i386: Support partial vectorized V2BF/V4BF plus/minus/mult/div/sqrt

This patch introduces new mode iterators and expands for the i386 architecture to support partial vectorization of bf16 operations using AVX10.2 instructions.

gcc/ChangeLog:

	* config/i386/mmx.md (VBF_32_64): New mode iterator for partial vectorized V2BF/V4BF.
	(<insn><mode>3): New define_expand for plusminusmultdiv.
	(sqrt<mode>2): New define_expand for sqrt.

gcc/testsuite/ChangeLog:

	* gcc.target/i386/avx10_2-partial-bf-vector-fast-math-1.c: New test.
	* gcc.target/i386/avx10_2-partial-bf-vector-operations-1.c: New test.

3 files changed, 116 insertions, 0 deletions

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index cb26975..076ea2e 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1958,6 +1958,8 @@
 
 (define_mode_iterator VHF_32_64 [V2HF (V4HF "TARGET_MMX_WITH_SSE")])
 
+(define_mode_iterator VBF_32_64 [V2BF (V4BF "TARGET_MMX_WITH_SSE")])
+
 (define_expand "divv4hf3"
   [(set (match_operand:V4HF 0 "register_operand")
 	(div:V4HF
@@ -2036,6 +2038,26 @@
   DONE;
 })
 
+;; VDIVNEPBF16 does not generate floating point exceptions.
+(define_expand "<insn><mode>3"
+  [(set (match_operand:VBF_32_64 0 "register_operand")
+    (plusminusmultdiv:VBF_32_64
+      (match_operand:VBF_32_64 1 "nonimmediate_operand")
+      (match_operand:VBF_32_64 2 "nonimmediate_operand")))]
+  "TARGET_AVX10_2_256"
+{
+  rtx op0 = gen_reg_rtx (V8BFmode);
+  rtx op1 = lowpart_subreg (V8BFmode,
+			    force_reg (<MODE>mode, operands[1]), <MODE>mode);
+  rtx op2 = lowpart_subreg (V8BFmode,
+			    force_reg (<MODE>mode, operands[2]), <MODE>mode);
+
+  emit_insn (gen_<insn>v8bf3 (op0, op1, op2));
+
+  emit_move_insn (operands[0], lowpart_subreg (<MODE>mode, op0, V8BFmode));
+  DONE;
+})
+
 (define_expand "divv2hf3"
   [(set (match_operand:V2HF 0 "register_operand")
 	(div:V2HF
@@ -2091,6 +2113,21 @@
   DONE;
 })
 
+(define_expand "sqrt<mode>2"
+  [(set (match_operand:VBF_32_64 0 "register_operand")
+	(sqrt:VBF_32_64 (match_operand:VBF_32_64 1 "vector_operand")))]
+  "TARGET_AVX10_2_256"
+{
+  rtx op0 = gen_reg_rtx (V8BFmode);
+  rtx op1 = lowpart_subreg (V8BFmode,
+			    force_reg (<MODE>mode, operands[1]), <MODE>mode);
+
+  emit_insn (gen_sqrtv8bf2 (op0, op1));
+
+  emit_move_insn (operands[0], lowpart_subreg (<MODE>mode, op0, V8BFmode));
+  DONE;
+})
+
 (define_expand "<code><mode>2"
   [(set (match_operand:VHF_32_64 0 "register_operand")
 	(absneg:VHF_32_64
diff --git a/gcc/testsuite/gcc.target/i386/avx10_2-partial-bf-vector-fast-math-1.c b/gcc/testsuite/gcc.target/i386/avx10_2-partial-bf-vector-fast-math-1.c
new file mode 100644
index 0000000..fd064f1
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avx10_2-partial-bf-vector-fast-math-1.c
@@ -0,0 +1,22 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-mavx10.2 -O2" } */
+/* { dg-final { scan-assembler-times "vmulnepbf16\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 2 } } */
+/* { dg-final { scan-assembler-times "vrcppbf16\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 2 } } */
+
+typedef __bf16 v4bf __attribute__ ((__vector_size__ (8)));
+typedef __bf16 v2bf __attribute__ ((__vector_size__ (4)));
+
+
+__attribute__((optimize("fast-math")))
+v4bf
+foo_div_fast_math_4 (v4bf a, v4bf b)
+{
+  return a / b;
+}
+
+__attribute__((optimize("fast-math")))
+v2bf
+foo_div_fast_math_2 (v2bf a, v2bf b)
+{
+  return a / b;
+}
diff --git a/gcc/testsuite/gcc.target/i386/avx10_2-partial-bf-vector-operations-1.c b/gcc/testsuite/gcc.target/i386/avx10_2-partial-bf-vector-operations-1.c
new file mode 100644
index 0000000..e7ee08a
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avx10_2-partial-bf-vector-operations-1.c
@@ -0,0 +1,57 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-mavx10.2 -O2" } */
+/* { dg-final { scan-assembler-times "vmulnepbf16\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 2 } } */
+/* { dg-final { scan-assembler-times "vaddnepbf16\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 2 } } */
+/* { dg-final { scan-assembler-times "vdivnepbf16\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 2 } } */
+/* { dg-final { scan-assembler-times "vsubnepbf16\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 2 } } */
+
+typedef __bf16 v4bf __attribute__ ((__vector_size__ (8)));
+typedef __bf16 v2bf __attribute__ ((__vector_size__ (4)));
+
+v4bf
+foo_mul_4 (v4bf a, v4bf b)
+{
+  return a * b;
+}
+
+v4bf
+foo_add_4 (v4bf a, v4bf b)
+{
+  return a + b;
+}
+
+v4bf
+foo_div_4 (v4bf a, v4bf b)
+{
+  return a / b;
+}
+
+v4bf
+foo_sub_4 (v4bf a, v4bf b)
+{
+  return a - b;
+}
+
+v2bf
+foo_mul_2 (v2bf a, v2bf b)
+{
+  return a * b;
+}
+
+v2bf
+foo_add_2 (v2bf a, v2bf b)
+{
+  return a + b;
+}
+
+v2bf
+foo_div_2 (v2bf a, v2bf b)
+{
+  return a / b;
+}
+
+v2bf
+foo_sub_2 (v2bf a, v2bf b)
+{
+  return a - b;
+}