diff options
-rw-r--r-- | src/amd/vulkan/radv_shader.c | 2 | ||||
-rw-r--r-- | src/compiler/nir/nir.h | 2 | ||||
-rw-r--r-- | src/compiler/nir/nir_deref.c | 63 | ||||
-rw-r--r-- | src/compiler/nir/nir_inline_functions.c | 12 | ||||
-rw-r--r-- | src/compiler/nir/nir_opt_copy_propagate.c | 22 | ||||
-rw-r--r-- | src/intel/vulkan/anv_pipeline.c | 2 | ||||
-rw-r--r-- | src/mesa/main/glspirv.c | 2 |
7 files changed, 75 insertions, 30 deletions
diff --git a/src/amd/vulkan/radv_shader.c b/src/amd/vulkan/radv_shader.c index 7ad9abe8df8..34bfa447930 100644 --- a/src/amd/vulkan/radv_shader.c +++ b/src/amd/vulkan/radv_shader.c @@ -265,7 +265,7 @@ radv_shader_compile_to_nir(struct radv_device *device, NIR_PASS_V(nir, nir_lower_constant_initializers, nir_var_local); NIR_PASS_V(nir, nir_lower_returns); NIR_PASS_V(nir, nir_inline_functions); - NIR_PASS_V(nir, nir_copy_prop); + NIR_PASS_V(nir, nir_opt_deref); /* Pick off the single entrypoint that we want */ foreach_list_typed_safe(nir_function, func, node, &nir->functions) { diff --git a/src/compiler/nir/nir.h b/src/compiler/nir/nir.h index e72585000d4..e991c625536 100644 --- a/src/compiler/nir/nir.h +++ b/src/compiler/nir/nir.h @@ -3196,6 +3196,8 @@ bool nir_opt_dead_cf(nir_shader *shader); bool nir_opt_dead_write_vars(nir_shader *shader); +bool nir_opt_deref(nir_shader *shader); + bool nir_opt_find_array_copies(nir_shader *shader); bool nir_opt_gcm(nir_shader *shader, bool value_number); diff --git a/src/compiler/nir/nir_deref.c b/src/compiler/nir/nir_deref.c index 7a625bcf92b..1dffa285037 100644 --- a/src/compiler/nir/nir_deref.c +++ b/src/compiler/nir/nir_deref.c @@ -510,3 +510,66 @@ nir_rematerialize_derefs_in_use_blocks_impl(nir_function_impl *impl) return state.progress; } + +static bool +is_trivial_deref_cast(nir_deref_instr *cast) +{ + nir_deref_instr *parent = nir_src_as_deref(cast->parent); + if (!parent) + return false; + + return cast->mode == parent->mode && + cast->type == parent->type && + cast->dest.ssa.num_components == parent->dest.ssa.num_components && + cast->dest.ssa.bit_size == parent->dest.ssa.bit_size; +} + +static bool +nir_opt_deref_impl(nir_function_impl *impl) +{ + bool progress = false; + + nir_foreach_block(block, impl) { + nir_foreach_instr_safe(instr, block) { + if (instr->type != nir_instr_type_deref) + continue; + + nir_deref_instr *deref = nir_instr_as_deref(instr); + switch (deref->deref_type) { + case nir_deref_type_cast: + if (is_trivial_deref_cast(deref)) { + assert(deref->parent.is_ssa); + nir_ssa_def_rewrite_uses(&deref->dest.ssa, + nir_src_for_ssa(deref->parent.ssa)); + nir_instr_remove(&deref->instr); + progress = true; + } + break; + + default: + /* Do nothing */ + break; + } + } + } + + if (progress) { + nir_metadata_preserve(impl, nir_metadata_block_index | + nir_metadata_dominance); + } + + return progress; +} + +bool +nir_opt_deref(nir_shader *shader) +{ + bool progress = false; + + nir_foreach_function(func, shader) { + if (func->impl && nir_opt_deref_impl(func->impl)) + progress = true; + } + + return progress; +} diff --git a/src/compiler/nir/nir_inline_functions.c b/src/compiler/nir/nir_inline_functions.c index 29474bb417b..864638d2315 100644 --- a/src/compiler/nir/nir_inline_functions.c +++ b/src/compiler/nir/nir_inline_functions.c @@ -164,14 +164,16 @@ inline_function_impl(nir_function_impl *impl, struct set *inlined) * This does the actual function inlining and the resulting shader will * contain no call instructions. * - * 4. nir_copy_prop(shader) + * 4. nir_opt_deref(shader) * * Most functions contain pointer parameters where the result of a deref * instruction is passed in as a parameter, loaded via a load_param - * intrinsic, and then turned back into a deref via a cast. Running copy - * propagation gets rid of the intermediate steps and results in a whole - * deref chain again. This is currently required by a number of - * optimizations and lowering passes at least for certain variable modes. + * intrinsic, and then turned back into a deref via a cast. Function + * inlining will get rid of the load_param but we are still left with a + * cast. Running nir_opt_deref gets rid of the intermediate cast and + * results in a whole deref chain again. This is currently required by a + * number of optimizations and lowering passes at least for certain + * variable modes. * * 5. Loop over the functions and delete all but the main entrypoint. * diff --git a/src/compiler/nir/nir_opt_copy_propagate.c b/src/compiler/nir/nir_opt_copy_propagate.c index 189d544979b..7673e9b62dd 100644 --- a/src/compiler/nir/nir_opt_copy_propagate.c +++ b/src/compiler/nir/nir_opt_copy_propagate.c @@ -99,22 +99,6 @@ is_swizzleless_move(nir_alu_instr *instr) } static bool -is_trivial_deref_cast(nir_deref_instr *cast) -{ - if (cast->deref_type != nir_deref_type_cast) - return false; - - nir_deref_instr *parent = nir_src_as_deref(cast->parent); - if (!parent) - return false; - - return cast->mode == parent->mode && - cast->type == parent->type && - cast->dest.ssa.num_components == parent->dest.ssa.num_components && - cast->dest.ssa.bit_size == parent->dest.ssa.bit_size; -} - -static bool copy_prop_src(nir_src *src, nir_instr *parent_instr, nir_if *parent_if, unsigned num_components) { @@ -135,12 +119,6 @@ copy_prop_src(nir_src *src, nir_instr *parent_instr, nir_if *parent_if, return false; copy_def= alu_instr->src[0].src.ssa; - } else if (src_instr->type == nir_instr_type_deref) { - nir_deref_instr *deref_instr = nir_instr_as_deref(src_instr); - if (!is_trivial_deref_cast(deref_instr)) - return false; - - copy_def = deref_instr->parent.ssa; } else { return false; } diff --git a/src/intel/vulkan/anv_pipeline.c b/src/intel/vulkan/anv_pipeline.c index 6db9945e0d4..10afe0825ab 100644 --- a/src/intel/vulkan/anv_pipeline.c +++ b/src/intel/vulkan/anv_pipeline.c @@ -184,7 +184,7 @@ anv_shader_compile_to_nir(struct anv_pipeline *pipeline, NIR_PASS_V(nir, nir_lower_constant_initializers, nir_var_local); NIR_PASS_V(nir, nir_lower_returns); NIR_PASS_V(nir, nir_inline_functions); - NIR_PASS_V(nir, nir_copy_prop); + NIR_PASS_V(nir, nir_opt_deref); /* Pick off the single entrypoint that we want */ foreach_list_typed_safe(nir_function, func, node, &nir->functions) { diff --git a/src/mesa/main/glspirv.c b/src/mesa/main/glspirv.c index 04e46ba571e..ec1edf11cfa 100644 --- a/src/mesa/main/glspirv.c +++ b/src/mesa/main/glspirv.c @@ -245,7 +245,7 @@ _mesa_spirv_to_nir(struct gl_context *ctx, NIR_PASS_V(nir, nir_lower_constant_initializers, nir_var_local); NIR_PASS_V(nir, nir_lower_returns); NIR_PASS_V(nir, nir_inline_functions); - NIR_PASS_V(nir, nir_copy_prop); + NIR_PASS_V(nir, nir_opt_deref); /* Pick off the single entrypoint that we want */ foreach_list_typed_safe(nir_function, func, node, &nir->functions) { |