[PATCH 2/2 v2] RISC-V: Constant synthesis of inverted halves

Changes since v1:
	- Fix synthesis-15.c.

-- >8 --

Improve handling of constants where the high half can be constructed by
inverting the lower half.

gcc/ChangeLog:
	* config/riscv/riscv.cc (riscv_build_integer): Detect constants
	were the higher half is the lower half inverted.

gcc/testsuite/ChangeLog:
	* gcc.target/riscv/synthesis-15.c: New test.

2 files changed, 56 insertions, 0 deletions

diff --git a/gcc/config/riscv/riscv.cc b/gcc/config/riscv/riscv.cc
index 064ffa4..6efe14f 100644
--- a/gcc/config/riscv/riscv.cc
+++ b/gcc/config/riscv/riscv.cc
@@ -1346,6 +1346,36 @@ riscv_build_integer (struct riscv_integer_op *codes, HOST_WIDE_INT value,
 		}
 	    }
 	}
+
+      if (cost > 5 && !bit31)
+	{
+	  /* For constants where the upper half is the lower half inverted we can flip
+	     it with an xor and do a shift 32 followed by an or.  */
+	  if (hival == (~loval & 0xffffffff))
+	    {
+	      alt_cost = 3 + riscv_build_integer_1 (alt_codes,
+						    sext_hwi (loval, 32), mode);
+	      if (alt_cost < cost)
+		{
+		  alt_codes[alt_cost - 4].save_temporary = true;
+		  alt_codes[alt_cost - 3].code = XOR;
+		  alt_codes[alt_cost - 3].value = -1;
+		  alt_codes[alt_cost - 3].use_uw = false;
+		  alt_codes[alt_cost - 3].save_temporary = false;
+		  alt_codes[alt_cost - 2].code = ASHIFT;
+		  alt_codes[alt_cost - 2].value = 32;
+		  alt_codes[alt_cost - 2].use_uw = false;
+		  alt_codes[alt_cost - 2].save_temporary = false;
+		  alt_codes[alt_cost - 1].code = CONCAT;
+		  alt_codes[alt_cost - 1].value = 0;
+		  alt_codes[alt_cost - 1].use_uw = false;
+		  alt_codes[alt_cost - 1].save_temporary = false;
+
+		  memcpy (codes, alt_codes, sizeof (alt_codes));
+		  cost = alt_cost;
+		}
+	    }
+	}
     }
 
   return cost;
diff --git a/gcc/testsuite/gcc.target/riscv/synthesis-15.c b/gcc/testsuite/gcc.target/riscv/synthesis-15.c
new file mode 100644
index 0000000..dc1f8b7
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/synthesis-15.c
@@ -0,0 +1,26 @@
+/* { dg-do compile } */
+/* { dg-require-effective-target rv64 } */
+/* We aggressively skip as we really just need to test the basic synthesis
+   which shouldn't vary based on the optimization level.  -O1 seems to work
+   and eliminates the usual sources of extraneous dead code that would throw
+   off the counts.  */
+/* { dg-skip-if "" { *-*-* } { "-O0" "-Og" "-O2" "-O3" "-Os" "-Oz" "-flto" } } */
+/* { dg-options "-march=rv64gc" } */
+
+/* Rather than test for a specific synthesis of all these constants or
+   having thousands of tests each testing one variant, we just test the
+   total number of instructions.
+
+   This isn't expected to change much and any change is worthy of a look.  */
+/* { dg-final { scan-assembler-times "\\t(add|addi|bseti|li|pack|ret|sh1add|sh2add|sh3add|slli|srli|xori|or)" 60 } } */
+
+unsigned long foo_0xcafe605f35019fa0(void) { return 0xcafe605f35019fa0UL; }
+unsigned long foo_0x87a80d217857f2de(void) { return 0x87a80d217857f2deUL; }
+unsigned long foo_0x6699f19c19660e63(void) { return 0xe699f19c19660e63UL; }
+unsigned long foo_0xec80e48a137f1b75(void) { return 0xec80e48a137f1b75UL; }
+unsigned long foo_0xc7d7193e3828e6c1(void) { return 0xc7d7193e3828e6c1UL; }
+unsigned long foo_0xfc627816039d87e9(void) { return 0xfc627816039d87e9UL; }
+unsigned long foo_0xbd69e83e429617c1(void) { return 0xbd69e83e429617c1UL; }
+unsigned long foo_0xdbee7ee624118119(void) { return 0xdbee7ee624118119UL; }
+unsigned long foo_0xf3fe20820c01df7d(void) { return 0xf3fe20820c01df7dUL; }
+unsigned long foo_0x8f1dc29470e23d6b(void) { return 0x8f1dc29470e23d6bUL; }