[ARM][3/3] Expand mod by power of 2

* config/arm/arm.md (*subsi3_compare0): Rename to...
	(subsi3_compare0): ... This.
	(modsi3): New define_expand.
	* config/arm/arm.c (arm_new_rtx_costs, MOD case): Handle case
	when operand is power of 2.

	* gcc.target/aarch64/mod_2.x: New file.
	* gcc.target/aarch64/mod_256.x: Likewise.
	* gcc.target/arm/mod_2.c: New test.
	* gcc.target/arm/mod_256.c: Likewise.
	* gcc.target/aarch64/mod_2.c: Likewise.
	* gcc.target/aarch64/mod_256.c: Likewise.

From-SVN: r227586

gcc/ChangeLog

@@ -1,3 +1,11 @@
+2015-09-09  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>
+
+	* config/arm/arm.md (*subsi3_compare0): Rename to...
+	(subsi3_compare0): ... This.
+	(modsi3): New define_expand.
+	* config/arm/arm.c (arm_new_rtx_costs, MOD case): Handle case
+	when operand is power of 2.
+
 2015-09-09  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>
 
 	* config/aarch64/aarch64.md (mod<mode>3): New define_expand.

gcc/config/arm/arm.c

@@ -9580,6 +9580,24 @@ arm_new_rtx_costs (rtx x, enum rtx_code code, enum rtx_code outer_code,
       return false;	/* All arguments must be in registers.  */
 
     case MOD:
+      /* MOD by a power of 2 can be expanded as:
+	 rsbs    r1, r0, #0
+	 and     r0, r0, #(n - 1)
+	 and     r1, r1, #(n - 1)
+	 rsbpl   r0, r1, #0.  */
+      if (CONST_INT_P (XEXP (x, 1))
+	  && exact_log2 (INTVAL (XEXP (x, 1))) > 0
+	  && mode == SImode)
+	{
+	  *cost += COSTS_N_INSNS (3);
+
+	  if (speed_p)
+	    *cost += 2 * extra_cost->alu.logical
+		     + extra_cost->alu.arith;
+	  return true;
+	}
+
+    /* Fall-through.  */
     case UMOD:
       *cost = LIBCALL_COST (2);
       return false;	/* All arguments must be in registers.  */

gcc/config/arm/arm.md

@@ -1229,7 +1229,7 @@
   ""
 )
 
-(define_insn "*subsi3_compare0"
+(define_insn "subsi3_compare0"
   [(set (reg:CC_NOOV CC_REGNUM)
 	(compare:CC_NOOV
 	 (minus:SI (match_operand:SI 1 "arm_rhs_operand" "r,r,I")
@@ -11142,6 +11142,75 @@
   ""
 )
 
+;; ARM-specific expansion of signed mod by power of 2
+;; using conditional negate.
+;; For r0 % n where n is a power of 2 produce:
+;; rsbs    r1, r0, #0
+;; and     r0, r0, #(n - 1)
+;; and     r1, r1, #(n - 1)
+;; rsbpl   r0, r1, #0
+
+(define_expand "modsi3"
+  [(match_operand:SI 0 "register_operand" "")
+   (match_operand:SI 1 "register_operand" "")
+   (match_operand:SI 2 "const_int_operand" "")]
+  "TARGET_32BIT"
+  {
+    HOST_WIDE_INT val = INTVAL (operands[2]);
+
+    if (val <= 0
+       || exact_log2 (val) <= 0)
+      FAIL;
+
+    rtx mask = GEN_INT (val - 1);
+
+    /* In the special case of x0 % 2 we can do the even shorter:
+	cmp     r0, #0
+	and     r0, r0, #1
+	rsblt   r0, r0, #0.  */
+
+    if (val == 2)
+      {
+	rtx cc_reg = arm_gen_compare_reg (LT,
+					  operands[1], const0_rtx, NULL_RTX);
+	rtx cond = gen_rtx_LT (SImode, cc_reg, const0_rtx);
+	rtx masked = gen_reg_rtx (SImode);
+
+	emit_insn (gen_andsi3 (masked, operands[1], mask));
+	emit_move_insn (operands[0],
+			gen_rtx_IF_THEN_ELSE (SImode, cond,
+					      gen_rtx_NEG (SImode,
+							   masked),
+					      masked));
+	DONE;
+      }
+
+    rtx neg_op = gen_reg_rtx (SImode);
+    rtx_insn *insn = emit_insn (gen_subsi3_compare0 (neg_op, const0_rtx,
+						      operands[1]));
+
+    /* Extract the condition register and mode.  */
+    rtx cmp = XVECEXP (PATTERN (insn), 0, 0);
+    rtx cc_reg = SET_DEST (cmp);
+    rtx cond = gen_rtx_GE (SImode, cc_reg, const0_rtx);
+
+    emit_insn (gen_andsi3 (operands[0], operands[1], mask));
+
+    rtx masked_neg = gen_reg_rtx (SImode);
+    emit_insn (gen_andsi3 (masked_neg, neg_op, mask));
+
+    /* We want a conditional negate here, but emitting COND_EXEC rtxes
+       during expand does not always work.  Do an IF_THEN_ELSE instead.  */
+    emit_move_insn (operands[0],
+		    gen_rtx_IF_THEN_ELSE (SImode, cond,
+					  gen_rtx_NEG (SImode, masked_neg),
+					  operands[0]));
+
+
+    DONE;
+  }
+)
+
 (define_expand "bswapsi2"
   [(set (match_operand:SI 0 "s_register_operand" "=r")
   	(bswap:SI (match_operand:SI 1 "s_register_operand" "r")))]

gcc/testsuite/ChangeLog

@@ -1,3 +1,12 @@
+2015-09-09  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>
+
+	* gcc.target/aarch64/mod_2.x: New file.
+	* gcc.target/aarch64/mod_256.x: Likewise.
+	* gcc.target/arm/mod_2.c: New test.
+	* gcc.target/arm/mod_256.c: Likewise.
+	* gcc.target/aarch64/mod_2.c: Likewise.
+	* gcc.target/aarch64/mod_256.c: Likewise.
+
 2015-09-09  Jakub Jelinek  <jakub@redhat.com>
 
 	PR c++/67504

gcc/testsuite/gcc.target/aarch64/mod_2.c

@@ -0,0 +1,7 @@
+/* { dg-do compile } */
+/* { dg-options "-O2 -mcpu=cortex-a57 -save-temps" } */
+
+#include "mod_2.x"
+
+/* { dg-final { scan-assembler "csneg\t\[wx\]\[0-9\]*" } } */
+/* { dg-final { scan-assembler-times "and\t\[wx\]\[0-9\]*" 1 } } */

gcc/testsuite/gcc.target/aarch64/mod_2.x

@@ -0,0 +1,5 @@
+int
+f (int x)
+{
+  return x % 2;
+}

gcc/testsuite/gcc.target/aarch64/mod_256.c

@@ -0,0 +1,6 @@
+/* { dg-do compile } */
+/* { dg-options "-O2 -mcpu=cortex-a57 -save-temps" } */
+
+#include "mod_256.x"
+
+/* { dg-final { scan-assembler "csneg\t\[wx\]\[0-9\]*" } } */

gcc/testsuite/gcc.target/aarch64/mod_256.x

@@ -0,0 +1,5 @@
+int
+f (int x)
+{
+  return x % 256;
+}

gcc/testsuite/gcc.target/arm/mod_2.c

@@ -0,0 +1,8 @@
+/* { dg-do compile } */
+/* { dg-require-effective-target arm32 } */
+/* { dg-options "-O2 -mcpu=cortex-a57 -save-temps" } */
+
+#include "../aarch64/mod_2.x"
+
+/* { dg-final { scan-assembler "rsblt\tr\[0-9\]*" } } */
+/* { dg-final { scan-assembler-times "and\tr\[0-9\].*1" 1 } } */

gcc/testsuite/gcc.target/arm/mod_256.c

@@ -0,0 +1,8 @@
+/* { dg-do compile } */
+/* { dg-require-effective-target arm32 } */
+/* { dg-options "-O2 -mcpu=cortex-a57 -save-temps" } */
+
+#include "../aarch64/mod_256.x"
+
+/* { dg-final { scan-assembler "rsbpl\tr\[0-9\]*" } } */
+