[AArch64] Add SVE conditional floating-point unary patterns

This patch adds patterns to match conditional unary operations
on floating-point modes.  At the moment we rely on combine to merge
separate arithmetic and vcond_mask operations, and since the latter
doesn't accept zero operands, we miss out on the opportunity to use
the movprfx /z alternative.  (This alternative is tested by the ACLE
patches though.)

2019-08-14  Richard Sandiford  <richard.sandiford@arm.com>
	    Kugan Vivekanandarajah  <kugan.vivekanandarajah@linaro.org>

gcc/
	* config/aarch64/aarch64-sve.md
	(*cond_<SVE_COND_FP_UNARY:optab><SVE_F:mode>_2): New pattern.
	(*cond_<SVE_COND_FP_UNARY:optab><SVE_F:mode>_any): Likewise.

gcc/testsuite/
	* gcc.target/aarch64/sve/cond_unary_1.c: Add tests for
	floating-point types.
	* gcc.target/aarch64/sve/cond_unary_2.c: Likewise.
	* gcc.target/aarch64/sve/cond_unary_3.c: Likewise.
	* gcc.target/aarch64/sve/cond_unary_4.c: Likewise.

Co-Authored-By: Kugan Vivekanandarajah <kuganv@linaro.org>

From-SVN: r274477

7 files changed, 158 insertions, 23 deletions

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index 4184b0f..bab95eb 100644
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -2,6 +2,13 @@
 	    Kugan Vivekanandarajah  <kugan.vivekanandarajah@linaro.org>
 
 	* config/aarch64/aarch64-sve.md
+	(*cond_<SVE_COND_FP_UNARY:optab><SVE_F:mode>_2): New pattern.
+	(*cond_<SVE_COND_FP_UNARY:optab><SVE_F:mode>_any): Likewise.
+
+2019-08-14  Richard Sandiford  <richard.sandiford@arm.com>
+	    Kugan Vivekanandarajah  <kugan.vivekanandarajah@linaro.org>
+
+	* config/aarch64/aarch64-sve.md
 	(*cond_<SVE_INT_UNARY:optab><SVE_I:mode>_2): New pattern.
 	(*cond_<SVE_INT_UNARY:optab><SVE_I:mode>_any): Likewise.
 
diff --git a/gcc/config/aarch64/aarch64-sve.md b/gcc/config/aarch64/aarch64-sve.md
index bed8da9..da49899 100644
--- a/gcc/config/aarch64/aarch64-sve.md
+++ b/gcc/config/aarch64/aarch64-sve.md
@@ -1624,6 +1624,62 @@
   "<sve_fp_op>\t%0.<Vetype>, %1/m, %2.<Vetype>"
 )
 
+;; Predicated floating-point unary arithmetic, merging with the first input.
+(define_insn_and_rewrite "*cond_<optab><mode>_2"
+  [(set (match_operand:SVE_F 0 "register_operand" "=w, ?&w")
+	(unspec:SVE_F
+	  [(match_operand:<VPRED> 1 "register_operand" "Upl, Upl")
+	   (unspec:SVE_F
+	     [(match_operand 3)
+	      (match_operand:SI 4 "aarch64_sve_gp_strictness")
+	      (match_operand:SVE_F 2 "register_operand" "0, w")]
+	     SVE_COND_FP_UNARY)
+	   (match_dup 2)]
+	  UNSPEC_SEL))]
+  "TARGET_SVE && aarch64_sve_pred_dominates_p (&operands[3], operands[1])"
+  "@
+   <sve_fp_op>\t%0.<Vetype>, %1/m, %0.<Vetype>
+   movprfx\t%0, %2\;<sve_fp_op>\t%0.<Vetype>, %1/m, %2.<Vetype>"
+  "&& !rtx_equal_p (operands[1], operands[3])"
+  {
+    operands[3] = copy_rtx (operands[1]);
+  }
+  [(set_attr "movprfx" "*,yes")]
+)
+
+;; Predicated floating-point unary arithmetic, merging with an independent
+;; value.
+;;
+;; The earlyclobber isn't needed for the first alternative, but omitting
+;; it would only help the case in which operands 2 and 3 are the same,
+;; which is handled above rather than here.  Marking all the alternatives
+;; as earlyclobber helps to make the instruction more regular to the
+;; register allocator.
+(define_insn_and_rewrite "*cond_<optab><mode>_any"
+  [(set (match_operand:SVE_F 0 "register_operand" "=&w, ?&w, ?&w")
+	(unspec:SVE_F
+	  [(match_operand:<VPRED> 1 "register_operand" "Upl, Upl, Upl")
+	   (unspec:SVE_F
+	     [(match_operand 4)
+	      (match_operand:SI 5 "aarch64_sve_gp_strictness")
+	      (match_operand:SVE_F 2 "register_operand" "w, w, w")]
+	     SVE_COND_FP_UNARY)
+	   (match_operand:SVE_F 3 "aarch64_simd_reg_or_zero" "0, Dz, w")]
+	  UNSPEC_SEL))]
+  "TARGET_SVE
+   && !rtx_equal_p (operands[2], operands[3])
+   && aarch64_sve_pred_dominates_p (&operands[4], operands[1])"
+  "@
+   <sve_fp_op>\t%0.<Vetype>, %1/m, %2.<Vetype>
+   movprfx\t%0.<Vetype>, %1/z, %2.<Vetype>\;<sve_fp_op>\t%0.<Vetype>, %1/m, %2.<Vetype>
+   movprfx\t%0, %3\;<sve_fp_op>\t%0.<Vetype>, %1/m, %2.<Vetype>"
+  "&& !rtx_equal_p (operands[1], operands[4])"
+  {
+    operands[4] = copy_rtx (operands[1]);
+  }
+  [(set_attr "movprfx" "*,yes,yes")]
+)
+
 ;; -------------------------------------------------------------------------
 ;; ---- [PRED] Inverse
 ;; -------------------------------------------------------------------------
diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog
index cf0f7a0..8d23fff 100644
--- a/gcc/testsuite/ChangeLog
+++ b/gcc/testsuite/ChangeLog
@@ -1,6 +1,15 @@
 2019-08-14  Richard Sandiford  <richard.sandiford@arm.com>
 	    Kugan Vivekanandarajah  <kugan.vivekanandarajah@linaro.org>
 
+	* gcc.target/aarch64/sve/cond_unary_1.c: Add tests for
+	floating-point types.
+	* gcc.target/aarch64/sve/cond_unary_2.c: Likewise.
+	* gcc.target/aarch64/sve/cond_unary_3.c: Likewise.
+	* gcc.target/aarch64/sve/cond_unary_4.c: Likewise.
+
+2019-08-14  Richard Sandiford  <richard.sandiford@arm.com>
+	    Kugan Vivekanandarajah  <kugan.vivekanandarajah@linaro.org>
+
 	* gcc.target/aarch64/sve/cond_unary_1.c: New test.
 	* gcc.target/aarch64/sve/cond_unary_1_run.c: Likewise.
 	* gcc.target/aarch64/sve/cond_unary_2.c: Likewise.
diff --git a/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_1.c b/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_1.c
index 2705a76..2b5f9c3 100644
--- a/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_1.c
+++ b/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_1.c
@@ -15,15 +15,22 @@
       r[i] = pred[i] ? OP (a[i]) : a[i];			\
   }
 
-#define TEST_TYPE(T, TYPE) \
+#define TEST_INT_TYPE(T, TYPE) \
   T (TYPE, abs) \
   T (TYPE, neg)
 
+#define TEST_FLOAT_TYPE(T, TYPE, SUFFIX) \
+  T (TYPE, __builtin_fabs##SUFFIX) \
+  T (TYPE, neg)
+
 #define TEST_ALL(T) \
-  TEST_TYPE (T, int8_t) \
-  TEST_TYPE (T, int16_t) \
-  TEST_TYPE (T, int32_t) \
-  TEST_TYPE (T, int64_t)
+  TEST_INT_TYPE (T, int8_t) \
+  TEST_INT_TYPE (T, int16_t) \
+  TEST_INT_TYPE (T, int32_t) \
+  TEST_INT_TYPE (T, int64_t) \
+  TEST_FLOAT_TYPE (T, _Float16, f16) \
+  TEST_FLOAT_TYPE (T, float, f) \
+  TEST_FLOAT_TYPE (T, double, )
 
 TEST_ALL (DEF_LOOP)
 
@@ -37,6 +44,14 @@ TEST_ALL (DEF_LOOP)
 /* { dg-final { scan-assembler-times {\tneg\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
 /* { dg-final { scan-assembler-times {\tneg\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
 
+/* { dg-final { scan-assembler-times {\tfabs\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfabs\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfabs\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-times {\tfneg\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfneg\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfneg\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
 /* { dg-final { scan-assembler-not {\tmov\tz} } } */
 /* { dg-final { scan-assembler-not {\tmovprfx\t} } } */
 /* XFAILed because the ?: gets canonicalized so that the operation is in
diff --git a/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_2.c b/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_2.c
index 3d346fb..adf8283 100644
--- a/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_2.c
+++ b/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_2.c
@@ -16,15 +16,22 @@
       r[i] = pred[i] ? OP (a[i]) : b[i];			\
   }
 
-#define TEST_TYPE(T, TYPE) \
+#define TEST_INT_TYPE(T, TYPE) \
   T (TYPE, abs) \
   T (TYPE, neg)
 
+#define TEST_FLOAT_TYPE(T, TYPE, SUFFIX) \
+  T (TYPE, __builtin_fabs##SUFFIX) \
+  T (TYPE, neg)
+
 #define TEST_ALL(T) \
-  TEST_TYPE (T, int8_t) \
-  TEST_TYPE (T, int16_t) \
-  TEST_TYPE (T, int32_t) \
-  TEST_TYPE (T, int64_t)
+  TEST_INT_TYPE (T, int8_t) \
+  TEST_INT_TYPE (T, int16_t) \
+  TEST_INT_TYPE (T, int32_t) \
+  TEST_INT_TYPE (T, int64_t) \
+  TEST_FLOAT_TYPE (T, _Float16, f16) \
+  TEST_FLOAT_TYPE (T, float, f) \
+  TEST_FLOAT_TYPE (T, double, )
 
 TEST_ALL (DEF_LOOP)
 
@@ -38,6 +45,17 @@ TEST_ALL (DEF_LOOP)
 /* { dg-final { scan-assembler-times {\tneg\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
 /* { dg-final { scan-assembler-times {\tneg\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
 
+/* { dg-final { scan-assembler-times {\tfabs\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfabs\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfabs\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-times {\tfneg\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfneg\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfneg\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
 /* { dg-final { scan-assembler-not {\tmov\tz} } } */
-/* { dg-final { scan-assembler-not {\tmovprfx\t} } } */
+/* At the moment we don't manage to avoid using MOVPRFX for the
+   floating-point functions.  */
+/* { dg-final { scan-assembler-not {\tmovprfx\t} { xfail *-*-* } } } */
+/* { dg-final { scan-assembler-times {\tmovprfx\t} 6 } } */
 /* { dg-final { scan-assembler-not {\tsel\t} } } */
diff --git a/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_3.c b/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_3.c
index ab2e99a..dde0fdd 100644
--- a/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_3.c
+++ b/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_3.c
@@ -15,15 +15,22 @@
       r[i] = pred[i] ? OP (a[i]) : 5;				\
   }
 
-#define TEST_TYPE(T, TYPE) \
+#define TEST_INT_TYPE(T, TYPE) \
   T (TYPE, abs) \
   T (TYPE, neg)
 
+#define TEST_FLOAT_TYPE(T, TYPE, SUFFIX) \
+  T (TYPE, __builtin_fabs##SUFFIX) \
+  T (TYPE, neg)
+
 #define TEST_ALL(T) \
-  TEST_TYPE (T, int8_t) \
-  TEST_TYPE (T, int16_t) \
-  TEST_TYPE (T, int32_t) \
-  TEST_TYPE (T, int64_t)
+  TEST_INT_TYPE (T, int8_t) \
+  TEST_INT_TYPE (T, int16_t) \
+  TEST_INT_TYPE (T, int32_t) \
+  TEST_INT_TYPE (T, int64_t) \
+  TEST_FLOAT_TYPE (T, _Float16, f16) \
+  TEST_FLOAT_TYPE (T, float, f) \
+  TEST_FLOAT_TYPE (T, double, )
 
 TEST_ALL (DEF_LOOP)
 
@@ -37,7 +44,15 @@ TEST_ALL (DEF_LOOP)
 /* { dg-final { scan-assembler-times {\tneg\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
 /* { dg-final { scan-assembler-times {\tneg\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
 
-/* { dg-final { scan-assembler-times {\tmovprfx\tz[0-9]+, z[0-9]+\n} 8 } } */
+/* { dg-final { scan-assembler-times {\tfabs\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfabs\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfabs\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-times {\tfneg\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfneg\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfneg\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-times {\tmovprfx\tz[0-9]+, z[0-9]+\n} 14 } } */
 
 /* { dg-final { scan-assembler-not {\tmov\tz[^\n]*z} } } */
 /* { dg-final { scan-assembler-not {\tsel\t} } } */
diff --git a/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_4.c b/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_4.c
index 5ad0d80..4604365 100644
--- a/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_4.c
+++ b/gcc/testsuite/gcc.target/aarch64/sve/cond_unary_4.c
@@ -15,15 +15,22 @@
       r[i] = pred[i] ? OP (a[i]) : 0;				\
   }
 
-#define TEST_TYPE(T, TYPE) \
+#define TEST_INT_TYPE(T, TYPE) \
   T (TYPE, abs) \
   T (TYPE, neg)
 
+#define TEST_FLOAT_TYPE(T, TYPE, SUFFIX) \
+  T (TYPE, __builtin_fabs##SUFFIX) \
+  T (TYPE, neg)
+
 #define TEST_ALL(T) \
-  TEST_TYPE (T, int8_t) \
-  TEST_TYPE (T, int16_t) \
-  TEST_TYPE (T, int32_t) \
-  TEST_TYPE (T, int64_t)
+  TEST_INT_TYPE (T, int8_t) \
+  TEST_INT_TYPE (T, int16_t) \
+  TEST_INT_TYPE (T, int32_t) \
+  TEST_INT_TYPE (T, int64_t) \
+  TEST_FLOAT_TYPE (T, _Float16, f16) \
+  TEST_FLOAT_TYPE (T, float, f) \
+  TEST_FLOAT_TYPE (T, double, )
 
 TEST_ALL (DEF_LOOP)
 
@@ -37,11 +44,19 @@ TEST_ALL (DEF_LOOP)
 /* { dg-final { scan-assembler-times {\tneg\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
 /* { dg-final { scan-assembler-times {\tneg\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
 
+/* { dg-final { scan-assembler-times {\tfabs\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfabs\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfabs\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-times {\tfneg\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfneg\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tfneg\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
 /* Really we should be able to use MOVPRFX /z here, but at the moment
    we're relying on combine to merge a SEL and an arithmetic operation,
    and the SEL doesn't allow the "false" value to be zero when the "true"
    value is a register.  */
-/* { dg-final { scan-assembler-times {\tmovprfx\tz[0-9]+, z[0-9]+\n} 8 } } */
+/* { dg-final { scan-assembler-times {\tmovprfx\tz[0-9]+, z[0-9]+\n} 14 } } */
 
 /* { dg-final { scan-assembler-not {\tmov\tz[^\n]*z} } } */
 /* { dg-final { scan-assembler-not {\tsel\t} } } */