHADER_TESS_CTRL &&
          instr->intrinsic == nir_intrinsic_load_per_vertex_input &&
          !src_is_invocation_id(nir_get_io_vertex_index_src(instr)))
         shader->info.tess.tcs_cross_invocation_inputs_read |= slot_mask;
      break;

   case nir_intrinsic_load_output:
   case nir_intrinsic_load_per_vertex_output:
   case nir_intrinsic_load_per_primitive_output:
      if (shader->info.stage == MESA_SHADER_TESS_CTRL &&
          instr->intrinsic == nir_intrinsic_load_output) {
         shader->info.patch_outputs_read |= slot_mask;
         if (!nir_src_is_const(*nir_get_io_offset_src(instr)))
            shader->info.patch_outputs_accessed_indirectly |= slot_mask;
      } else {
         shader->info.outputs_read |= slot_mask;
         shader->info.outputs_read_16bit |= slot_mask_16bit;
         if (!nir_src_is_const(*nir_get_io_offset_src(instr))) {
            shader->info.outputs_accessed_indirectly |= slot_mask;
            shader->info.outputs_accessed_indirectly_16bit |= slot_mask_16bit;
         }
      }

      if (shader->info.stage == MESA_SHADER_TESS_CTRL &&
          instr->intrinsic == nir_intrinsic_load_per_vertex_output &&
          !src_is_invocation_id(nir_get_io_vertex_index_src(instr)))
         shader->info.tess.tcs_cross_invocation_outputs_read |= slot_mask;

      if (shader->info.stage == MESA_SHADER_FRAGMENT &&
          nir_intrinsic_io_semantics(instr).fb_fetch_output)
         shader->info.fs.uses_fbfetch_output = true;
      break;

   case nir_intrinsic_store_output:
   case nir_intrinsic_store_per_vertex_output:
   case nir_intrinsic_store_per_primitive_output:
      if (shader->info.stage == MESA_SHADER_TESS_CTRL &&
          instr->intrinsic == nir_intrinsic_store_output) {
         shader->info.patch_outputs_written |= slot_mask;
         if (!nir_src_is_const(*nir_get_io_offset_src(instr)))
            shader->info.patch_outputs_accessed_indirectly |= slot_mask;
      } else {
         shader->info.outputs_written |= slot_mask;
         shader->info.outputs_written_16bit |= slot_mask_16bit;
         if (!nir_src_is_const(*nir_get_io_offset_src(instr))) {
            shader->info.outputs_accessed_indirectly |= slot_mask;
            shader->info.outputs_accessed_indirectly_16bit |= slot_mask_16bit;
         }
      }

      if (shader->info.stage == MESA_SHADER_FRAGMENT &&
          nir_intrinsic_io_semantics(instr).dual_source_blend_index)
         shader->info.fs.color_is_dual_source = true;
      break;

   case nir_intrinsic_load_color0:
   case nir_intrinsic_load_color1:
      shader->info.inputs_read |=
         BITFIELD64_BIT(VARYING_SLOT_COL0 <<
                        (instr->intrinsic == nir_intrinsic_load_color1));
      FALLTHROUGH;
   case nir_intrinsic_load_subgroup_size:
   case nir_intrinsic_load_subgroup_invocation:
   case nir_intrinsic_load_subgroup_eq_mask:
   case nir_intrinsic_load_subgroup_ge_mask:
   case nir_intrinsic_load_subgroup_gt_mask:
   case nir_intrinsic_load_subgroup_le_mask:
   case nir_intrinsic_load_subgroup_lt_mask:
   case nir_intrinsic_load_num_subgroups:
   case nir_intrinsic_load_subgroup_id:
   case nir_intrinsic_load_vertex_id:
   case nir_intrinsic_load_instance_id:
   case nir_intrinsic_load_vertex_id_zero_base:
   case nir_intrinsic_load_base_vertex:
   case nir_intrinsic_load_first_vertex:
   case nir_intrinsic_load_is_indexed_draw:
   case nir_intrinsic_load_base_instance:
   case nir_intrinsic_load_draw_id:
   case nir_intrinsic_load_invocation_id:
   case nir_intrinsic_load_frag_coord:
   case nir_intrinsic_load_frag_shading_rate:
   case nir_intrinsic_load_point_coord:
   case nir_intrinsic_load_line_coord:
   case nir_intrinsic_load_front_face:
   case nir_intrinsic_load_sample_id:
   case nir_intrinsic_load_sample_pos:
   case nir_intrinsic_load_sample_mask_in:
   case nir_intrinsic_load_helper_invocation:
   case nir_intrinsic_load_tess_coord:
   case nir_intrinsic_load_patch_vertices_in:
   case nir_intrinsic_load_primitive_id:
   case nir_intrinsic_load_tess_level_outer:
   case nir_intrinsic_load_tess_level_inner:
   case nir_intrinsic_load_tess_level_outer_default:
   case nir_intrinsic_load_tess_level_inner_default:
   case nir_intrinsic_load_local_invocation_id:
   case nir_intrinsic_load_local_invocation_index:
   case nir_intrinsic_load_global_invocation_id:
   case nir_intrinsic_load_base_global_invocation_id:
   case nir_intrinsic_load_global_invocation_index:
   case nir_intrinsic_load_workgroup_id:
   case nir_intrinsic_load_num_workgroups:
   case nir_intrinsic_load_workgroup_size:
   case nir_intrinsic_load_work_dim:
   case nir_intrinsic_load_user_data_amd:
   case nir_intrinsic_load_view_index:
   case nir_intrinsic_load_barycentric_model:
   case nir_intrinsic_load_gs_header_ir3:
   case nir_intrinsic_load_tcs_header_ir3:
      BITSET_SET(shader->info.system_values_read,
                 nir_system_value_from_intrinsic(instr->intrinsic));
      break;

   case nir_intrinsic_load_barycentric_pixel:
      if (nir_intrinsic_interp_mode(instr) == INTERP_MODE_SMOOTH ||
          nir_intrinsic_interp_mode(instr) == INTERP_MODE_NONE) {
         BITSET_SET(shader->info.system_values_read,
                    SYSTEM_VALUE_BARYCENTRIC_PERSP_PIXEL);
      } else if (nir_intrinsic_interp_mode(instr) == INTERP_MODE_NOPERSPECTIVE) {
         BITSET_SET(shader->info.system_values_read,
                    SYSTEM_VALUE_BARYCENTRIC_LINEAR_PIXEL);
      }
      break;

   case nir_intrinsic_load_barycentric_centroid:
      if (nir_intrinsic_interp_mode(instr) == INTERP_MODE_SMOOTH ||
          nir_intrinsic_interp_mode(instr) == INTERP_MODE_NONE) {
         BITSET_SET(shader->info.system_values_read,
                    SYSTEM_VALUE_BARYCENTRIC_PERSP_CENTROID);
      } else if (nir_intrinsic_interp_mode(instr) == INTERP_MODE_NOPERSPECTIVE) {
         BITSET_SET(shader->info.system_values_read,
                    SYSTEM_VALUE_BARYCENTRIC_LINEAR_CENTROID);
      }
      break;

   case nir_intrinsic_load_barycentric_sample:
      if (nir_intrinsic_interp_mode(instr) == INTERP_MODE_SMOOTH ||
          nir_intrinsic_interp_mode(instr) == INTERP_MODE_NONE) {
         BITSET_SET(shader->info.system_values_read,
                    SYSTEM_VALUE_BARYCENTRIC_PERSP_SAMPLE);
      } else if (nir_intrinsic_interp_mode(instr) == INTERP_MODE_NOPERSPECTIVE) {
         BITSET_SET(shader->info.system_values_read,
                    SYSTEM_VALUE_BARYCENTRIC_LINEAR_SAMPLE);
      }
      if (shader->info.stage == MESA_SHADER_FRAGMENT)
         shader->info.fs.uses_sample_qualifier = true;
      break;

   case nir_intrinsic_quad_broadcast:
   case nir_intrinsic_quad_swap_horizontal:
   case nir_intrinsic_quad_swap_vertical:
   case nir_intrinsic_quad_swap_diagonal:
   case nir_intrinsic_quad_swizzle_amd:
      if (shader->info.stage == MESA_SHADER_FRAGMENT)
         shader->info.fs.needs_quad_helper_invocations = true;
      break;

   case nir_intrinsic_vote_any:
   case nir_intrinsic_vote_all:
   case nir_intrinsic_vote_feq:
   case nir_intrinsic_vote_ieq:
   case nir_intrinsic_ballot:
   case nir_intrinsic_ballot_bit_count_exclusive:
   case nir_intrinsic_ballot_bit_count_inclusive:
   case nir_intrinsic_ballot_bitfield_extract:
   case nir_intrinsic_ballot_bit_count_reduce:
   case nir_intrinsic_ballot_find_lsb:
   case nir_intrinsic_ballot_find_msb:
   case nir_intrinsic_first_invocation:
   case nir_intrinsic_read_invocation:
   case nir_intrinsic_read_first_invocation:
   case nir_intrinsic_elect:
   case nir_intrinsic_reduce:
   case nir_intrinsic_inclusive_scan:
   case nir_intrinsic_exclusive_scan:
   case nir_intrinsic_shuffle:
   case nir_intrinsic_shuffle_xor:
   case nir_intrinsic_shuffle_up:
   case nir_intrinsic_shuffle_down:
   case nir_intrinsic_write_invocation_amd:
      if (shader->info.stage == MESA_SHADER_FRAGMENT)
         shader->info.fs.needs_all_helper_invocations = true;
      if (shader->info.stage == MESA_SHADER_COMPUTE)
         shader->info.cs.uses_wide_subgroup_intrinsics = true;
      break;

   case nir_intrinsic_end_primitive:
   case nir_intrinsic_end_primitive_with_counter:
      assert(shader->info.stage == MESA_SHADER_GEOMETRY);
      shader->info.gs.uses_end_primitive = 1;
      FALLTHROUGH;

   case nir_intrinsic_emit_vertex:
   case nir_intrinsic_emit_vertex_with_counter:
      shader->info.gs.active_stream_mask |= 1 << nir_intrinsic_stream_id(instr);

      break;

   case nir_intrinsic_control_barrier:
      shader->info.uses_control_barrier = true;
      break;

   case nir_intrinsic_scoped_barrier:
      shader->info.uses_control_barrier |=
         nir_intrinsic_execution_scope(instr) != NIR_SCOPE_NONE;

      shader->info.uses_memory_barrier |=
         nir_intrinsic_memory_scope(instr) != NIR_SCOPE_NONE;
      break;

   case nir_intrinsic_memory_barrier:
   case nir_intrinsic_group_memory_barrier:
   case nir_intrinsic_memory_barrier_atomic_counter:
   case nir_intrinsic_memory_barrier_buffer:
   case nir_intrinsic_memory_barrier_image:
   case nir_intrinsic_memory_barrier_shared:
   case nir_intrinsic_memory_barrier_tcs_patch:
      shader->info.uses_memory_barrier = true;
      break;

   default:
      if (nir_intrinsic_writes_external_memory(instr))
         shader->info.writes_memory = true;
      break;
   }
}

static void
gather_tex_info(nir_tex_instr *instr, nir_shader *shader)
{
   if (shader->info.stage == MESA_SHADER_FRAGMENT &&
       nir_tex_instr_has_implicit_derivative(instr))
      shader->info.fs.needs_quad_helper_invocations = true;

   switch (instr->op) {
   case nir_texop_tg4:
      shader->info.uses_texture_gather = true;
      break;
   default:
      break;
   }
}

static void
gather_alu_info(nir_alu_instr *instr, nir_shader *shader)
{
   switch (instr->op) {
   case nir_op_fddx:
   case nir_op_fddy:
      shader->info.uses_fddx_fddy = true;
      FALLTHROUGH;
   case nir_op_fddx_fine:
   case nir_op_fddy_fine:
   case nir_op_fddx_coarse:
   case nir_op_fddy_coarse:
      if (shader->info.stage == MESA_SHADER_FRAGMENT)
         shader->info.fs.needs_quad_helper_invocations = true;
      break;
   default:
      break;
   }

   const nir_op_info *info = &nir_op_infos[instr->op];

   for (unsigned i = 0; i < info->num_inputs; i++) {
      if (nir_alu_type_get_base_type(info->input_types[i]) == nir_type_float)
         shader->info.bit_sizes_float |= nir_src_bit_size(instr->src[i].src);
      else
         shader->info.bit_sizes_int |= nir_src_bit_size(instr->src[i].src);
   }
   if (nir_alu_type_get_base_type(info->output_type) == nir_type_float)
      shader->info.bit_sizes_float |= nir_dest_bit_size(instr->dest.dest);
   else
      shader->info.bit_sizes_int |= nir_dest_bit_size(instr->dest.dest);
}

static void
gather_info_block(nir_block *block, nir_shader *shader, void *dead_ctx)
{
   nir_foreach_instr(instr, block) {
      switch (instr->type) {
      case nir_instr_type_alu:
         gather_alu_info(nir_instr_as_alu(instr), shader);
         break;
      case nir_instr_type_intrinsic:
         gather_intrinsic_info(nir_instr_as_intrinsic(instr), shader, dead_ctx);
         break;
      case nir_instr_type_tex:
         gather_tex_info(nir_instr_as_tex(instr), shader);
         break;
      case nir_instr_type_call:
         assert(!"nir_shader_gather_info only works if functions are inlined");
         break;
      default:
         break;
      }
   }
}

void
nir_shader_gather_info(nir_shader *shader, nir_function_impl *entrypoint)
{
   shader->info.num_textures = 0;
   shader->info.num_images = 0;
   shader->info.image_buffers = 0;
   shader->info.msaa_images = 0;
   shader->info.bit_sizes_float = 0;
   shader->info.bit_sizes_int = 0;

   nir_foreach_uniform_variable(var, shader) {
      /* Bindless textures and images don't use non-bindless slots.
       * Interface blocks imply inputs, outputs, UBO, or SSBO, which can only
       * mean bindless.
       */
      if (var->data.bindless || var->interface_type)
         continue;

      shader->info.num_textures += glsl_type_get_sampler_count(var->type);

      unsigned num_image_slots = glsl_type_get_image_count(var->type);
      if (num_image_slots) {
         const struct glsl_type *image_type = glsl_without_array(var->type);

         if (glsl_get_sampler_dim(image_type) == GLSL_SAMPLER_DIM_BUF) {
            shader->info.image_buffers |=
               BITFIELD_RANGE(shader->info.num_images, num_image_slots);
         }
         if (glsl_get_sampler_dim(image_type) == GLSL_SAMPLER_DIM_MS) {
            shader->info.msaa_images |=
               BITFIELD_RANGE(shader->info.num_images, num_image_slots);
         }
         shader->info.num_images += num_image_slots;
      }
   }

   shader->info.inputs_read = 0;
   shader->info.outputs_written = 0;
   shader->info.outputs_read = 0;
   shader->info.inputs_read_16bit = 0;
   shader->info.outputs_written_16bit = 0;
   shader->info.outputs_read_16bit = 0;
   shader->info.inputs_read_indirectly_16bit = 0;
   shader->info.outputs_accessed_indirectly_16bit = 0;
   shader->info.patch_outputs_read = 0;
   shader->info.patch_inputs_read = 0;
   shader->info.patch_outputs_written = 0;
   BITSET_ZERO(shader->info.system_values_read);
   shader->info.inputs_read_indirectly = 0;
   shader->info.outputs_accessed_indirectly = 0;
   shader->info.patch_inputs_read_indirectly = 0;
   shader->info.patch_outputs_accessed_indirectly = 0;

   if (shader->info.stage == MESA_SHADER_VERTEX) {
      shader->info.vs.double_inputs = 0;
   }
   if (shader->info.stage == MESA_SHADER_FRAGMENT) {
      shader->info.fs.uses_sample_qualifier = false;
      shader->info.fs.uses_discard = false;
      shader->info.fs.uses_demote = false;
      shader->info.fs.color_is_dual_source = false;
      shader->info.fs.uses_fbfetch_output = false;
      shader->info.fs.needs_quad_helper_invocations = false;
      shader->info.fs.needs_all_helper_invocations = false;
   }
   if (shader->info.stage == MESA_SHADER_TESS_CTRL) {
      shader->info.tess.tcs_cross_invocation_inputs_read = 0;
      shader->info.tess.tcs_cross_invocation_outputs_read = 0;
   }

   shader->info.writes_memory = shader->info.has_transform_feedback_varyings;

   void *dead_ctx = ralloc_context(NULL);
   nir_foreach_block(block, entrypoint) {
      gather_info_block(block, shader, dead_ctx);
   }
   ralloc_free(dead_ctx);

   if (shader->info.stage == MESA_SHADER_FRAGMENT &&
       (shader->info.fs.uses_sample_qualifier ||
        (BITSET_TEST(shader->info.system_values_read, SYSTEM_VALUE_SAMPLE_ID) ||
         BITSET_TEST(shader->info.system_values_read, SYSTEM_VALUE_SAMPLE_POS)))) {
      /* This shouldn't be cleared because if optimizations remove all
       * sample-qualified inputs and that pass is run again, the sample
       * shading must stay enabled.
       */
      shader->info.fs.uses_sample_shading = true;
   }

   shader->info.per_primitive_outputs = 0;
   if (shader->info.stage == MESA_SHADER_MESH) {
      nir_foreach_shader_out_variable(var, shader) {
         if (var->data.per_primitive) {
            assert(nir_is_arrayed_io(var, shader->info.stage));
            const unsigned slots =
               glsl_count_attribute_slots(glsl_get_array_element(var->type), false);
            shader->info.per_primitive_outputs |= BITFIELD64_RANGE(var->data.location, slots);
         }
      }
   }

   shader->info.per_primitive_inputs = 0;
   if (shader->info.stage == MESA_SHADER_FRAGMENT) {
      nir_foreach_shader_in_variable(var, shader) {
         if (var->data.per_primitive) {
            const unsigned slots =
               glsl_count_attribute_slots(var->type, false);
            shader->info.per_primitive_inputs |= BITFIELD64_RANGE(var->data.location, slots);
         }
      }
   }
}