[20/21] Arm: Add Advanced SIMD cbranch implementation
Checks
Commit Message
Hi All,
This adds an implementation for conditional branch optab for AArch32.
The previous version only allowed operand 0 but it looks like cbranch
expansion does not check with the target and so we have to implement all.
I therefore did not commit it. This is a larger version.
For e.g.
void f1 ()
{
for (int i = 0; i < N; i++)
{
b[i] += a[i];
if (a[i] > 0)
break;
}
}
For 128-bit vectors we generate:
vcgt.s32 q8, q9, #0
vpmax.u32 d7, d16, d17
vpmax.u32 d7, d7, d7
vmov r3, s14 @ int
cmp r3, #0
and of 64-bit vector we can omit one vpmax as we still need to compress to
32-bits.
Bootstrapped Regtested on arm-none-linux-gnueabihf and no issues.
Ok for master?
Thanks,
Tamar
gcc/ChangeLog:
* config/arm/neon.md (cbranch<mode>4): New.
gcc/testsuite/ChangeLog:
* gcc.dg/vect/vect-early-break_2.c: Skip Arm.
* gcc.dg/vect/vect-early-break_7.c: Likewise.
* gcc.dg/vect/vect-early-break_75.c: Likewise.
* gcc.dg/vect/vect-early-break_77.c: Likewise.
* gcc.dg/vect/vect-early-break_82.c: Likewise.
* gcc.dg/vect/vect-early-break_88.c: Likewise.
* lib/target-supports.exp (add_options_for_vect_early_break,
check_effective_target_vect_early_break_hw,
check_effective_target_vect_early_break): Support AArch32.
* gcc.target/arm/vect-early-break-cbranch.c: New test.
--- inline copy of patch --
diff --git a/gcc/config/arm/neon.md b/gcc/config/arm/neon.md
index d213369ffc38fb88ad0357d848cc7da5af73bab7..0f088a51d31e6882bc0fabbad99862b8b465dd22 100644
--
diff --git a/gcc/config/arm/neon.md b/gcc/config/arm/neon.md
index d213369ffc38fb88ad0357d848cc7da5af73bab7..0f088a51d31e6882bc0fabbad99862b8b465dd22 100644
--- a/gcc/config/arm/neon.md
+++ b/gcc/config/arm/neon.md
@@ -408,6 +408,54 @@ (define_insn "vec_extract<mode><V_elem_l>"
[(set_attr "type" "neon_store1_one_lane<q>,neon_to_gp<q>")]
)
+;; Patterns comparing two vectors and conditionally jump.
+;; Avdanced SIMD lacks a vector != comparison, but this is a quite common
+;; operation. To not pay the penalty for inverting == we can map our any
+;; comparisons to all i.e. any(~x) => all(x).
+;;
+;; However unlike the AArch64 version, we can't optimize this further as the
+;; chain is too long for combine due to these being unspecs so it doesn't fold
+;; the operation to something simpler.
+(define_expand "cbranch<mode>4"
+ [(set (pc) (if_then_else
+ (match_operator 0 "expandable_comparison_operator"
+ [(match_operand:VDQI 1 "register_operand")
+ (match_operand:VDQI 2 "reg_or_zero_operand")])
+ (label_ref (match_operand 3 "" ""))
+ (pc)))]
+ "TARGET_NEON"
+{
+ rtx mask = operands[1];
+
+ /* If comparing against a non-zero vector we have to do a comparison first
+ so we can have a != 0 comparison with the result. */
+ if (operands[2] != CONST0_RTX (<MODE>mode))
+ {
+ mask = gen_reg_rtx (<MODE>mode);
+ emit_insn (gen_xor<mode>3 (mask, operands[1], operands[2]));
+ }
+
+ /* For 128-bit vectors we need an additional reductions. */
+ if (known_eq (128, GET_MODE_BITSIZE (<MODE>mode)))
+ {
+ /* Always reduce using a V4SI. */
+ mask = gen_reg_rtx (V2SImode);
+ rtx low = gen_reg_rtx (V2SImode);
+ rtx high = gen_reg_rtx (V2SImode);
+ rtx op1 = simplify_gen_subreg (V4SImode, operands[1], <MODE>mode, 0);
+ emit_insn (gen_neon_vget_lowv4si (low, op1));
+ emit_insn (gen_neon_vget_highv4si (high, op1));
+ emit_insn (gen_neon_vpumaxv2si (mask, low, high));
+ }
+
+ emit_insn (gen_neon_vpumaxv2si (mask, mask, mask));
+
+ rtx val = gen_reg_rtx (SImode);
+ emit_move_insn (val, gen_lowpart (SImode, mask));
+ emit_jump_insn (gen_cbranch_cc (operands[0], val, const0_rtx, operands[3]));
+ DONE;
+})
+
;; This pattern is renamed from "vec_extract<mode><V_elem_l>" to
;; "neon_vec_extract<mode><V_elem_l>" and this pattern is called
;; by define_expand in vec-common.md file.
diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
index 5c32bf94409e9743e72429985ab3bf13aab8f2c1..dec0b492ab883de6e02944a95fd554a109a68a39 100644
--- a/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
+++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
@@ -5,7 +5,7 @@
/* { dg-additional-options "-Ofast" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include <complex.h>
diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
index 8c86c5034d7522b3733543fb384a23c5d6ed0fcf..d218a0686719fee4c167684dcf26402851b53260 100644
--- a/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
+++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
@@ -5,7 +5,7 @@
/* { dg-additional-options "-Ofast" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include <complex.h>
diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
index ed27f8635730ff0d8803517c72693625a2feddef..9dcc3372acd657458df8d94ce36c4bd96f02fd52 100644
--- a/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
+++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
@@ -3,7 +3,7 @@
/* { dg-require-effective-target vect_int } */
/* { dg-additional-options "-O3" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "x86_64-*-* i?86-*-*" } } } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "x86_64-*-* i?86-*-* arm*-*-*" } } } } */
#include <limits.h>
#include <assert.h>
diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
index 225106aab0a3efc7536de6f6e45bc6ff16210ea8..9fa7e6948ebfb5f1723833653fd6ad1fc65f4e8e 100644
--- a/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
+++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
@@ -3,7 +3,7 @@
/* { dg-require-effective-target vect_int } */
/* { dg-additional-options "-O3" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include "tree-vect.h"
diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
index 0e9b2d8d385c556063a3c6fcb14383317b056a79..7cd21d33485f3abb823e1943c87e9481c41fd2c3 100644
--- a/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
+++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
@@ -5,7 +5,7 @@
/* { dg-additional-options "-Ofast" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include <complex.h>
diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
index b392dd46553994d813761da41c42989a79b90119..59ed57c5fb5f3e8197fc20058eeb0a81a55815cc 100644
--- a/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
+++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
@@ -3,7 +3,7 @@
/* { dg-require-effective-target vect_int } */
/* { dg-additional-options "-Ofast --param vect-partial-vector-usage=2" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include "tree-vect.h"
diff --git a/gcc/testsuite/gcc.target/arm/vect-early-break-cbranch.c b/gcc/testsuite/gcc.target/arm/vect-early-break-cbranch.c
new file mode 100644
index 0000000000000000000000000000000000000000..0e9a39d231fdf4cb56590945e7cedfabd11d39b5
--- /dev/null
+++ b/gcc/testsuite/gcc.target/arm/vect-early-break-cbranch.c
@@ -0,0 +1,138 @@
+/* { dg-do compile } */
+/* { dg-require-effective-target vect_early_break } */
+/* { dg-require-effective-target arm_neon_ok } */
+/* { dg-require-effective-target arm32 } */
+/* { dg-options "-O3 -march=armv8-a+simd -mfpu=auto -mfloat-abi=hard -fno-schedule-insns -fno-reorder-blocks -fno-schedule-insns2" } */
+/* { dg-final { check-function-bodies "**" "" "" } } */
+
+#define N 640
+int a[N] = {0};
+int b[N] = {0};
+
+/*
+** f1:
+** ...
+** vcgt.s32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f1 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] > 0)
+ break;
+ }
+}
+
+/*
+** f2:
+** ...
+** vcge.s32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f2 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] >= 0)
+ break;
+ }
+}
+
+/*
+** f3:
+** ...
+** vceq.i32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f3 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] == 0)
+ break;
+ }
+}
+
+/*
+** f4:
+** ...
+** vceq.i32 q[0-9]+, q[0-9]+, #0
+** vmvn q[0-9]+, q[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f4 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] != 0)
+ break;
+ }
+}
+
+/*
+** f5:
+** ...
+** vclt.s32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f5 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] < 0)
+ break;
+ }
+}
+
+/*
+** f6:
+** ...
+** vcle.s32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f6 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] <= 0)
+ break;
+ }
+}
+
diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
index 05fc417877bcd658931061b7245eb8ba5abd2e09..24a937dbb59b5723af038bd9e0b89369595fcf87 100644
--- a/gcc/testsuite/lib/target-supports.exp
+++ b/gcc/testsuite/lib/target-supports.exp
@@ -4059,6 +4059,7 @@ proc check_effective_target_vect_early_break { } {
return [check_cached_effective_target_indexed vect_early_break {
expr {
[istarget aarch64*-*-*]
+ || [check_effective_target_arm_v8_neon_ok]
|| [check_effective_target_sse4]
}}]
}
@@ -4072,6 +4073,7 @@ proc check_effective_target_vect_early_break_hw { } {
return [check_cached_effective_target_indexed vect_early_break_hw {
expr {
[istarget aarch64*-*-*]
+ || [check_effective_target_arm_v8_neon_hw]
|| [check_sse4_hw_available]
}}]
}
@@ -4081,6 +4083,11 @@ proc add_options_for_vect_early_break { flags } {
return "$flags"
}
+ if { [check_effective_target_arm_v8_neon_ok] } {
+ global et_arm_v8_neon_flags
+ return "$flags $et_arm_v8_neon_flags -march=armv8-a"
+ }
+
if { [check_effective_target_sse4] } {
return "$flags -msse4.1"
}
Comments
Ping,
---
Hi All,
This adds an implementation for conditional branch optab for AArch32.
The previous version only allowed operand 0 but it looks like cbranch
expansion does not check with the target and so we have to implement all.
I therefore did not commit it. This is a larger version. I've also dropped the MVE
version because the mid-end can rewrite the comparison into comparing two
predicates without checking with the backend. Since MVE only has 1 predicate
register this would need to go through memory and two MRS calls. It's unlikely
to be beneficial and so that's for GCC 15 when I can fix the middle-end.
The cases where AArch32 is skipped in the testsuite are all missed-optimizations
due to AArch32 missing some optabs.
For e.g.
void f1 ()
{
for (int i = 0; i < N; i++)
{
b[i] += a[i];
if (a[i] > 0)
break;
}
}
For 128-bit vectors we generate:
vcgt.s32 q8, q9, #0
vpmax.u32 d7, d16, d17
vpmax.u32 d7, d7, d7
vmov r3, s14 @ int
cmp r3, #0
and of 64-bit vector we can omit one vpmax as we still need to compress to
32-bits.
Bootstrapped Regtested on arm-none-linux-gnueabihf and no issues.
Ok for master?
Thanks,
Tamar
gcc/ChangeLog:
* config/arm/neon.md (cbranch<mode>4): New.
gcc/testsuite/ChangeLog:
* gcc.dg/vect/vect-early-break_2.c: Skip Arm.
* gcc.dg/vect/vect-early-break_7.c: Likewise.
* gcc.dg/vect/vect-early-break_75.c: Likewise.
* gcc.dg/vect/vect-early-break_77.c: Likewise.
* gcc.dg/vect/vect-early-break_82.c: Likewise.
* gcc.dg/vect/vect-early-break_88.c: Likewise.
* lib/target-supports.exp (add_options_for_vect_early_break,
check_effective_target_vect_early_break_hw,
check_effective_target_vect_early_break): Support AArch32.
* gcc.target/arm/vect-early-break-cbranch.c: New test.
--- inline version of patch ---
diff --git a/gcc/config/arm/neon.md b/gcc/config/arm/neon.md
index d213369ffc38fb88ad0357d848cc7da5af73bab7..ed659ab736862da416d1ff6241d0d3e6c6b96ff1 100644
--- a/gcc/config/arm/neon.md
+++ b/gcc/config/arm/neon.md
@@ -408,6 +408,55 @@ (define_insn "vec_extract<mode><V_elem_l>"
[(set_attr "type" "neon_store1_one_lane<q>,neon_to_gp<q>")]
)
+;; Patterns comparing two vectors and conditionally jump.
+;; Avdanced SIMD lacks a vector != comparison, but this is a quite common
+;; operation. To not pay the penalty for inverting == we can map our any
+;; comparisons to all i.e. any(~x) => all(x).
+;;
+;; However unlike the AArch64 version, we can't optimize this further as the
+;; chain is too long for combine due to these being unspecs so it doesn't fold
+;; the operation to something simpler.
+(define_expand "cbranch<mode>4"
+ [(set (pc) (if_then_else
+ (match_operator 0 "expandable_comparison_operator"
+ [(match_operand:VDQI 1 "register_operand")
+ (match_operand:VDQI 2 "reg_or_zero_operand")])
+ (label_ref (match_operand 3 "" ""))
+ (pc)))]
+ "TARGET_NEON"
+{
+ rtx mask = operands[1];
+
+ /* If comparing against a non-zero vector we have to do a comparison first
+ so we can have a != 0 comparison with the result. */
+ if (operands[2] != CONST0_RTX (<MODE>mode))
+ {
+ mask = gen_reg_rtx (<MODE>mode);
+ emit_insn (gen_xor<mode>3 (mask, operands[1], operands[2]));
+ }
+
+ /* For 128-bit vectors we need an additional reductions. */
+ if (known_eq (128, GET_MODE_BITSIZE (<MODE>mode)))
+ {
+ /* Always reduce using a V4SI. */
+ mask = gen_reg_rtx (V2SImode);
+ rtx low = gen_reg_rtx (V2SImode);
+ rtx high = gen_reg_rtx (V2SImode);
+ rtx op1 = lowpart_subreg (V4SImode, operands[1], <MODE>mode);
+ emit_insn (gen_neon_vget_lowv4si (low, op1));
+ emit_insn (gen_neon_vget_highv4si (high, op1));
+ emit_insn (gen_neon_vpumaxv2si (mask, low, high));
+ }
+
+ rtx op1 = lowpart_subreg (V2SImode, mask, GET_MODE (mask));
+ emit_insn (gen_neon_vpumaxv2si (op1, op1, op1));
+
+ rtx val = gen_reg_rtx (SImode);
+ emit_move_insn (val, gen_lowpart (SImode, mask));
+ emit_jump_insn (gen_cbranch_cc (operands[0], val, const0_rtx, operands[3]));
+ DONE;
+})
+
;; This pattern is renamed from "vec_extract<mode><V_elem_l>" to
;; "neon_vec_extract<mode><V_elem_l>" and this pattern is called
;; by define_expand in vec-common.md file.
diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
index 5c32bf94409e9743e72429985ab3bf13aab8f2c1..dec0b492ab883de6e02944a95fd554a109a68a39 100644
--- a/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
+++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
@@ -5,7 +5,7 @@
/* { dg-additional-options "-Ofast" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include <complex.h>
diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
index 8c86c5034d7522b3733543fb384a23c5d6ed0fcf..d218a0686719fee4c167684dcf26402851b53260 100644
--- a/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
+++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
@@ -5,7 +5,7 @@
/* { dg-additional-options "-Ofast" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include <complex.h>
diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
index ed27f8635730ff0d8803517c72693625a2feddef..9dcc3372acd657458df8d94ce36c4bd96f02fd52 100644
--- a/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
+++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
@@ -3,7 +3,7 @@
/* { dg-require-effective-target vect_int } */
/* { dg-additional-options "-O3" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "x86_64-*-* i?86-*-*" } } } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "x86_64-*-* i?86-*-* arm*-*-*" } } } } */
#include <limits.h>
#include <assert.h>
diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
index 225106aab0a3efc7536de6f6e45bc6ff16210ea8..9fa7e6948ebfb5f1723833653fd6ad1fc65f4e8e 100644
--- a/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
+++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
@@ -3,7 +3,7 @@
/* { dg-require-effective-target vect_int } */
/* { dg-additional-options "-O3" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include "tree-vect.h"
diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
index 0e9b2d8d385c556063a3c6fcb14383317b056a79..7cd21d33485f3abb823e1943c87e9481c41fd2c3 100644
--- a/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
+++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
@@ -5,7 +5,7 @@
/* { dg-additional-options "-Ofast" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include <complex.h>
diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
index b392dd46553994d813761da41c42989a79b90119..59ed57c5fb5f3e8197fc20058eeb0a81a55815cc 100644
--- a/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
+++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
@@ -3,7 +3,7 @@
/* { dg-require-effective-target vect_int } */
/* { dg-additional-options "-Ofast --param vect-partial-vector-usage=2" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include "tree-vect.h"
diff --git a/gcc/testsuite/gcc.target/arm/vect-early-break-cbranch.c b/gcc/testsuite/gcc.target/arm/vect-early-break-cbranch.c
new file mode 100644
index 0000000000000000000000000000000000000000..0e9a39d231fdf4cb56590945e7cedfabd11d39b5
--- /dev/null
+++ b/gcc/testsuite/gcc.target/arm/vect-early-break-cbranch.c
@@ -0,0 +1,138 @@
+/* { dg-do compile } */
+/* { dg-require-effective-target vect_early_break } */
+/* { dg-require-effective-target arm_neon_ok } */
+/* { dg-require-effective-target arm32 } */
+/* { dg-options "-O3 -march=armv8-a+simd -mfpu=auto -mfloat-abi=hard -fno-schedule-insns -fno-reorder-blocks -fno-schedule-insns2" } */
+/* { dg-final { check-function-bodies "**" "" "" } } */
+
+#define N 640
+int a[N] = {0};
+int b[N] = {0};
+
+/*
+** f1:
+** ...
+** vcgt.s32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f1 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] > 0)
+ break;
+ }
+}
+
+/*
+** f2:
+** ...
+** vcge.s32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f2 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] >= 0)
+ break;
+ }
+}
+
+/*
+** f3:
+** ...
+** vceq.i32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f3 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] == 0)
+ break;
+ }
+}
+
+/*
+** f4:
+** ...
+** vceq.i32 q[0-9]+, q[0-9]+, #0
+** vmvn q[0-9]+, q[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f4 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] != 0)
+ break;
+ }
+}
+
+/*
+** f5:
+** ...
+** vclt.s32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f5 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] < 0)
+ break;
+ }
+}
+
+/*
+** f6:
+** ...
+** vcle.s32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f6 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] <= 0)
+ break;
+ }
+}
+
diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
index 05fc417877bcd658931061b7245eb8ba5abd2e09..24a937dbb59b5723af038bd9e0b89369595fcf87 100644
--- a/gcc/testsuite/lib/target-supports.exp
+++ b/gcc/testsuite/lib/target-supports.exp
@@ -4059,6 +4059,7 @@ proc check_effective_target_vect_early_break { } {
return [check_cached_effective_target_indexed vect_early_break {
expr {
[istarget aarch64*-*-*]
+ || [check_effective_target_arm_v8_neon_ok]
|| [check_effective_target_sse4]
}}]
}
@@ -4072,6 +4073,7 @@ proc check_effective_target_vect_early_break_hw { } {
return [check_cached_effective_target_indexed vect_early_break_hw {
expr {
[istarget aarch64*-*-*]
+ || [check_effective_target_arm_v8_neon_hw]
|| [check_sse4_hw_available]
}}]
}
@@ -4081,6 +4083,11 @@ proc add_options_for_vect_early_break { flags } {
return "$flags"
}
+ if { [check_effective_target_arm_v8_neon_ok] } {
+ global et_arm_v8_neon_flags
+ return "$flags $et_arm_v8_neon_flags -march=armv8-a"
+ }
+
if { [check_effective_target_sse4] } {
return "$flags -msse4.1"
}
Hi Tamar,
> -----Original Message-----
> From: Tamar Christina <Tamar.Christina@arm.com>
> Sent: Thursday, January 4, 2024 11:06 AM
> To: Tamar Christina <Tamar.Christina@arm.com>; gcc-patches@gcc.gnu.org
> Cc: nd <nd@arm.com>; Ramana Radhakrishnan
> <Ramana.Radhakrishnan@arm.com>; Richard Earnshaw
> <Richard.Earnshaw@arm.com>; nickc@redhat.com; Kyrylo Tkachov
> <Kyrylo.Tkachov@arm.com>
> Subject: RE: [PATCH 20/21]Arm: Add Advanced SIMD cbranch implementation
>
> Ping,
>
> ---
>
> Hi All,
>
> This adds an implementation for conditional branch optab for AArch32.
> The previous version only allowed operand 0 but it looks like cbranch
> expansion does not check with the target and so we have to implement all.
>
> I therefore did not commit it. This is a larger version. I've also dropped the MVE
> version because the mid-end can rewrite the comparison into comparing two
> predicates without checking with the backend. Since MVE only has 1 predicate
> register this would need to go through memory and two MRS calls. It's unlikely
> to be beneficial and so that's for GCC 15 when I can fix the middle-end.
>
> The cases where AArch32 is skipped in the testsuite are all missed-optimizations
> due to AArch32 missing some optabs.
Does the testsuite have vect_* checks that can be used instead of target arm*?
If so let's use those.
Otherwise it's okay as is.
Thanks,
Kyrill
>
> For e.g.
>
> void f1 ()
> {
> for (int i = 0; i < N; i++)
> {
> b[i] += a[i];
> if (a[i] > 0)
> break;
> }
> }
>
> For 128-bit vectors we generate:
>
> vcgt.s32 q8, q9, #0
> vpmax.u32 d7, d16, d17
> vpmax.u32 d7, d7, d7
> vmov r3, s14 @ int
> cmp r3, #0
>
> and of 64-bit vector we can omit one vpmax as we still need to compress to
> 32-bits.
>
> Bootstrapped Regtested on arm-none-linux-gnueabihf and no issues.
>
> Ok for master?
>
> Thanks,
> Tamar
>
> gcc/ChangeLog:
>
> * config/arm/neon.md (cbranch<mode>4): New.
>
> gcc/testsuite/ChangeLog:
>
> * gcc.dg/vect/vect-early-break_2.c: Skip Arm.
> * gcc.dg/vect/vect-early-break_7.c: Likewise.
> * gcc.dg/vect/vect-early-break_75.c: Likewise.
> * gcc.dg/vect/vect-early-break_77.c: Likewise.
> * gcc.dg/vect/vect-early-break_82.c: Likewise.
> * gcc.dg/vect/vect-early-break_88.c: Likewise.
> * lib/target-supports.exp (add_options_for_vect_early_break,
> check_effective_target_vect_early_break_hw,
> check_effective_target_vect_early_break): Support AArch32.
> * gcc.target/arm/vect-early-break-cbranch.c: New test.
>
> --- inline version of patch ---
>
> diff --git a/gcc/config/arm/neon.md b/gcc/config/arm/neon.md
> index
> d213369ffc38fb88ad0357d848cc7da5af73bab7..ed659ab736862da416d1ff6241d
> 0d3e6c6b96ff1 100644
> --- a/gcc/config/arm/neon.md
> +++ b/gcc/config/arm/neon.md
> @@ -408,6 +408,55 @@ (define_insn "vec_extract<mode><V_elem_l>"
> [(set_attr "type" "neon_store1_one_lane<q>,neon_to_gp<q>")]
> )
>
> +;; Patterns comparing two vectors and conditionally jump.
> +;; Avdanced SIMD lacks a vector != comparison, but this is a quite common
> +;; operation. To not pay the penalty for inverting == we can map our any
> +;; comparisons to all i.e. any(~x) => all(x).
> +;;
> +;; However unlike the AArch64 version, we can't optimize this further as the
> +;; chain is too long for combine due to these being unspecs so it doesn't fold
> +;; the operation to something simpler.
> +(define_expand "cbranch<mode>4"
> + [(set (pc) (if_then_else
> + (match_operator 0 "expandable_comparison_operator"
> + [(match_operand:VDQI 1 "register_operand")
> + (match_operand:VDQI 2 "reg_or_zero_operand")])
> + (label_ref (match_operand 3 "" ""))
> + (pc)))]
> + "TARGET_NEON"
> +{
> + rtx mask = operands[1];
> +
> + /* If comparing against a non-zero vector we have to do a comparison first
> + so we can have a != 0 comparison with the result. */
> + if (operands[2] != CONST0_RTX (<MODE>mode))
> + {
> + mask = gen_reg_rtx (<MODE>mode);
> + emit_insn (gen_xor<mode>3 (mask, operands[1], operands[2]));
> + }
> +
> + /* For 128-bit vectors we need an additional reductions. */
> + if (known_eq (128, GET_MODE_BITSIZE (<MODE>mode)))
> + {
> + /* Always reduce using a V4SI. */
> + mask = gen_reg_rtx (V2SImode);
> + rtx low = gen_reg_rtx (V2SImode);
> + rtx high = gen_reg_rtx (V2SImode);
> + rtx op1 = lowpart_subreg (V4SImode, operands[1], <MODE>mode);
> + emit_insn (gen_neon_vget_lowv4si (low, op1));
> + emit_insn (gen_neon_vget_highv4si (high, op1));
> + emit_insn (gen_neon_vpumaxv2si (mask, low, high));
> + }
> +
> + rtx op1 = lowpart_subreg (V2SImode, mask, GET_MODE (mask));
> + emit_insn (gen_neon_vpumaxv2si (op1, op1, op1));
> +
> + rtx val = gen_reg_rtx (SImode);
> + emit_move_insn (val, gen_lowpart (SImode, mask));
> + emit_jump_insn (gen_cbranch_cc (operands[0], val, const0_rtx, operands[3]));
> + DONE;
> +})
> +
> ;; This pattern is renamed from "vec_extract<mode><V_elem_l>" to
> ;; "neon_vec_extract<mode><V_elem_l>" and this pattern is called
> ;; by define_expand in vec-common.md file.
> diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
> index
> 5c32bf94409e9743e72429985ab3bf13aab8f2c1..dec0b492ab883de6e02944a95f
> d554a109a68a39 100644
> --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
> +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
> @@ -5,7 +5,7 @@
>
> /* { dg-additional-options "-Ofast" } */
>
> -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
> +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-
> *" } } } } */
>
> #include <complex.h>
>
> diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
> index
> 8c86c5034d7522b3733543fb384a23c5d6ed0fcf..d218a0686719fee4c167684dcf2
> 6402851b53260 100644
> --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
> +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
> @@ -5,7 +5,7 @@
>
> /* { dg-additional-options "-Ofast" } */
>
> -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
> +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-
> *" } } } } */
>
> #include <complex.h>
>
> diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
> index
> ed27f8635730ff0d8803517c72693625a2feddef..9dcc3372acd657458df8d94ce36
> c4bd96f02fd52 100644
> --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
> +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
> @@ -3,7 +3,7 @@
> /* { dg-require-effective-target vect_int } */
>
> /* { dg-additional-options "-O3" } */
> -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "x86_64-*-
> * i?86-*-*" } } } } */
> +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "x86_64-
> *-* i?86-*-* arm*-*-*" } } } } */
>
> #include <limits.h>
> #include <assert.h>
> diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
> index
> 225106aab0a3efc7536de6f6e45bc6ff16210ea8..9fa7e6948ebfb5f1723833653fd
> 6ad1fc65f4e8e 100644
> --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
> +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
> @@ -3,7 +3,7 @@
> /* { dg-require-effective-target vect_int } */
>
> /* { dg-additional-options "-O3" } */
> -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
> +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-
> *" } } } } */
>
> #include "tree-vect.h"
>
> diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
> index
> 0e9b2d8d385c556063a3c6fcb14383317b056a79..7cd21d33485f3abb823e1943c
> 87e9481c41fd2c3 100644
> --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
> +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
> @@ -5,7 +5,7 @@
>
> /* { dg-additional-options "-Ofast" } */
>
> -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
> +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-
> *" } } } } */
>
> #include <complex.h>
>
> diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
> b/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
> index
> b392dd46553994d813761da41c42989a79b90119..59ed57c5fb5f3e8197fc20058
> eeb0a81a55815cc 100644
> --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
> +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
> @@ -3,7 +3,7 @@
> /* { dg-require-effective-target vect_int } */
>
> /* { dg-additional-options "-Ofast --param vect-partial-vector-usage=2" } */
> -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
> +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-
> *" } } } } */
>
> #include "tree-vect.h"
>
> diff --git a/gcc/testsuite/gcc.target/arm/vect-early-break-cbranch.c
> b/gcc/testsuite/gcc.target/arm/vect-early-break-cbranch.c
> new file mode 100644
> index
> 0000000000000000000000000000000000000000..0e9a39d231fdf4cb56590945e
> 7cedfabd11d39b5
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/arm/vect-early-break-cbranch.c
> @@ -0,0 +1,138 @@
> +/* { dg-do compile } */
> +/* { dg-require-effective-target vect_early_break } */
> +/* { dg-require-effective-target arm_neon_ok } */
> +/* { dg-require-effective-target arm32 } */
> +/* { dg-options "-O3 -march=armv8-a+simd -mfpu=auto -mfloat-abi=hard -fno-
> schedule-insns -fno-reorder-blocks -fno-schedule-insns2" } */
> +/* { dg-final { check-function-bodies "**" "" "" } } */
> +
> +#define N 640
> +int a[N] = {0};
> +int b[N] = {0};
> +
> +/*
> +** f1:
> +** ...
> +** vcgt.s32 q[0-9]+, q[0-9]+, #0
> +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> +** vmov r[0-9]+, s[0-9]+ @ int
> +** cmp r[0-9]+, #0
> +** bne \.L[0-9]+
> +** ...
> +*/
> +void f1 ()
> +{
> + for (int i = 0; i < N; i++)
> + {
> + b[i] += a[i];
> + if (a[i] > 0)
> + break;
> + }
> +}
> +
> +/*
> +** f2:
> +** ...
> +** vcge.s32 q[0-9]+, q[0-9]+, #0
> +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> +** vmov r[0-9]+, s[0-9]+ @ int
> +** cmp r[0-9]+, #0
> +** bne \.L[0-9]+
> +** ...
> +*/
> +void f2 ()
> +{
> + for (int i = 0; i < N; i++)
> + {
> + b[i] += a[i];
> + if (a[i] >= 0)
> + break;
> + }
> +}
> +
> +/*
> +** f3:
> +** ...
> +** vceq.i32 q[0-9]+, q[0-9]+, #0
> +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> +** vmov r[0-9]+, s[0-9]+ @ int
> +** cmp r[0-9]+, #0
> +** bne \.L[0-9]+
> +** ...
> +*/
> +void f3 ()
> +{
> + for (int i = 0; i < N; i++)
> + {
> + b[i] += a[i];
> + if (a[i] == 0)
> + break;
> + }
> +}
> +
> +/*
> +** f4:
> +** ...
> +** vceq.i32 q[0-9]+, q[0-9]+, #0
> +** vmvn q[0-9]+, q[0-9]+
> +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> +** vmov r[0-9]+, s[0-9]+ @ int
> +** cmp r[0-9]+, #0
> +** bne \.L[0-9]+
> +** ...
> +*/
> +void f4 ()
> +{
> + for (int i = 0; i < N; i++)
> + {
> + b[i] += a[i];
> + if (a[i] != 0)
> + break;
> + }
> +}
> +
> +/*
> +** f5:
> +** ...
> +** vclt.s32 q[0-9]+, q[0-9]+, #0
> +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> +** vmov r[0-9]+, s[0-9]+ @ int
> +** cmp r[0-9]+, #0
> +** bne \.L[0-9]+
> +** ...
> +*/
> +void f5 ()
> +{
> + for (int i = 0; i < N; i++)
> + {
> + b[i] += a[i];
> + if (a[i] < 0)
> + break;
> + }
> +}
> +
> +/*
> +** f6:
> +** ...
> +** vcle.s32 q[0-9]+, q[0-9]+, #0
> +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> +** vmov r[0-9]+, s[0-9]+ @ int
> +** cmp r[0-9]+, #0
> +** bne \.L[0-9]+
> +** ...
> +*/
> +void f6 ()
> +{
> + for (int i = 0; i < N; i++)
> + {
> + b[i] += a[i];
> + if (a[i] <= 0)
> + break;
> + }
> +}
> +
> diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-
> supports.exp
> index
> 05fc417877bcd658931061b7245eb8ba5abd2e09..24a937dbb59b5723af038bd9e
> 0b89369595fcf87 100644
> --- a/gcc/testsuite/lib/target-supports.exp
> +++ b/gcc/testsuite/lib/target-supports.exp
> @@ -4059,6 +4059,7 @@ proc check_effective_target_vect_early_break { } {
> return [check_cached_effective_target_indexed vect_early_break {
> expr {
> [istarget aarch64*-*-*]
> + || [check_effective_target_arm_v8_neon_ok]
> || [check_effective_target_sse4]
> }}]
> }
> @@ -4072,6 +4073,7 @@ proc check_effective_target_vect_early_break_hw { }
> {
> return [check_cached_effective_target_indexed vect_early_break_hw {
> expr {
> [istarget aarch64*-*-*]
> + || [check_effective_target_arm_v8_neon_hw]
> || [check_sse4_hw_available]
> }}]
> }
> @@ -4081,6 +4083,11 @@ proc add_options_for_vect_early_break { flags } {
> return "$flags"
> }
>
> + if { [check_effective_target_arm_v8_neon_ok] } {
> + global et_arm_v8_neon_flags
> + return "$flags $et_arm_v8_neon_flags -march=armv8-a"
> + }
> +
> if { [check_effective_target_sse4] } {
> return "$flags -msse4.1"
> }
> -----Original Message-----
> From: Kyrylo Tkachov <Kyrylo.Tkachov@arm.com>
> Sent: Thursday, January 4, 2024 11:12 AM
> To: Tamar Christina <Tamar.Christina@arm.com>; gcc-patches@gcc.gnu.org
> Cc: nd <nd@arm.com>; Ramana Radhakrishnan
> <Ramana.Radhakrishnan@arm.com>; Richard Earnshaw
> <Richard.Earnshaw@arm.com>; nickc@redhat.com
> Subject: RE: [PATCH 20/21]Arm: Add Advanced SIMD cbranch implementation
>
> Hi Tamar,
>
> > -----Original Message-----
> > From: Tamar Christina <Tamar.Christina@arm.com>
> > Sent: Thursday, January 4, 2024 11:06 AM
> > To: Tamar Christina <Tamar.Christina@arm.com>; gcc-patches@gcc.gnu.org
> > Cc: nd <nd@arm.com>; Ramana Radhakrishnan
> > <Ramana.Radhakrishnan@arm.com>; Richard Earnshaw
> > <Richard.Earnshaw@arm.com>; nickc@redhat.com; Kyrylo Tkachov
> > <Kyrylo.Tkachov@arm.com>
> > Subject: RE: [PATCH 20/21]Arm: Add Advanced SIMD cbranch implementation
> >
> > Ping,
> >
> > ---
> >
> > Hi All,
> >
> > This adds an implementation for conditional branch optab for AArch32.
> > The previous version only allowed operand 0 but it looks like cbranch
> > expansion does not check with the target and so we have to implement all.
> >
> > I therefore did not commit it. This is a larger version. I've also dropped the MVE
> > version because the mid-end can rewrite the comparison into comparing two
> > predicates without checking with the backend. Since MVE only has 1 predicate
> > register this would need to go through memory and two MRS calls. It's unlikely
> > to be beneficial and so that's for GCC 15 when I can fix the middle-end.
> >
> > The cases where AArch32 is skipped in the testsuite are all missed-optimizations
> > due to AArch32 missing some optabs.
>
> Does the testsuite have vect_* checks that can be used instead of target arm*?
> If so let's use those.
Unfortunately not, a lot of them center around handling of complex doubles.
Some tests work and some fail, which makes it hard to disable based on a
target effective test. They are things that look easy to fix so I may file some tickets
for them.
Cheers,
Tamar
> Otherwise it's okay as is.
> Thanks,
> Kyrill
>
> >
> > For e.g.
> >
> > void f1 ()
> > {
> > for (int i = 0; i < N; i++)
> > {
> > b[i] += a[i];
> > if (a[i] > 0)
> > break;
> > }
> > }
> >
> > For 128-bit vectors we generate:
> >
> > vcgt.s32 q8, q9, #0
> > vpmax.u32 d7, d16, d17
> > vpmax.u32 d7, d7, d7
> > vmov r3, s14 @ int
> > cmp r3, #0
> >
> > and of 64-bit vector we can omit one vpmax as we still need to compress to
> > 32-bits.
> >
> > Bootstrapped Regtested on arm-none-linux-gnueabihf and no issues.
> >
> > Ok for master?
> >
> > Thanks,
> > Tamar
> >
> > gcc/ChangeLog:
> >
> > * config/arm/neon.md (cbranch<mode>4): New.
> >
> > gcc/testsuite/ChangeLog:
> >
> > * gcc.dg/vect/vect-early-break_2.c: Skip Arm.
> > * gcc.dg/vect/vect-early-break_7.c: Likewise.
> > * gcc.dg/vect/vect-early-break_75.c: Likewise.
> > * gcc.dg/vect/vect-early-break_77.c: Likewise.
> > * gcc.dg/vect/vect-early-break_82.c: Likewise.
> > * gcc.dg/vect/vect-early-break_88.c: Likewise.
> > * lib/target-supports.exp (add_options_for_vect_early_break,
> > check_effective_target_vect_early_break_hw,
> > check_effective_target_vect_early_break): Support AArch32.
> > * gcc.target/arm/vect-early-break-cbranch.c: New test.
> >
> > --- inline version of patch ---
> >
> > diff --git a/gcc/config/arm/neon.md b/gcc/config/arm/neon.md
> > index
> >
> d213369ffc38fb88ad0357d848cc7da5af73bab7..ed659ab736862da416d1ff624
> 1d
> > 0d3e6c6b96ff1 100644
> > --- a/gcc/config/arm/neon.md
> > +++ b/gcc/config/arm/neon.md
> > @@ -408,6 +408,55 @@ (define_insn "vec_extract<mode><V_elem_l>"
> > [(set_attr "type" "neon_store1_one_lane<q>,neon_to_gp<q>")]
> > )
> >
> > +;; Patterns comparing two vectors and conditionally jump.
> > +;; Avdanced SIMD lacks a vector != comparison, but this is a quite common
> > +;; operation. To not pay the penalty for inverting == we can map our any
> > +;; comparisons to all i.e. any(~x) => all(x).
> > +;;
> > +;; However unlike the AArch64 version, we can't optimize this further as the
> > +;; chain is too long for combine due to these being unspecs so it doesn't fold
> > +;; the operation to something simpler.
> > +(define_expand "cbranch<mode>4"
> > + [(set (pc) (if_then_else
> > + (match_operator 0 "expandable_comparison_operator"
> > + [(match_operand:VDQI 1 "register_operand")
> > + (match_operand:VDQI 2 "reg_or_zero_operand")])
> > + (label_ref (match_operand 3 "" ""))
> > + (pc)))]
> > + "TARGET_NEON"
> > +{
> > + rtx mask = operands[1];
> > +
> > + /* If comparing against a non-zero vector we have to do a comparison first
> > + so we can have a != 0 comparison with the result. */
> > + if (operands[2] != CONST0_RTX (<MODE>mode))
> > + {
> > + mask = gen_reg_rtx (<MODE>mode);
> > + emit_insn (gen_xor<mode>3 (mask, operands[1], operands[2]));
> > + }
> > +
> > + /* For 128-bit vectors we need an additional reductions. */
> > + if (known_eq (128, GET_MODE_BITSIZE (<MODE>mode)))
> > + {
> > + /* Always reduce using a V4SI. */
> > + mask = gen_reg_rtx (V2SImode);
> > + rtx low = gen_reg_rtx (V2SImode);
> > + rtx high = gen_reg_rtx (V2SImode);
> > + rtx op1 = lowpart_subreg (V4SImode, operands[1], <MODE>mode);
> > + emit_insn (gen_neon_vget_lowv4si (low, op1));
> > + emit_insn (gen_neon_vget_highv4si (high, op1));
> > + emit_insn (gen_neon_vpumaxv2si (mask, low, high));
> > + }
> > +
> > + rtx op1 = lowpart_subreg (V2SImode, mask, GET_MODE (mask));
> > + emit_insn (gen_neon_vpumaxv2si (op1, op1, op1));
> > +
> > + rtx val = gen_reg_rtx (SImode);
> > + emit_move_insn (val, gen_lowpart (SImode, mask));
> > + emit_jump_insn (gen_cbranch_cc (operands[0], val, const0_rtx,
> operands[3]));
> > + DONE;
> > +})
> > +
> > ;; This pattern is renamed from "vec_extract<mode><V_elem_l>" to
> > ;; "neon_vec_extract<mode><V_elem_l>" and this pattern is called
> > ;; by define_expand in vec-common.md file.
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
> > b/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
> > index
> >
> 5c32bf94409e9743e72429985ab3bf13aab8f2c1..dec0b492ab883de6e02944a9
> 5f
> > d554a109a68a39 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_2.c
> > @@ -5,7 +5,7 @@
> >
> > /* { dg-additional-options "-Ofast" } */
> >
> > -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
> > +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-
> > *" } } } } */
> >
> > #include <complex.h>
> >
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
> > b/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
> > index
> >
> 8c86c5034d7522b3733543fb384a23c5d6ed0fcf..d218a0686719fee4c167684dc
> f2
> > 6402851b53260 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_7.c
> > @@ -5,7 +5,7 @@
> >
> > /* { dg-additional-options "-Ofast" } */
> >
> > -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
> > +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-
> > *" } } } } */
> >
> > #include <complex.h>
> >
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
> > b/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
> > index
> >
> ed27f8635730ff0d8803517c72693625a2feddef..9dcc3372acd657458df8d94ce
> 36
> > c4bd96f02fd52 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_75.c
> > @@ -3,7 +3,7 @@
> > /* { dg-require-effective-target vect_int } */
> >
> > /* { dg-additional-options "-O3" } */
> > -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "x86_64-*-
> > * i?86-*-*" } } } } */
> > +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "x86_64-
> > *-* i?86-*-* arm*-*-*" } } } } */
> >
> > #include <limits.h>
> > #include <assert.h>
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
> > b/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
> > index
> >
> 225106aab0a3efc7536de6f6e45bc6ff16210ea8..9fa7e6948ebfb5f1723833653f
> d
> > 6ad1fc65f4e8e 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_77.c
> > @@ -3,7 +3,7 @@
> > /* { dg-require-effective-target vect_int } */
> >
> > /* { dg-additional-options "-O3" } */
> > -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
> > +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-
> > *" } } } } */
> >
> > #include "tree-vect.h"
> >
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
> > b/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
> > index
> >
> 0e9b2d8d385c556063a3c6fcb14383317b056a79..7cd21d33485f3abb823e194
> 3c
> > 87e9481c41fd2c3 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_82.c
> > @@ -5,7 +5,7 @@
> >
> > /* { dg-additional-options "-Ofast" } */
> >
> > -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
> > +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-
> > *" } } } } */
> >
> > #include <complex.h>
> >
> > diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
> > b/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
> > index
> >
> b392dd46553994d813761da41c42989a79b90119..59ed57c5fb5f3e8197fc200
> 58
> > eeb0a81a55815cc 100644
> > --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
> > +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_88.c
> > @@ -3,7 +3,7 @@
> > /* { dg-require-effective-target vect_int } */
> >
> > /* { dg-additional-options "-Ofast --param vect-partial-vector-usage=2" } */
> > -/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
> > +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-
> > *" } } } } */
> >
> > #include "tree-vect.h"
> >
> > diff --git a/gcc/testsuite/gcc.target/arm/vect-early-break-cbranch.c
> > b/gcc/testsuite/gcc.target/arm/vect-early-break-cbranch.c
> > new file mode 100644
> > index
> >
> 0000000000000000000000000000000000000000..0e9a39d231fdf4cb565909
> 45e
> > 7cedfabd11d39b5
> > --- /dev/null
> > +++ b/gcc/testsuite/gcc.target/arm/vect-early-break-cbranch.c
> > @@ -0,0 +1,138 @@
> > +/* { dg-do compile } */
> > +/* { dg-require-effective-target vect_early_break } */
> > +/* { dg-require-effective-target arm_neon_ok } */
> > +/* { dg-require-effective-target arm32 } */
> > +/* { dg-options "-O3 -march=armv8-a+simd -mfpu=auto -mfloat-abi=hard -
> fno-
> > schedule-insns -fno-reorder-blocks -fno-schedule-insns2" } */
> > +/* { dg-final { check-function-bodies "**" "" "" } } */
> > +
> > +#define N 640
> > +int a[N] = {0};
> > +int b[N] = {0};
> > +
> > +/*
> > +** f1:
> > +** ...
> > +** vcgt.s32 q[0-9]+, q[0-9]+, #0
> > +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> > +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> > +** vmov r[0-9]+, s[0-9]+ @ int
> > +** cmp r[0-9]+, #0
> > +** bne \.L[0-9]+
> > +** ...
> > +*/
> > +void f1 ()
> > +{
> > + for (int i = 0; i < N; i++)
> > + {
> > + b[i] += a[i];
> > + if (a[i] > 0)
> > + break;
> > + }
> > +}
> > +
> > +/*
> > +** f2:
> > +** ...
> > +** vcge.s32 q[0-9]+, q[0-9]+, #0
> > +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> > +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> > +** vmov r[0-9]+, s[0-9]+ @ int
> > +** cmp r[0-9]+, #0
> > +** bne \.L[0-9]+
> > +** ...
> > +*/
> > +void f2 ()
> > +{
> > + for (int i = 0; i < N; i++)
> > + {
> > + b[i] += a[i];
> > + if (a[i] >= 0)
> > + break;
> > + }
> > +}
> > +
> > +/*
> > +** f3:
> > +** ...
> > +** vceq.i32 q[0-9]+, q[0-9]+, #0
> > +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> > +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> > +** vmov r[0-9]+, s[0-9]+ @ int
> > +** cmp r[0-9]+, #0
> > +** bne \.L[0-9]+
> > +** ...
> > +*/
> > +void f3 ()
> > +{
> > + for (int i = 0; i < N; i++)
> > + {
> > + b[i] += a[i];
> > + if (a[i] == 0)
> > + break;
> > + }
> > +}
> > +
> > +/*
> > +** f4:
> > +** ...
> > +** vceq.i32 q[0-9]+, q[0-9]+, #0
> > +** vmvn q[0-9]+, q[0-9]+
> > +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> > +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> > +** vmov r[0-9]+, s[0-9]+ @ int
> > +** cmp r[0-9]+, #0
> > +** bne \.L[0-9]+
> > +** ...
> > +*/
> > +void f4 ()
> > +{
> > + for (int i = 0; i < N; i++)
> > + {
> > + b[i] += a[i];
> > + if (a[i] != 0)
> > + break;
> > + }
> > +}
> > +
> > +/*
> > +** f5:
> > +** ...
> > +** vclt.s32 q[0-9]+, q[0-9]+, #0
> > +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> > +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> > +** vmov r[0-9]+, s[0-9]+ @ int
> > +** cmp r[0-9]+, #0
> > +** bne \.L[0-9]+
> > +** ...
> > +*/
> > +void f5 ()
> > +{
> > + for (int i = 0; i < N; i++)
> > + {
> > + b[i] += a[i];
> > + if (a[i] < 0)
> > + break;
> > + }
> > +}
> > +
> > +/*
> > +** f6:
> > +** ...
> > +** vcle.s32 q[0-9]+, q[0-9]+, #0
> > +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> > +** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
> > +** vmov r[0-9]+, s[0-9]+ @ int
> > +** cmp r[0-9]+, #0
> > +** bne \.L[0-9]+
> > +** ...
> > +*/
> > +void f6 ()
> > +{
> > + for (int i = 0; i < N; i++)
> > + {
> > + b[i] += a[i];
> > + if (a[i] <= 0)
> > + break;
> > + }
> > +}
> > +
> > diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-
> > supports.exp
> > index
> >
> 05fc417877bcd658931061b7245eb8ba5abd2e09..24a937dbb59b5723af038bd
> 9e
> > 0b89369595fcf87 100644
> > --- a/gcc/testsuite/lib/target-supports.exp
> > +++ b/gcc/testsuite/lib/target-supports.exp
> > @@ -4059,6 +4059,7 @@ proc check_effective_target_vect_early_break { } {
> > return [check_cached_effective_target_indexed vect_early_break {
> > expr {
> > [istarget aarch64*-*-*]
> > + || [check_effective_target_arm_v8_neon_ok]
> > || [check_effective_target_sse4]
> > }}]
> > }
> > @@ -4072,6 +4073,7 @@ proc check_effective_target_vect_early_break_hw { }
> > {
> > return [check_cached_effective_target_indexed vect_early_break_hw {
> > expr {
> > [istarget aarch64*-*-*]
> > + || [check_effective_target_arm_v8_neon_hw]
> > || [check_sse4_hw_available]
> > }}]
> > }
> > @@ -4081,6 +4083,11 @@ proc add_options_for_vect_early_break { flags } {
> > return "$flags"
> > }
> >
> > + if { [check_effective_target_arm_v8_neon_ok] } {
> > + global et_arm_v8_neon_flags
> > + return "$flags $et_arm_v8_neon_flags -march=armv8-a"
> > + }
> > +
> > if { [check_effective_target_sse4] } {
> > return "$flags -msse4.1"
> > }
@@ -408,6 +408,54 @@ (define_insn "vec_extract<mode><V_elem_l>"
[(set_attr "type" "neon_store1_one_lane<q>,neon_to_gp<q>")]
)
+;; Patterns comparing two vectors and conditionally jump.
+;; Avdanced SIMD lacks a vector != comparison, but this is a quite common
+;; operation. To not pay the penalty for inverting == we can map our any
+;; comparisons to all i.e. any(~x) => all(x).
+;;
+;; However unlike the AArch64 version, we can't optimize this further as the
+;; chain is too long for combine due to these being unspecs so it doesn't fold
+;; the operation to something simpler.
+(define_expand "cbranch<mode>4"
+ [(set (pc) (if_then_else
+ (match_operator 0 "expandable_comparison_operator"
+ [(match_operand:VDQI 1 "register_operand")
+ (match_operand:VDQI 2 "reg_or_zero_operand")])
+ (label_ref (match_operand 3 "" ""))
+ (pc)))]
+ "TARGET_NEON"
+{
+ rtx mask = operands[1];
+
+ /* If comparing against a non-zero vector we have to do a comparison first
+ so we can have a != 0 comparison with the result. */
+ if (operands[2] != CONST0_RTX (<MODE>mode))
+ {
+ mask = gen_reg_rtx (<MODE>mode);
+ emit_insn (gen_xor<mode>3 (mask, operands[1], operands[2]));
+ }
+
+ /* For 128-bit vectors we need an additional reductions. */
+ if (known_eq (128, GET_MODE_BITSIZE (<MODE>mode)))
+ {
+ /* Always reduce using a V4SI. */
+ mask = gen_reg_rtx (V2SImode);
+ rtx low = gen_reg_rtx (V2SImode);
+ rtx high = gen_reg_rtx (V2SImode);
+ rtx op1 = simplify_gen_subreg (V4SImode, operands[1], <MODE>mode, 0);
+ emit_insn (gen_neon_vget_lowv4si (low, op1));
+ emit_insn (gen_neon_vget_highv4si (high, op1));
+ emit_insn (gen_neon_vpumaxv2si (mask, low, high));
+ }
+
+ emit_insn (gen_neon_vpumaxv2si (mask, mask, mask));
+
+ rtx val = gen_reg_rtx (SImode);
+ emit_move_insn (val, gen_lowpart (SImode, mask));
+ emit_jump_insn (gen_cbranch_cc (operands[0], val, const0_rtx, operands[3]));
+ DONE;
+})
+
;; This pattern is renamed from "vec_extract<mode><V_elem_l>" to
;; "neon_vec_extract<mode><V_elem_l>" and this pattern is called
;; by define_expand in vec-common.md file.
@@ -5,7 +5,7 @@
/* { dg-additional-options "-Ofast" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include <complex.h>
@@ -5,7 +5,7 @@
/* { dg-additional-options "-Ofast" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include <complex.h>
@@ -3,7 +3,7 @@
/* { dg-require-effective-target vect_int } */
/* { dg-additional-options "-O3" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "x86_64-*-* i?86-*-*" } } } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "x86_64-*-* i?86-*-* arm*-*-*" } } } } */
#include <limits.h>
#include <assert.h>
@@ -3,7 +3,7 @@
/* { dg-require-effective-target vect_int } */
/* { dg-additional-options "-O3" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include "tree-vect.h"
@@ -5,7 +5,7 @@
/* { dg-additional-options "-Ofast" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include <complex.h>
@@ -3,7 +3,7 @@
/* { dg-require-effective-target vect_int } */
/* { dg-additional-options "-Ofast --param vect-partial-vector-usage=2" } */
-/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */
+/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target { ! "arm*-*-*" } } } } */
#include "tree-vect.h"
new file mode 100644
@@ -0,0 +1,138 @@
+/* { dg-do compile } */
+/* { dg-require-effective-target vect_early_break } */
+/* { dg-require-effective-target arm_neon_ok } */
+/* { dg-require-effective-target arm32 } */
+/* { dg-options "-O3 -march=armv8-a+simd -mfpu=auto -mfloat-abi=hard -fno-schedule-insns -fno-reorder-blocks -fno-schedule-insns2" } */
+/* { dg-final { check-function-bodies "**" "" "" } } */
+
+#define N 640
+int a[N] = {0};
+int b[N] = {0};
+
+/*
+** f1:
+** ...
+** vcgt.s32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f1 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] > 0)
+ break;
+ }
+}
+
+/*
+** f2:
+** ...
+** vcge.s32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f2 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] >= 0)
+ break;
+ }
+}
+
+/*
+** f3:
+** ...
+** vceq.i32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f3 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] == 0)
+ break;
+ }
+}
+
+/*
+** f4:
+** ...
+** vceq.i32 q[0-9]+, q[0-9]+, #0
+** vmvn q[0-9]+, q[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f4 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] != 0)
+ break;
+ }
+}
+
+/*
+** f5:
+** ...
+** vclt.s32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f5 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] < 0)
+ break;
+ }
+}
+
+/*
+** f6:
+** ...
+** vcle.s32 q[0-9]+, q[0-9]+, #0
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vpmax.u32 d[0-9]+, d[0-9]+, d[0-9]+
+** vmov r[0-9]+, s[0-9]+ @ int
+** cmp r[0-9]+, #0
+** bne \.L[0-9]+
+** ...
+*/
+void f6 ()
+{
+ for (int i = 0; i < N; i++)
+ {
+ b[i] += a[i];
+ if (a[i] <= 0)
+ break;
+ }
+}
+
@@ -4059,6 +4059,7 @@ proc check_effective_target_vect_early_break { } {
return [check_cached_effective_target_indexed vect_early_break {
expr {
[istarget aarch64*-*-*]
+ || [check_effective_target_arm_v8_neon_ok]
|| [check_effective_target_sse4]
}}]
}
@@ -4072,6 +4073,7 @@ proc check_effective_target_vect_early_break_hw { } {
return [check_cached_effective_target_indexed vect_early_break_hw {
expr {
[istarget aarch64*-*-*]
+ || [check_effective_target_arm_v8_neon_hw]
|| [check_sse4_hw_available]
}}]
}
@@ -4081,6 +4083,11 @@ proc add_options_for_vect_early_break { flags } {
return "$flags"
}
+ if { [check_effective_target_arm_v8_neon_ok] } {
+ global et_arm_v8_neon_flags
+ return "$flags $et_arm_v8_neon_flags -march=armv8-a"
+ }
+
if { [check_effective_target_sse4] } {
return "$flags -msse4.1"
}