+2019-08-15 Richard Sandiford <richard.sandiford@arm.com>
+ Kugan Vivekanandarajah <kugan.vivekanandarajah@linaro.org>
+
+ * config/aarch64/aarch64-sve.md (*aarch64_cond_<su>abd<mode>_2)
+ (*aarch64_cond_<su>abd<mode>_any): New patterns.
+
2019-08-15 Richard Sandiford <richard.sandiford@arm.com>
Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
[(set_attr "movprfx" "*,yes")]
)
+;; Predicated integer absolute difference, merging with the first input.
+(define_insn_and_rewrite "*aarch64_cond_<su>abd<mode>_2"
+ [(set (match_operand:SVE_I 0 "register_operand" "=w, ?&w")
+ (unspec:SVE_I
+ [(match_operand:<VPRED> 1 "register_operand" "Upl, Upl")
+ (minus:SVE_I
+ (unspec:SVE_I
+ [(match_operand 4)
+ (USMAX:SVE_I
+ (match_operand:SVE_I 2 "register_operand" "0, w")
+ (match_operand:SVE_I 3 "register_operand" "w, w"))]
+ UNSPEC_PRED_X)
+ (unspec:SVE_I
+ [(match_operand 5)
+ (<max_opp>:SVE_I
+ (match_dup 2)
+ (match_dup 3))]
+ UNSPEC_PRED_X))
+ (match_dup 2)]
+ UNSPEC_SEL))]
+ "TARGET_SVE"
+ "@
+ <su>abd\t%0.<Vetype>, %1/m, %0.<Vetype>, %3.<Vetype>
+ movprfx\t%0, %2\;<su>abd\t%0.<Vetype>, %1/m, %0.<Vetype>, %3.<Vetype>"
+ "&& (!CONSTANT_P (operands[4]) || !CONSTANT_P (operands[5]))"
+ {
+ operands[4] = operands[5] = CONSTM1_RTX (<VPRED>mode);
+ }
+ [(set_attr "movprfx" "*,yes")]
+)
+
+;; Predicated integer absolute difference, merging with an independent value.
+(define_insn_and_rewrite "*aarch64_cond_<su>abd<mode>_any"
+ [(set (match_operand:SVE_I 0 "register_operand" "=&w, &w, &w, &w, ?&w")
+ (unspec:SVE_I
+ [(match_operand:<VPRED> 1 "register_operand" "Upl, Upl, Upl, Upl, Upl")
+ (minus:SVE_I
+ (unspec:SVE_I
+ [(match_operand 5)
+ (USMAX:SVE_I
+ (match_operand:SVE_I 2 "register_operand" "0, w, w, w, w")
+ (match_operand:SVE_I 3 "register_operand" "w, 0, w, w, w"))]
+ UNSPEC_PRED_X)
+ (unspec:SVE_I
+ [(match_operand 6)
+ (<max_opp>:SVE_I
+ (match_dup 2)
+ (match_dup 3))]
+ UNSPEC_PRED_X))
+ (match_operand:SVE_I 4 "aarch64_simd_reg_or_zero" "Dz, Dz, Dz, 0, w")]
+ UNSPEC_SEL))]
+ "TARGET_SVE
+ && !rtx_equal_p (operands[2], operands[4])
+ && !rtx_equal_p (operands[3], operands[4])"
+ "@
+ movprfx\t%0.<Vetype>, %1/z, %0.<Vetype>\;<su>abd\t%0.<Vetype>, %1/m, %0.<Vetype>, %3.<Vetype>
+ movprfx\t%0.<Vetype>, %1/z, %0.<Vetype>\;<su>abd\t%0.<Vetype>, %1/m, %0.<Vetype>, %2.<Vetype>
+ movprfx\t%0.<Vetype>, %1/z, %2.<Vetype>\;<su>abd\t%0.<Vetype>, %1/m, %0.<Vetype>, %3.<Vetype>
+ movprfx\t%0.<Vetype>, %1/m, %2.<Vetype>\;<su>abd\t%0.<Vetype>, %1/m, %0.<Vetype>, %3.<Vetype>
+ #"
+ "&& 1"
+ {
+ if (!CONSTANT_P (operands[5]) || !CONSTANT_P (operands[6]))
+ operands[5] = operands[6] = CONSTM1_RTX (<VPRED>mode);
+ else if (reload_completed
+ && register_operand (operands[4], <MODE>mode)
+ && !rtx_equal_p (operands[0], operands[4]))
+ {
+ emit_insn (gen_vcond_mask_<mode><vpred> (operands[0], operands[2],
+ operands[4], operands[1]));
+ operands[4] = operands[2] = operands[0];
+ }
+ else
+ FAIL;
+ }
+ [(set_attr "movprfx" "yes")]
+)
+
;; -------------------------------------------------------------------------
;; ---- [INT] Highpart multiplication
;; -------------------------------------------------------------------------
+2019-08-15 Richard Sandiford <richard.sandiford@arm.com>
+ Kugan Vivekanandarajah <kugan.vivekanandarajah@linaro.org>
+
+ * gcc.target/aarch64/sve/cond_abd_1.c: New test.
+ * gcc.target/aarch64/sve/cond_abd_1_run.c: Likewise.
+ * gcc.target/aarch64/sve/cond_abd_2.c: Likewise.
+ * gcc.target/aarch64/sve/cond_abd_2_run.c: Likewise.
+ * gcc.target/aarch64/sve/cond_abd_3.c: Likewise.
+ * gcc.target/aarch64/sve/cond_abd_3_run.c: Likewise.
+ * gcc.target/aarch64/sve/cond_abd_4.c: Likewise.
+ * gcc.target/aarch64/sve/cond_abd_4_run.c: Likewise.
+ * gcc.target/aarch64/sve/cond_abd_5.c: Likewise.
+ * gcc.target/aarch64/sve/cond_abd_5_run.c: Likewise.
+
2019-08-15 Richard Sandiford <richard.sandiford@arm.com>
Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
--- /dev/null
+/* { dg-do compile } */
+/* { dg-options "-O2 -ftree-vectorize" } */
+
+#include <stdint.h>
+
+#define abd(A, B) (((A) < (B) ? (B) : (A)) - ((A) < (B) ? (A) : (B)))
+
+#define DEF_LOOP(TYPE) \
+ void __attribute__ ((noinline, noclone)) \
+ test_##TYPE (TYPE *__restrict r, TYPE *__restrict a, \
+ TYPE *__restrict b, TYPE *__restrict c, \
+ int n) \
+ { \
+ for (int i = 0; i < n; ++i) \
+ r[i] = a[i] < 20 ? abd (b[i], c[i]) : b[i]; \
+ }
+
+#define TEST_ALL(T) \
+ T (int8_t) \
+ T (uint8_t) \
+ T (int16_t) \
+ T (uint16_t) \
+ T (int32_t) \
+ T (uint32_t) \
+ T (int64_t) \
+ T (uint64_t)
+
+TEST_ALL (DEF_LOOP)
+
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.b, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.b, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-not {\tmov\tz[^,]*z} } } */
+/* { dg-final { scan-assembler-not {\tmovprfx\t} } } */
+/* { dg-final { scan-assembler-not {\tsel\t} } } */
--- /dev/null
+/* { dg-do run { target aarch64_sve_hw } } */
+/* { dg-options "-O2 -ftree-vectorize" } */
+
+#include "cond_abd_1.c"
+
+#define N 99
+
+#define TEST_LOOP(TYPE) \
+ { \
+ TYPE r[N], a[N], b[N], c[N]; \
+ for (int i = 0; i < N; ++i) \
+ { \
+ a[i] = (i & 1 ? i : 3 * i); \
+ b[i] = (i >> 4) << (i & 15); \
+ c[i] = ((i + 2) % 3) * (i + 1); \
+ asm volatile ("" ::: "memory"); \
+ } \
+ test_##TYPE (r, a, b, c, N); \
+ for (int i = 0; i < N; ++i) \
+ { \
+ TYPE expected = a[i] < 20 ? abd (b[i], c[i]) : b[i]; \
+ if (r[i] != expected) \
+ __builtin_abort (); \
+ asm volatile ("" ::: "memory"); \
+ } \
+ }
+
+int
+main (void)
+{
+ TEST_ALL (TEST_LOOP)
+ return 0;
+}
--- /dev/null
+/* { dg-do compile } */
+/* { dg-options "-O2 -ftree-vectorize" } */
+
+#include <stdint.h>
+
+#define abd(A, B) (((A) < (B) ? (B) : (A)) - ((A) < (B) ? (A) : (B)))
+
+#define DEF_LOOP(TYPE) \
+ void __attribute__ ((noinline, noclone)) \
+ test_##TYPE (TYPE *__restrict r, TYPE *__restrict a, \
+ TYPE *__restrict b, TYPE *__restrict c, \
+ int n) \
+ { \
+ for (int i = 0; i < n; ++i) \
+ r[i] = a[i] < 20 ? abd (b[i], c[i]) : c[i]; \
+ }
+
+#define TEST_ALL(T) \
+ T (int8_t) \
+ T (uint8_t) \
+ T (int16_t) \
+ T (uint16_t) \
+ T (int32_t) \
+ T (uint32_t) \
+ T (int64_t) \
+ T (uint64_t)
+
+TEST_ALL (DEF_LOOP)
+
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.b, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.b, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-not {\tmov\tz[^,]*z} } } */
+/* { dg-final { scan-assembler-not {\tmovprfx\t} } } */
+/* { dg-final { scan-assembler-not {\tsel\t} } } */
--- /dev/null
+/* { dg-do run { target aarch64_sve_hw } } */
+/* { dg-options "-O2 -ftree-vectorize" } */
+
+#include "cond_abd_2.c"
+
+#define N 99
+
+#define TEST_LOOP(TYPE) \
+ { \
+ TYPE r[N], a[N], b[N], c[N]; \
+ for (int i = 0; i < N; ++i) \
+ { \
+ a[i] = (i & 1 ? i : 3 * i); \
+ b[i] = (i >> 4) << (i & 15); \
+ c[i] = ((i + 2) % 3) * (i + 1); \
+ asm volatile ("" ::: "memory"); \
+ } \
+ test_##TYPE (r, a, b, c, N); \
+ for (int i = 0; i < N; ++i) \
+ { \
+ TYPE expected = a[i] < 20 ? abd (b[i], c[i]) : c[i]; \
+ if (r[i] != expected) \
+ __builtin_abort (); \
+ asm volatile ("" ::: "memory"); \
+ } \
+ }
+
+int
+main (void)
+{
+ TEST_ALL (TEST_LOOP)
+ return 0;
+}
--- /dev/null
+/* { dg-do compile } */
+/* { dg-options "-O2 -ftree-vectorize" } */
+
+#include <stdint.h>
+
+#define abd(A, B) (((A) < (B) ? (B) : (A)) - ((A) < (B) ? (A) : (B)))
+
+#define DEF_LOOP(TYPE) \
+ void __attribute__ ((noinline, noclone)) \
+ test_##TYPE (TYPE *__restrict r, TYPE *__restrict a, \
+ TYPE *__restrict b, TYPE *__restrict c, \
+ int n) \
+ { \
+ for (int i = 0; i < n; ++i) \
+ r[i] = a[i] < 20 ? abd (b[i], c[i]) : a[i]; \
+ }
+
+#define TEST_ALL(T) \
+ T (int8_t) \
+ T (uint8_t) \
+ T (int16_t) \
+ T (uint16_t) \
+ T (int32_t) \
+ T (uint32_t) \
+ T (int64_t) \
+ T (uint64_t)
+
+TEST_ALL (DEF_LOOP)
+
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.b, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.b, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-times {\tmovprfx\tz[0-9]+\.b, p[0-7]/m, z[0-9]+\.b\n} 2 } } */
+/* { dg-final { scan-assembler-times {\tmovprfx\tz[0-9]+\.h, p[0-7]/m, z[0-9]+\.h\n} 2 } } */
+/* { dg-final { scan-assembler-times {\tmovprfx\tz[0-9]+\.s, p[0-7]/m, z[0-9]+\.s\n} 2 } } */
+/* { dg-final { scan-assembler-times {\tmovprfx\tz[0-9]+\.d, p[0-7]/m, z[0-9]+\.d\n} 2 } } */
+
+/* { dg-final { scan-assembler-not {\tmov\tz[^,]*z} } } */
+/* { dg-final { scan-assembler-not {\tsel\t} } } */
--- /dev/null
+/* { dg-do run { target aarch64_sve_hw } } */
+/* { dg-options "-O2 -ftree-vectorize" } */
+
+#include "cond_abd_3.c"
+
+#define N 99
+
+#define TEST_LOOP(TYPE) \
+ { \
+ TYPE r[N], a[N], b[N], c[N]; \
+ for (int i = 0; i < N; ++i) \
+ { \
+ a[i] = (i & 1 ? i : 3 * i); \
+ b[i] = (i >> 4) << (i & 15); \
+ c[i] = ((i + 2) % 3) * (i + 1); \
+ asm volatile ("" ::: "memory"); \
+ } \
+ test_##TYPE (r, a, b, c, N); \
+ for (int i = 0; i < N; ++i) \
+ { \
+ TYPE expected = a[i] < 20 ? abd (b[i], c[i]) : a[i]; \
+ if (r[i] != expected) \
+ __builtin_abort (); \
+ asm volatile ("" ::: "memory"); \
+ } \
+ }
+
+int
+main (void)
+{
+ TEST_ALL (TEST_LOOP)
+ return 0;
+}
--- /dev/null
+/* { dg-do compile } */
+/* { dg-options "-O2 -ftree-vectorize" } */
+
+#include <stdint.h>
+
+#define abd(A, B) (((A) < (B) ? (B) : (A)) - ((A) < (B) ? (A) : (B)))
+
+#define DEF_LOOP(TYPE) \
+ void __attribute__ ((noinline, noclone)) \
+ test_##TYPE (TYPE *__restrict r, TYPE *__restrict a, \
+ TYPE *__restrict b, TYPE *__restrict c, \
+ int n) \
+ { \
+ for (int i = 0; i < n; ++i) \
+ r[i] = a[i] < 20 ? abd (b[i], c[i]) : 79; \
+ }
+
+#define TEST_ALL(T) \
+ T (int8_t) \
+ T (uint8_t) \
+ T (int16_t) \
+ T (uint16_t) \
+ T (int32_t) \
+ T (uint32_t) \
+ T (int64_t) \
+ T (uint64_t)
+
+TEST_ALL (DEF_LOOP)
+
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.b, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.b, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-not {\tmov\tz[^,]*z} } } */
+/* { dg-final { scan-assembler-not {\tmovprfx\t} } } */
+/* { dg-final { scan-assembler-times {\tsel\t} 8 } } */
--- /dev/null
+/* { dg-do run { target aarch64_sve_hw } } */
+/* { dg-options "-O2 -ftree-vectorize" } */
+
+#include "cond_abd_4.c"
+
+#define N 99
+
+#define TEST_LOOP(TYPE) \
+ { \
+ TYPE r[N], a[N], b[N], c[N]; \
+ for (int i = 0; i < N; ++i) \
+ { \
+ a[i] = (i & 1 ? i : 3 * i); \
+ b[i] = (i >> 4) << (i & 15); \
+ c[i] = ((i + 2) % 3) * (i + 1); \
+ asm volatile ("" ::: "memory"); \
+ } \
+ test_##TYPE (r, a, b, c, N); \
+ for (int i = 0; i < N; ++i) \
+ { \
+ TYPE expected = a[i] < 20 ? abd (b[i], c[i]) : 79; \
+ if (r[i] != expected) \
+ __builtin_abort (); \
+ asm volatile ("" ::: "memory"); \
+ } \
+ }
+
+int
+main (void)
+{
+ TEST_ALL (TEST_LOOP)
+ return 0;
+}
--- /dev/null
+/* { dg-do compile } */
+/* { dg-options "-O2 -ftree-vectorize" } */
+
+#include <stdint.h>
+
+#define abd(A, B) (((A) < (B) ? (B) : (A)) - ((A) < (B) ? (A) : (B)))
+
+#define DEF_LOOP(TYPE) \
+ void __attribute__ ((noinline, noclone)) \
+ test_##TYPE (TYPE *__restrict r, TYPE *__restrict a, \
+ TYPE *__restrict b, TYPE *__restrict c, \
+ int n) \
+ { \
+ for (int i = 0; i < n; ++i) \
+ r[i] = a[i] < 20 ? abd (b[i], c[i]) : 0; \
+ }
+
+#define TEST_ALL(T) \
+ T (int8_t) \
+ T (uint8_t) \
+ T (int16_t) \
+ T (uint16_t) \
+ T (int32_t) \
+ T (uint32_t) \
+ T (int64_t) \
+ T (uint64_t)
+
+TEST_ALL (DEF_LOOP)
+
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.b, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tsabd\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.b, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.h, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.s, p[0-7]/m,} 1 } } */
+/* { dg-final { scan-assembler-times {\tuabd\tz[0-9]+\.d, p[0-7]/m,} 1 } } */
+
+/* { dg-final { scan-assembler-times {\tmovprfx\tz[0-9]+\.b, p[0-7]/z, z[0-9]+\.b\n} 2 } } */
+/* { dg-final { scan-assembler-times {\tmovprfx\tz[0-9]+\.h, p[0-7]/z, z[0-9]+\.h\n} 2 } } */
+/* { dg-final { scan-assembler-times {\tmovprfx\tz[0-9]+\.s, p[0-7]/z, z[0-9]+\.s\n} 2 } } */
+/* { dg-final { scan-assembler-times {\tmovprfx\tz[0-9]+\.d, p[0-7]/z, z[0-9]+\.d\n} 2 } } */
+
+/* { dg-final { scan-assembler-not {\tmov\tz[^,]*z} } } */
+/* { dg-final { scan-assembler-not {\tsel\t} } } */
--- /dev/null
+/* { dg-do run { target aarch64_sve_hw } } */
+/* { dg-options "-O2 -ftree-vectorize" } */
+
+#include "cond_abd_5.c"
+
+#define N 99
+
+#define TEST_LOOP(TYPE) \
+ { \
+ TYPE r[N], a[N], b[N], c[N]; \
+ for (int i = 0; i < N; ++i) \
+ { \
+ a[i] = (i & 1 ? i : 3 * i); \
+ b[i] = (i >> 4) << (i & 15); \
+ c[i] = ((i + 2) % 3) * (i + 1); \
+ asm volatile ("" ::: "memory"); \
+ } \
+ test_##TYPE (r, a, b, c, N); \
+ for (int i = 0; i < N; ++i) \
+ { \
+ TYPE expected = a[i] < 20 ? abd (b[i], c[i]) : 0; \
+ if (r[i] != expected) \
+ __builtin_abort (); \
+ asm volatile ("" ::: "memory"); \
+ } \
+ }
+
+int
+main (void)
+{
+ TEST_ALL (TEST_LOOP)
+ return 0;
+}
projects / gcc.git / commitdiff