arm: [MVE intrinsics] Add load_extending and store_truncating function bases

This patch adds the load_extending and store_truncating function bases
for MVE intrinsics.

The constructors have parameters describing the memory element
type/width which is part of the function base name (e.g. "h" in
vldrhq).

2024-09-11  Alfie Richards <Alfie.Richards@arm.com>

	gcc/

	* config/arm/arm-mve-builtins-functions.h
	(load_extending): New class.
	(store_truncating): New class.
	* config/arm/arm-protos.h (arm_mve_data_mode): New helper function.
	* config/arm/arm.cc (arm_mve_data_mode): New helper function.

diff --git a/gcc/config/arm/arm-mve-builtins-functions.h b/gcc/config/arm/arm-mve-builtins-functions.h
index 57e59e30c369db5f310fedbda4aa8e166ae0ccaf..48b9e79dbafa17593a433ed3363db8461322fd24 100644 (file)
--- a/gcc/config/arm/arm-mve-builtins-functions.h
+++ b/gcc/config/arm/arm-mve-builtins-functions.h
@@ -20,6 +20,8 @@
 #ifndef GCC_ARM_MVE_BUILTINS_FUNCTIONS_H
 #define GCC_ARM_MVE_BUILTINS_FUNCTIONS_H
 
+#include "arm-protos.h"
+
 namespace arm_mve {
 
 /* Wrap T, which is derived from function_base, and indicate that the
@@ -977,6 +979,107 @@ public:
   }
 };
 
+/* A function_base that loads elements from memory and extends them
+   to a wider element.  The memory element type is a fixed part of
+   the function base name.  */
+class load_extending : public function_base
+{
+public:
+  CONSTEXPR load_extending (type_suffix_index signed_memory_type,
+                           type_suffix_index unsigned_memory_type,
+                           type_suffix_index float_memory_type)
+    : m_signed_memory_type (signed_memory_type),
+      m_unsigned_memory_type (unsigned_memory_type),
+      m_float_memory_type (float_memory_type)
+  {}
+  CONSTEXPR load_extending (type_suffix_index signed_memory_type,
+                           type_suffix_index unsigned_memory_type)
+    : m_signed_memory_type (signed_memory_type),
+      m_unsigned_memory_type (unsigned_memory_type),
+      m_float_memory_type (NUM_TYPE_SUFFIXES)
+  {}
+
+  unsigned int call_properties (const function_instance &) const override
+  {
+    return CP_READ_MEMORY;
+  }
+
+  tree memory_scalar_type (const function_instance &fi) const override
+  {
+    type_suffix_index memory_type_suffix
+      = (fi.type_suffix (0).integer_p
+        ? (fi.type_suffix (0).unsigned_p
+           ? m_unsigned_memory_type
+           : m_signed_memory_type)
+        : m_float_memory_type);
+    return scalar_types[type_suffixes[memory_type_suffix].vector_type];
+  }
+
+  machine_mode memory_vector_mode (const function_instance &fi) const override
+  {
+    type_suffix_index memory_type_suffix
+      = (fi.type_suffix (0).integer_p
+        ? (fi.type_suffix (0).unsigned_p
+           ? m_unsigned_memory_type
+           : m_signed_memory_type)
+        : m_float_memory_type);
+    machine_mode mem_mode = type_suffixes[memory_type_suffix].vector_mode;
+    machine_mode reg_mode = fi.vector_mode (0);
+
+    return arm_mve_data_mode (GET_MODE_INNER (mem_mode),
+                             GET_MODE_NUNITS (reg_mode)).require ();
+  }
+
+  /* The type of the memory elements.  This is part of the function base
+     name rather than a true type suffix.  */
+  type_suffix_index m_signed_memory_type;
+  type_suffix_index m_unsigned_memory_type;
+  type_suffix_index m_float_memory_type;
+};
+
+/* A function_base that truncates vector elements and stores them to memory.
+   The memory element width is a fixed part of the function base name.  */
+class store_truncating : public function_base
+{
+public:
+  CONSTEXPR store_truncating (scalar_mode to_int_mode,
+                             opt_scalar_mode to_float_mode)
+    : m_to_int_mode (to_int_mode), m_to_float_mode (to_float_mode)
+  {}
+
+  unsigned int call_properties (const function_instance &) const override
+  {
+    return CP_WRITE_MEMORY;
+  }
+
+  tree memory_scalar_type (const function_instance &fi) const override
+  {
+    /* In truncating stores, the signedness of the memory element is defined
+       to be the same as the signedness of the vector element.  The signedness
+       doesn't make any difference to the behavior of the function.  */
+    type_class_index tclass = fi.type_suffix (0).tclass;
+    unsigned int element_bits
+      = GET_MODE_BITSIZE (fi.type_suffix (0).integer_p
+                         ? m_to_int_mode
+                         : m_to_float_mode.require ());
+    type_suffix_index suffix = find_type_suffix (tclass, element_bits);
+    return scalar_types[type_suffixes[suffix].vector_type];
+  }
+
+  machine_mode memory_vector_mode (const function_instance &fi) const override
+  {
+    poly_uint64 nunits = GET_MODE_NUNITS (fi.vector_mode (0));
+    scalar_mode mode = (fi.type_suffix (0).integer_p
+                       ? m_to_int_mode
+                       : m_to_float_mode.require ());
+    return arm_mve_data_mode (mode, nunits).require ();
+  }
+
+  /* The mode of a single memory element.  */
+  scalar_mode m_to_int_mode;
+  opt_scalar_mode m_to_float_mode;
+};
+
 } /* end namespace arm_mve */
 
 /* Declare the global function base NAME, creating it from an instance

diff --git a/gcc/config/arm/arm-protos.h b/gcc/config/arm/arm-protos.h
index b694589cab4b45730b257baacfdbd9db00584cdc..7311ad4d8e4a4fae88e3d194a553252fbef3fae9 100644 (file)
--- a/gcc/config/arm/arm-protos.h
+++ b/gcc/config/arm/arm-protos.h
@@ -616,4 +616,7 @@ void arm_initialize_isa (sbitmap, const enum isa_feature *);
 const char * arm_gen_far_branch (rtx *, int, const char * , const char *);
 
 bool arm_mve_immediate_check(rtx, machine_mode, bool);
+
+opt_machine_mode arm_mve_data_mode (scalar_mode, poly_uint64);
+
 #endif /* ! GCC_ARM_PROTOS_H */

diff --git a/gcc/config/arm/arm.cc b/gcc/config/arm/arm.cc
index 8ccdf1ae93cb14110640be385d67ca7a2459d513..6f11b6c816d4c01a798e85202f2f256260ff416b 100644 (file)
--- a/gcc/config/arm/arm.cc
+++ b/gcc/config/arm/arm.cc
@@ -76,6 +76,7 @@
 #include "opts.h"
 #include "aarch-common.h"
 #include "aarch-common-protos.h"
+#include "machmode.h"
 
 /* This file should be included last.  */
 #include "target-def.h"
@@ -36212,4 +36213,18 @@ arm_output_load_tpidr (rtx dst, bool pred_p)
   return "";
 }
 
+/* Return the MVE vector mode that has NUNITS elements of mode INNER_MODE.  */
+opt_machine_mode
+arm_mve_data_mode (scalar_mode inner_mode, poly_uint64 nunits)
+{
+  enum mode_class mclass
+    = (SCALAR_FLOAT_MODE_P (inner_mode) ? MODE_VECTOR_FLOAT : MODE_VECTOR_INT);
+  machine_mode mode;
+  FOR_EACH_MODE_IN_CLASS (mode, mclass)
+    if (inner_mode == GET_MODE_INNER (mode)
+       && known_eq (nunits, GET_MODE_NUNITS (mode)))
+      return mode;
+  return opt_machine_mode ();
+}
+
 #include "gt-arm.h"