File gcc7-aarch64-moutline-atomics.patch of Package cross-arm-gcc7

Overview Repositories Revisions Requests Users Attributes Meta

File gcc7-aarch64-moutline-atomics.patch of Package cross-arm-gcc7

From 56c60ff0b1ee5d15e9e1673eddda6cb450e4253c Mon Sep 17 00:00:00 2001
From: Kyrylo Tkachov <kyrylo.tkachov@arm.com>
Date: Tue, 6 Jun 2017 13:26:46 +0000
Subject: [PATCH 01/24] Allow const0_rtx operand for atomic compare-exchange
 patterns
To: gcc-patches@gcc.gnu.org

2017-06-06  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>

gcc/
	* config/aarch64/atomics.md (atomic_compare_and_swap<mode> expander):
	Use aarch64_reg_or_zero predicate for operand 4.
	(aarch64_compare_and_swap<mode> define_insn_and_split):
	Use aarch64_reg_or_zero predicate for operand 3.  Add 'Z' constraint.
	(aarch64_store_exclusive<mode>): Likewise for operand 2.

gcc/testsuite/
	* gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c: New test.

(cherry picked from commit 4ebcc903bf03705099cd4b50231dc8fe444d70b9)
---
 gcc/config/aarch64/atomics.md                        |  8 ++++----
 .../aarch64/atomic_cmp_exchange_zero_reg_1.c         | 12 ++++++++++++
 2 files changed, 16 insertions(+), 4 deletions(-)
 create mode 100644 gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c

diff --git a/gcc/config/aarch64/atomics.md b/gcc/config/aarch64/atomics.md
index 09d441075f0..27fc1933ce3 100644
--- a/gcc/config/aarch64/atomics.md
+++ b/gcc/config/aarch64/atomics.md
@@ -25,7 +25,7 @@
    (match_operand:ALLI 1 "register_operand" "")			;; val out
    (match_operand:ALLI 2 "aarch64_sync_memory_operand" "")	;; memory
    (match_operand:ALLI 3 "general_operand" "")			;; expected
-   (match_operand:ALLI 4 "register_operand" "")			;; desired
+   (match_operand:ALLI 4 "aarch64_reg_or_zero" "")			;; desired
    (match_operand:SI 5 "const_int_operand")			;; is_weak
    (match_operand:SI 6 "const_int_operand")			;; mod_s
    (match_operand:SI 7 "const_int_operand")]			;; mod_f
@@ -45,7 +45,7 @@
    (set (match_dup 1)
     (unspec_volatile:SHORT
       [(match_operand:SI 2 "aarch64_plus_operand" "rI")	;; expected
-       (match_operand:SHORT 3 "register_operand" "r")	;; desired
+       (match_operand:SHORT 3 "aarch64_reg_or_zero" "rZ")	;; desired
        (match_operand:SI 4 "const_int_operand")		;; is_weak
        (match_operand:SI 5 "const_int_operand")		;; mod_s
        (match_operand:SI 6 "const_int_operand")]	;; mod_f
@@ -69,7 +69,7 @@
    (set (match_dup 1)
     (unspec_volatile:GPI
       [(match_operand:GPI 2 "aarch64_plus_operand" "rI")	;; expect
-       (match_operand:GPI 3 "register_operand" "r")		;; desired
+       (match_operand:GPI 3 "aarch64_reg_or_zero" "rZ")		;; desired
        (match_operand:SI 4 "const_int_operand")			;; is_weak
        (match_operand:SI 5 "const_int_operand")			;; mod_s
        (match_operand:SI 6 "const_int_operand")]		;; mod_f
@@ -534,7 +534,7 @@
     (unspec_volatile:SI [(const_int 0)] UNSPECV_SX))
    (set (match_operand:ALLI 1 "aarch64_sync_memory_operand" "=Q")
     (unspec_volatile:ALLI
-      [(match_operand:ALLI 2 "register_operand" "r")
+      [(match_operand:ALLI 2 "aarch64_reg_or_zero" "rZ")
        (match_operand:SI 3 "const_int_operand")]
       UNSPECV_SX))]
   ""
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c b/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c
new file mode 100644
index 00000000000..15606b68990
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c
@@ -0,0 +1,12 @@
+/* { dg-do compile } */
+/* { dg-options "-O2" } */
+
+int
+foo (int *a)
+{
+  int x = 3;
+  return __atomic_compare_exchange_n (a, &x, 0, 1, __ATOMIC_ACQUIRE, __ATOMIC_ACQUIRE);
+}
+
+/* { dg-final { scan-assembler "stxr\\tw\[0-9\]+, wzr,.*" } } */
+/* { dg-final { scan-assembler-not "mov\\tw\[0-9\]+, 0" } } */
-- 
2.26.2

From b8e2b779d1815147073e9dcb04a1f8f9d96b1a62 Mon Sep 17 00:00:00 2001
From: Ramana Radhakrishnan <ramana.radhakrishnan@arm.com>
Date: Mon, 16 Jul 2018 09:03:48 +0000
Subject: [PATCH 02/24] Add early clobber for aarch64_store_exclusive.
To: gcc-patches@gcc.gnu.org

2018-07-16  Ramana Radhakrishnan  <ramana.radhakrishnan@arm.com>

gcc/
       * config/aarch64/atomics.md (aarch64_store_execlusive<mode>): Add
       early clobber.

(cherry picked from commit 1d896f48fa2f796ba13773ab735ef40fa3afb257)
---
 gcc/config/aarch64/atomics.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gcc/config/aarch64/atomics.md b/gcc/config/aarch64/atomics.md
index 27fc1933ce3..eb4b95c2453 100644
--- a/gcc/config/aarch64/atomics.md
+++ b/gcc/config/aarch64/atomics.md
@@ -530,7 +530,7 @@
 )
 
 (define_insn "aarch64_store_exclusive<mode>"
-  [(set (match_operand:SI 0 "register_operand" "=r")
+  [(set (match_operand:SI 0 "register_operand" "=&r")
     (unspec_volatile:SI [(const_int 0)] UNSPECV_SX))
    (set (match_operand:ALLI 1 "aarch64_sync_memory_operand" "=Q")
     (unspec_volatile:ALLI
-- 
2.26.2

From 03ffcedc78870eaf89e0f45b3d7a2e3af003e45a Mon Sep 17 00:00:00 2001
From: Richard Henderson <richard.henderson@linaro.org>
Date: Wed, 31 Oct 2018 09:29:29 +0000
Subject: [PATCH 03/24] aarch64: Simplify LSE cas generation
To: gcc-patches@gcc.gnu.org

The cas insn is a single insn, and if expanded properly need not
be split after reload.  Use the proper inputs for the insn.

2018-10-31  Richard Henderson  <richard.henderson@linaro.org>

gcc/
	* config/aarch64/aarch64.c (aarch64_expand_compare_and_swap):
	Force oldval into the rval register for TARGET_LSE; emit the compare
	during initial expansion so that it may be deleted if unused.
	(aarch64_gen_atomic_cas): Remove.
	* config/aarch64/atomics.md (@aarch64_compare_and_swap<SHORT>_lse):
	Change =&r to +r for operand 0; use match_dup for operand 2;
	remove is_weak and mod_f operands as unused.  Drop the split
	and merge with...
	(@aarch64_atomic_cas<SHORT>): ... this pattern's output; remove.
	(@aarch64_compare_and_swap<GPI>_lse): Similarly.
	(@aarch64_atomic_cas<GPI>): Similarly.

(cherry picked from commit 77f33f44baf24c22848197aa80962c003dd7b3e2)
---
 gcc/config/aarch64/aarch64-protos.h |   1 -
 gcc/config/aarch64/aarch64.c        |  62 ++++--------
 gcc/config/aarch64/atomics.md       | 143 +++++++++-------------------
 3 files changed, 63 insertions(+), 143 deletions(-)

diff --git a/gcc/config/aarch64/aarch64-protos.h b/gcc/config/aarch64/aarch64-protos.h
index 9543f8c9f29..e25f9833af4 100644
--- a/gcc/config/aarch64/aarch64-protos.h
+++ b/gcc/config/aarch64/aarch64-protos.h
@@ -425,7 +425,6 @@ rtx aarch64_load_tp (rtx);
 
 void aarch64_expand_compare_and_swap (rtx op[]);
 void aarch64_split_compare_and_swap (rtx op[]);
-void aarch64_gen_atomic_cas (rtx, rtx, rtx, rtx, rtx);
 
 bool aarch64_atomic_ldop_supported_p (enum rtx_code);
 void aarch64_gen_atomic_ldop (enum rtx_code, rtx, rtx, rtx, rtx, rtx);
diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index 05eaef3d2e6..c03d0ff9b40 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -11865,7 +11865,6 @@ aarch64_expand_compare_and_swap (rtx operands[])
   machine_mode mode, cmp_mode;
   typedef rtx (*gen_cas_fn) (rtx, rtx, rtx, rtx, rtx, rtx, rtx);
   int idx;
-  gen_cas_fn gen;
   const gen_cas_fn split_cas[] =
   {
     gen_aarch64_compare_and_swapqi,
@@ -11873,7 +11872,8 @@ aarch64_expand_compare_and_swap (rtx operands[])
     gen_aarch64_compare_and_swapsi,
     gen_aarch64_compare_and_swapdi
   };
-  const gen_cas_fn atomic_cas[] =
+  typedef rtx (*gen_lse_fn) (rtx, rtx, rtx, rtx);
+  const gen_lse_fn atomic_cas[] =
   {
     gen_aarch64_compare_and_swapqi_lse,
     gen_aarch64_compare_and_swaphi_lse,
@@ -11932,14 +11932,26 @@ aarch64_expand_compare_and_swap (rtx operands[])
       gcc_unreachable ();
     }
   if (TARGET_LSE)
-    gen = atomic_cas[idx];
+    {
+      /* The CAS insn requires oldval and rval overlap, but we need to
+	 have a copy of oldval saved across the operation to tell if
+	 the operation is successful.  */
+      if (mode == QImode || mode == HImode)
+	rval = copy_to_mode_reg (SImode, gen_lowpart (SImode, oldval));
+      else if (reg_overlap_mentioned_p (rval, oldval))
+        rval = copy_to_mode_reg (mode, oldval);
+      else
+	emit_move_insn (rval, oldval);
+      emit_insn (atomic_cas[idx] (rval, mem, newval, mod_s));
+      aarch64_gen_compare_reg (EQ, rval, oldval);
+    }
   else
-    gen = split_cas[idx];
-
-  emit_insn (gen (rval, mem, oldval, newval, is_weak, mod_s, mod_f));
+    emit_insn (split_cas[idx] (rval, mem, oldval, newval, is_weak, mod_s,
+			       mod_f));
 
   if (mode == QImode || mode == HImode)
-    emit_move_insn (operands[1], gen_lowpart (mode, rval));
+    rval = gen_lowpart (mode, rval);
+  emit_move_insn (operands[1], rval);
 
   x = gen_rtx_REG (CCmode, CC_REGNUM);
   x = gen_rtx_EQ (SImode, x, const0_rtx);
@@ -11989,42 +12001,6 @@ aarch64_emit_post_barrier (enum memmodel model)
     }
 }
 
-/* Emit an atomic compare-and-swap operation.  RVAL is the destination register
-   for the data in memory.  EXPECTED is the value expected to be in memory.
-   DESIRED is the value to store to memory.  MEM is the memory location.  MODEL
-   is the memory ordering to use.  */
-
-void
-aarch64_gen_atomic_cas (rtx rval, rtx mem,
-			rtx expected, rtx desired,
-			rtx model)
-{
-  rtx (*gen) (rtx, rtx, rtx, rtx);
-  machine_mode mode;
-
-  mode = GET_MODE (mem);
-
-  switch (mode)
-    {
-    case QImode: gen = gen_aarch64_atomic_casqi; break;
-    case HImode: gen = gen_aarch64_atomic_cashi; break;
-    case SImode: gen = gen_aarch64_atomic_cassi; break;
-    case DImode: gen = gen_aarch64_atomic_casdi; break;
-    default:
-      gcc_unreachable ();
-    }
-
-  /* Move the expected value into the CAS destination register.  */
-  emit_insn (gen_rtx_SET (rval, expected));
-
-  /* Emit the CAS.  */
-  emit_insn (gen (rval, mem, desired, model));
-
-  /* Compare the expected value with the value loaded by the CAS, to establish
-     whether the swap was made.  */
-  aarch64_gen_compare_reg (EQ, rval, expected);
-}
-
 /* Split a compare and swap pattern.  */
 
 void
diff --git a/gcc/config/aarch64/atomics.md b/gcc/config/aarch64/atomics.md
index eb4b95c2453..713aec618a2 100644
--- a/gcc/config/aarch64/atomics.md
+++ b/gcc/config/aarch64/atomics.md
@@ -85,56 +85,50 @@
   }
 )
 
-(define_insn_and_split "aarch64_compare_and_swap<mode>_lse"
-  [(set (reg:CC CC_REGNUM)					;; bool out
-    (unspec_volatile:CC [(const_int 0)] UNSPECV_ATOMIC_CMPSW))
-   (set (match_operand:SI 0 "register_operand" "=&r")		;; val out
-    (zero_extend:SI
-      (match_operand:SHORT 1 "aarch64_sync_memory_operand" "+Q"))) ;; memory
-   (set (match_dup 1)
-    (unspec_volatile:SHORT
-      [(match_operand:SI 2 "aarch64_plus_operand" "rI")	;; expected
-       (match_operand:SHORT 3 "register_operand" "r")	;; desired
-       (match_operand:SI 4 "const_int_operand")		;; is_weak
-       (match_operand:SI 5 "const_int_operand")		;; mod_s
-       (match_operand:SI 6 "const_int_operand")]	;; mod_f
-      UNSPECV_ATOMIC_CMPSW))]
-  "TARGET_LSE"
-  "#"
-  "&& reload_completed"
-  [(const_int 0)]
-  {
-    aarch64_gen_atomic_cas (operands[0], operands[1],
-			    operands[2], operands[3],
-			    operands[5]);
-    DONE;
-  }
-)
-
-(define_insn_and_split "aarch64_compare_and_swap<mode>_lse"
-  [(set (reg:CC CC_REGNUM)					;; bool out
-    (unspec_volatile:CC [(const_int 0)] UNSPECV_ATOMIC_CMPSW))
-   (set (match_operand:GPI 0 "register_operand" "=&r")		;; val out
-    (match_operand:GPI 1 "aarch64_sync_memory_operand" "+Q"))   ;; memory
-   (set (match_dup 1)
-    (unspec_volatile:GPI
-      [(match_operand:GPI 2 "aarch64_plus_operand" "rI")	;; expect
-       (match_operand:GPI 3 "register_operand" "r")		;; desired
-       (match_operand:SI 4 "const_int_operand")			;; is_weak
-       (match_operand:SI 5 "const_int_operand")			;; mod_s
-       (match_operand:SI 6 "const_int_operand")]		;; mod_f
-      UNSPECV_ATOMIC_CMPSW))]
-  "TARGET_LSE"
-  "#"
-  "&& reload_completed"
-  [(const_int 0)]
-  {
-    aarch64_gen_atomic_cas (operands[0], operands[1],
-			    operands[2], operands[3],
-			    operands[5]);
-    DONE;
-  }
-)
+(define_insn "aarch64_compare_and_swap<mode>_lse"
+  [(set (match_operand:SI 0 "register_operand" "+r")		;; val out
+     (zero_extend:SI
+     (match_operand:SHORT 1 "aarch64_sync_memory_operand" "+Q"))) ;; memory
+    (set (match_dup 1)
+     (unspec_volatile:SHORT
+      [(match_dup 0)						;; expected
+       (match_operand:SHORT 2 "aarch64_reg_or_zero" "rZ")	;; desired
+       (match_operand:SI 3 "const_int_operand")]		;; mod_s
+       UNSPECV_ATOMIC_CMPSW))]
+   "TARGET_LSE"
+{
+  enum memmodel model = memmodel_from_int (INTVAL (operands[3]));
+  if (is_mm_relaxed (model))
+    return "cas<atomic_sfx>\t%<w>0, %<w>2, %1";
+  else if (is_mm_acquire (model) || is_mm_consume (model))
+    return "casa<atomic_sfx>\t%<w>0, %<w>2, %1";
+  else if (is_mm_release (model))
+    return "casl<atomic_sfx>\t%<w>0, %<w>2, %1";
+  else
+    return "casal<atomic_sfx>\t%<w>0, %<w>2, %1";
+})
+ 
+(define_insn "aarch64_compare_and_swap<mode>_lse"
+  [(set (match_operand:GPI 0 "register_operand" "+r")		;; val out
+     (match_operand:GPI 1 "aarch64_sync_memory_operand" "+Q"))   ;; memory
+    (set (match_dup 1)
+     (unspec_volatile:GPI
+      [(match_dup 0)						;; expected
+       (match_operand:GPI 2 "aarch64_reg_or_zero" "rZ")		;; desired
+       (match_operand:SI 3 "const_int_operand")]		;; mod_s
+       UNSPECV_ATOMIC_CMPSW))]
+   "TARGET_LSE"
+{
+  enum memmodel model = memmodel_from_int (INTVAL (operands[3]));
+  if (is_mm_relaxed (model))
+    return "cas<atomic_sfx>\t%<w>0, %<w>2, %1";
+  else if (is_mm_acquire (model) || is_mm_consume (model))
+    return "casa<atomic_sfx>\t%<w>0, %<w>2, %1";
+  else if (is_mm_release (model))
+    return "casl<atomic_sfx>\t%<w>0, %<w>2, %1";
+  else
+    return "casal<atomic_sfx>\t%<w>0, %<w>2, %1";
+})
 
 (define_expand "atomic_exchange<mode>"
  [(match_operand:ALLI 0 "register_operand" "")
@@ -607,55 +601,6 @@
       return "swpal<atomic_sfx>\t%<w>2, %<w>0, %1";
   })
 
-;; Atomic compare-and-swap: HI and smaller modes.
-
-(define_insn "aarch64_atomic_cas<mode>"
- [(set (match_operand:SI 0 "register_operand" "+&r")		  ;; out
-   (zero_extend:SI
-    (match_operand:SHORT 1 "aarch64_sync_memory_operand" "+Q")))  ;; memory.
-  (set (match_dup 1)
-   (unspec_volatile:SHORT
-    [(match_dup 0)
-     (match_operand:SHORT 2 "register_operand" "r")	;; value.
-     (match_operand:SI 3 "const_int_operand" "")]	;; model.
-    UNSPECV_ATOMIC_CAS))]
- "TARGET_LSE && reload_completed"
-{
-  enum memmodel model = memmodel_from_int (INTVAL (operands[3]));
-  if (is_mm_relaxed (model))
-    return "cas<atomic_sfx>\t%<w>0, %<w>2, %1";
-  else if (is_mm_acquire (model) || is_mm_consume (model))
-    return "casa<atomic_sfx>\t%<w>0, %<w>2, %1";
-  else if (is_mm_release (model))
-    return "casl<atomic_sfx>\t%<w>0, %<w>2, %1";
-  else
-    return "casal<atomic_sfx>\t%<w>0, %<w>2, %1";
-})
-
-;; Atomic compare-and-swap: SI and larger modes.
-
-(define_insn "aarch64_atomic_cas<mode>"
- [(set (match_operand:GPI 0 "register_operand" "+&r")	      ;; out
-   (match_operand:GPI 1 "aarch64_sync_memory_operand" "+Q"))  ;; memory.
-  (set (match_dup 1)
-   (unspec_volatile:GPI
-    [(match_dup 0)
-     (match_operand:GPI 2 "register_operand" "r")	;; value.
-     (match_operand:SI 3 "const_int_operand" "")]	;; model.
-    UNSPECV_ATOMIC_CAS))]
-  "TARGET_LSE && reload_completed"
-{
-    enum memmodel model = memmodel_from_int (INTVAL (operands[3]));
-    if (is_mm_relaxed (model))
-      return "cas<atomic_sfx>\t%<w>0, %<w>2, %1";
-    else if (is_mm_acquire (model) || is_mm_consume (model))
-      return "casa<atomic_sfx>\t%<w>0, %<w>2, %1";
-    else if (is_mm_release (model))
-      return "casl<atomic_sfx>\t%<w>0, %<w>2, %1";
-    else
-      return "casal<atomic_sfx>\t%<w>0, %<w>2, %1";
-})
-
 ;; Atomic load-op: Load data, operate, store result, keep data.
 
 (define_insn "aarch64_atomic_load<atomic_ldop><mode>"
-- 
2.26.2

From 19c478bafa944772683d4fdf6a4be962b9fb5d7e Mon Sep 17 00:00:00 2001
From: Kyrylo Tkachov <kyrylo.tkachov@arm.com>
Date: Fri, 2 Jun 2017 14:41:52 +0000
Subject: [PATCH 04/24] Emit tighter strong atomic compare-exchange loop when
 comparing against zero
To: gcc-patches@gcc.gnu.org

2017-06-02  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>

gcc/
	* config/aarch64/aarch64.c (aarch64_split_compare_and_swap):
	Emit CBNZ inside loop when doing a strong exchange and comparing
	against zero.  Generate the CC flags after the loop.

gcc/testsuite/
	* gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c: New test.

(cherry picked from commit 17f47f8621099fdc4b0fc6681924f97230c8a114)
---
 gcc/config/aarch64/aarch64.c                  | 41 ++++++++++++++++---
 .../atomic_cmp_exchange_zero_strong_1.c       | 12 ++++++
 2 files changed, 47 insertions(+), 6 deletions(-)
 create mode 100644 gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c

diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index c03d0ff9b40..c2789665728 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -12024,6 +12024,17 @@ aarch64_split_compare_and_swap (rtx operands[])
   mode = GET_MODE (mem);
   model = memmodel_from_int (INTVAL (model_rtx));
 
+  /* When OLDVAL is zero and we want the strong version we can emit a tighter
+    loop:
+    .label1:
+	LD[A]XR	rval, [mem]
+	CBNZ	rval, .label2
+	ST[L]XR	scratch, newval, [mem]
+	CBNZ	scratch, .label1
+    .label2:
+	CMP	rval, 0.  */
+  bool strong_zero_p = !is_weak && oldval == const0_rtx;
+
   label1 = NULL;
   if (!is_weak)
     {
@@ -12040,11 +12051,21 @@ aarch64_split_compare_and_swap (rtx operands[])
   else
     aarch64_emit_load_exclusive (mode, rval, mem, model_rtx);
 
-  cond = aarch64_gen_compare_reg (NE, rval, oldval);
-  x = gen_rtx_NE (VOIDmode, cond, const0_rtx);
-  x = gen_rtx_IF_THEN_ELSE (VOIDmode, x,
-			    gen_rtx_LABEL_REF (Pmode, label2), pc_rtx);
-  aarch64_emit_unlikely_jump (gen_rtx_SET (pc_rtx, x));
+  if (strong_zero_p)
+    {
+      x = gen_rtx_NE (VOIDmode, rval, const0_rtx);
+      x = gen_rtx_IF_THEN_ELSE (VOIDmode, x,
+				gen_rtx_LABEL_REF (Pmode, label2), pc_rtx);
+      aarch64_emit_unlikely_jump (gen_rtx_SET (pc_rtx, x));
+    }
+  else
+    {
+      cond = aarch64_gen_compare_reg (NE, rval, oldval);
+      x = gen_rtx_NE (VOIDmode, cond, const0_rtx);
+      x = gen_rtx_IF_THEN_ELSE (VOIDmode, x,
+				 gen_rtx_LABEL_REF (Pmode, label2), pc_rtx);
+      aarch64_emit_unlikely_jump (gen_rtx_SET (pc_rtx, x));
+    }
 
   aarch64_emit_store_exclusive (mode, scratch, mem, newval, model_rtx);
 
@@ -12063,7 +12084,15 @@ aarch64_split_compare_and_swap (rtx operands[])
     }
 
   emit_label (label2);
-
+  /* If we used a CBNZ in the exchange loop emit an explicit compare with RVAL
+     to set the condition flags.  If this is not used it will be removed by
+     later passes.  */
+  if (strong_zero_p)
+    {
+      cond = gen_rtx_REG (CCmode, CC_REGNUM);
+      x = gen_rtx_COMPARE (CCmode, rval, const0_rtx);
+      emit_insn (gen_rtx_SET (cond, x));
+    }
   /* Emit any final barrier needed for a __sync operation.  */
   if (is_mm_sync (model))
     aarch64_emit_post_barrier (model);
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c b/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c
new file mode 100644
index 00000000000..b14a7c29437
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c
@@ -0,0 +1,12 @@
+/* { dg-do compile } */
+/* { dg-options "-O2" } */
+
+int
+foo (int *a)
+{
+  int x = 0;
+  return __atomic_compare_exchange_n (a, &x, 4, 0,
+				      __ATOMIC_ACQUIRE, __ATOMIC_ACQUIRE);
+}
+
+/* { dg-final { scan-assembler-times "cbnz\\tw\[0-9\]+" 2 } } */
-- 
2.26.2

From 1b90e5f91c930b124f1d4940b515a7ea64809904 Mon Sep 17 00:00:00 2001
From: Richard Henderson <richard.henderson@linaro.org>
Date: Wed, 31 Oct 2018 09:42:39 +0000
Subject: [PATCH 05/24] aarch64: Improve cas generation
To: gcc-patches@gcc.gnu.org

Do not zero-extend the input to the cas for subword operations;
instead, use the appropriate zero-extending compare insns.
Correct the predicates and constraints for immediate expected operand.

2018-10-31  Richard Henderson  <richard.henderson@linaro.org>

gcc/
	* config/aarch64/aarch64.c (aarch64_gen_compare_reg_maybe_ze): New.
	(aarch64_split_compare_and_swap): Use it.
	(aarch64_expand_compare_and_swap): Likewise.  Remove convert_modes;
	test oldval against the proper predicate.
	* config/aarch64/atomics.md (@atomic_compare_and_swap<ALLI>):
	Use nonmemory_operand for expected.
	(cas_short_expected_pred): New.
	(@aarch64_compare_and_swap<SHORT>): Use it; use "rn" not "rI" to match.
	(@aarch64_compare_and_swap<GPI>): Use "rn" not "rI" for expected.
	* config/aarch64/predicates.md (aarch64_plushi_immediate): New.
	(aarch64_plushi_operand): New.

(cherry picked from commit d400fda3a8c3330f77eb9d51874f5482d3819a9f)
---
 gcc/config/aarch64/aarch64.c     | 97 ++++++++++++++++++++------------
 gcc/config/aarch64/atomics.md    | 19 ++++---
 gcc/config/aarch64/predicates.md | 12 ++++
 3 files changed, 84 insertions(+), 44 deletions(-)

diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index c2789665728..afde9e291f9 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -1198,6 +1198,33 @@ aarch64_gen_compare_reg (RTX_CODE code, rtx x, rtx y)
   return cc_reg;
 }
 
+/* Similarly, but maybe zero-extend Y if Y_MODE < SImode.  */
+
+static rtx
+aarch64_gen_compare_reg_maybe_ze (RTX_CODE code, rtx x, rtx y,
+                                  machine_mode y_mode)
+{
+  if (y_mode == QImode || y_mode == HImode)
+    {
+      if (CONST_INT_P (y))
+	y = GEN_INT (INTVAL (y) & GET_MODE_MASK (y_mode));
+      else
+	{
+	  rtx t, cc_reg;
+	  machine_mode cc_mode;
+
+	  t = gen_rtx_ZERO_EXTEND (SImode, y);
+	  t = gen_rtx_COMPARE (CC_SWPmode, t, x);
+	  cc_mode = CC_SWPmode;
+	  cc_reg = gen_rtx_REG (cc_mode, CC_REGNUM);
+	  emit_set_insn (cc_reg, t);
+	  return cc_reg;
+	}
+    }
+
+  return aarch64_gen_compare_reg (code, x, y);
+}
+
 /* Build the SYMBOL_REF for __tls_get_addr.  */
 
 static GTY(()) rtx tls_get_addr_libfunc;
@@ -11861,8 +11888,8 @@ aarch64_emit_unlikely_jump (rtx insn)
 void
 aarch64_expand_compare_and_swap (rtx operands[])
 {
-  rtx bval, rval, mem, oldval, newval, is_weak, mod_s, mod_f, x;
-  machine_mode mode, cmp_mode;
+  rtx bval, rval, mem, oldval, newval, is_weak, mod_s, mod_f, x, cc_reg;
+  machine_mode mode, r_mode;
   typedef rtx (*gen_cas_fn) (rtx, rtx, rtx, rtx, rtx, rtx, rtx);
   int idx;
   const gen_cas_fn split_cas[] =
@@ -11890,36 +11917,19 @@ aarch64_expand_compare_and_swap (rtx operands[])
   mod_s = operands[6];
   mod_f = operands[7];
   mode = GET_MODE (mem);
-  cmp_mode = mode;
 
   /* Normally the succ memory model must be stronger than fail, but in the
      unlikely event of fail being ACQUIRE and succ being RELEASE we need to
      promote succ to ACQ_REL so that we don't lose the acquire semantics.  */
-
   if (is_mm_acquire (memmodel_from_int (INTVAL (mod_f)))
       && is_mm_release (memmodel_from_int (INTVAL (mod_s))))
     mod_s = GEN_INT (MEMMODEL_ACQ_REL);
 
-  switch (mode)
+  r_mode = mode;
+  if (mode == QImode || mode == HImode)
     {
-    case QImode:
-    case HImode:
-      /* For short modes, we're going to perform the comparison in SImode,
-	 so do the zero-extension now.  */
-      cmp_mode = SImode;
-      rval = gen_reg_rtx (SImode);
-      oldval = convert_modes (SImode, mode, oldval, true);
-      /* Fall through.  */
-
-    case SImode:
-    case DImode:
-      /* Force the value into a register if needed.  */
-      if (!aarch64_plus_operand (oldval, mode))
-	oldval = force_reg (cmp_mode, oldval);
-      break;
-
-    default:
-      gcc_unreachable ();
+      r_mode = SImode;
+      rval = gen_reg_rtx (r_mode);
     }
 
   switch (mode)
@@ -11936,25 +11946,40 @@ aarch64_expand_compare_and_swap (rtx operands[])
       /* The CAS insn requires oldval and rval overlap, but we need to
 	 have a copy of oldval saved across the operation to tell if
 	 the operation is successful.  */
-      if (mode == QImode || mode == HImode)
-	rval = copy_to_mode_reg (SImode, gen_lowpart (SImode, oldval));
-      else if (reg_overlap_mentioned_p (rval, oldval))
-        rval = copy_to_mode_reg (mode, oldval);
+      if (reg_overlap_mentioned_p (rval, oldval))
+        rval = copy_to_mode_reg (r_mode, oldval);
       else
-	emit_move_insn (rval, oldval);
+	emit_move_insn (rval, gen_lowpart (r_mode, oldval));
+
       emit_insn (atomic_cas[idx] (rval, mem, newval, mod_s));
-      aarch64_gen_compare_reg (EQ, rval, oldval);
+      cc_reg = aarch64_gen_compare_reg_maybe_ze (NE, rval, oldval, mode);
     }
   else
-    emit_insn (split_cas[idx] (rval, mem, oldval, newval, is_weak, mod_s,
-			       mod_f));
+    {
+      /* The oldval predicate varies by mode.  Test it and force to reg.  */
+      insn_code code;
+      switch (mode)
+	{
+	case QImode: code = CODE_FOR_aarch64_compare_and_swapqi; break;
+	case HImode: code = CODE_FOR_aarch64_compare_and_swaphi; break;
+	case SImode: code = CODE_FOR_aarch64_compare_and_swapsi; break;
+	case DImode: code = CODE_FOR_aarch64_compare_and_swapdi; break;
+	default:
+	  gcc_unreachable ();
+	}
+      if (!insn_data[code].operand[2].predicate (oldval, mode))
+	oldval = force_reg (mode, oldval);
 
-  if (mode == QImode || mode == HImode)
+      emit_insn (split_cas[idx] (rval, mem, oldval, newval, is_weak, mod_s,
+				 mod_f));
+      cc_reg = gen_rtx_REG (CCmode, CC_REGNUM);
+    }
+
+  if (r_mode != mode)
     rval = gen_lowpart (mode, rval);
   emit_move_insn (operands[1], rval);
 
-  x = gen_rtx_REG (CCmode, CC_REGNUM);
-  x = gen_rtx_EQ (SImode, x, const0_rtx);
+  x = gen_rtx_EQ (SImode, cc_reg, const0_rtx);
   emit_insn (gen_rtx_SET (bval, x));
 }
 
@@ -12060,10 +12085,10 @@ aarch64_split_compare_and_swap (rtx operands[])
     }
   else
     {
-      cond = aarch64_gen_compare_reg (NE, rval, oldval);
+      cond = aarch64_gen_compare_reg_maybe_ze (NE, rval, oldval, mode);
       x = gen_rtx_NE (VOIDmode, cond, const0_rtx);
       x = gen_rtx_IF_THEN_ELSE (VOIDmode, x,
-				 gen_rtx_LABEL_REF (Pmode, label2), pc_rtx);
+				gen_rtx_LABEL_REF (Pmode, label2), pc_rtx);
       aarch64_emit_unlikely_jump (gen_rtx_SET (pc_rtx, x));
     }
 
diff --git a/gcc/config/aarch64/atomics.md b/gcc/config/aarch64/atomics.md
index 713aec618a2..577000fa6a4 100644
--- a/gcc/config/aarch64/atomics.md
+++ b/gcc/config/aarch64/atomics.md
@@ -24,8 +24,8 @@
   [(match_operand:SI 0 "register_operand" "")			;; bool out
    (match_operand:ALLI 1 "register_operand" "")			;; val out
    (match_operand:ALLI 2 "aarch64_sync_memory_operand" "")	;; memory
-   (match_operand:ALLI 3 "general_operand" "")			;; expected
-   (match_operand:ALLI 4 "aarch64_reg_or_zero" "")			;; desired
+   (match_operand:ALLI 3 "nonmemory_operand" "")		;; expected
+   (match_operand:ALLI 4 "aarch64_reg_or_zero" "")		;; desired
    (match_operand:SI 5 "const_int_operand")			;; is_weak
    (match_operand:SI 6 "const_int_operand")			;; mod_s
    (match_operand:SI 7 "const_int_operand")]			;; mod_f
@@ -36,19 +36,22 @@
   }
 )
 
+(define_mode_attr cas_short_expected_pred
+  [(QI "aarch64_reg_or_imm") (HI "aarch64_plushi_operand")])
+
 (define_insn_and_split "aarch64_compare_and_swap<mode>"
   [(set (reg:CC CC_REGNUM)					;; bool out
     (unspec_volatile:CC [(const_int 0)] UNSPECV_ATOMIC_CMPSW))
-   (set (match_operand:SI 0 "register_operand" "=&r")	   ;; val out
+   (set (match_operand:SI 0 "register_operand" "=&r")		;; val out
     (zero_extend:SI
       (match_operand:SHORT 1 "aarch64_sync_memory_operand" "+Q"))) ;; memory
    (set (match_dup 1)
     (unspec_volatile:SHORT
-      [(match_operand:SI 2 "aarch64_plus_operand" "rI")	;; expected
+      [(match_operand:SHORT 2 "<cas_short_expected_pred>" "rn")	;; expected
        (match_operand:SHORT 3 "aarch64_reg_or_zero" "rZ")	;; desired
-       (match_operand:SI 4 "const_int_operand")		;; is_weak
-       (match_operand:SI 5 "const_int_operand")		;; mod_s
-       (match_operand:SI 6 "const_int_operand")]	;; mod_f
+       (match_operand:SI 4 "const_int_operand")			;; is_weak
+       (match_operand:SI 5 "const_int_operand")			;; mod_s
+       (match_operand:SI 6 "const_int_operand")]		;; mod_f
       UNSPECV_ATOMIC_CMPSW))
    (clobber (match_scratch:SI 7 "=&r"))]
   ""
@@ -68,7 +71,7 @@
     (match_operand:GPI 1 "aarch64_sync_memory_operand" "+Q"))   ;; memory
    (set (match_dup 1)
     (unspec_volatile:GPI
-      [(match_operand:GPI 2 "aarch64_plus_operand" "rI")	;; expect
+      [(match_operand:GPI 2 "aarch64_plus_operand" "rn")	;; expect
        (match_operand:GPI 3 "aarch64_reg_or_zero" "rZ")		;; desired
        (match_operand:SI 4 "const_int_operand")			;; is_weak
        (match_operand:SI 5 "const_int_operand")			;; mod_s
diff --git a/gcc/config/aarch64/predicates.md b/gcc/config/aarch64/predicates.md
index e83d45b3945..20429e5d04c 100644
--- a/gcc/config/aarch64/predicates.md
+++ b/gcc/config/aarch64/predicates.md
@@ -86,6 +86,18 @@
   (ior (match_operand 0 "register_operand")
        (match_operand 0 "aarch64_plus_immediate")))
 
+(define_predicate "aarch64_plushi_immediate"
+  (match_code "const_int")
+{
+  HOST_WIDE_INT val = INTVAL (op);
+  /* The HImode value must be zero-extendable to an SImode plus_operand.  */
+  return ((val & 0xfff) == val || sext_hwi (val & 0xf000, 16) == val);
+})
+
+(define_predicate "aarch64_plushi_operand"
+  (ior (match_operand 0 "register_operand")
+       (match_operand 0 "aarch64_plushi_immediate")))
+
 (define_predicate "aarch64_pluslong_immediate"
   (and (match_code "const_int")
        (match_test "(INTVAL (op) < 0xffffff && INTVAL (op) > -0xffffff)")))
-- 
2.26.2

From 585818b9697910d3c136db9805f129f4d735e28d Mon Sep 17 00:00:00 2001
From: Richard Henderson <richard.henderson@linaro.org>
Date: Wed, 31 Oct 2018 09:47:21 +0000
Subject: [PATCH 06/24] aarch64: Improve swp generation
To: gcc-patches@gcc.gnu.org

Allow zero as an input; fix constraints; avoid unnecessary split.

2018-10-31  Richard Henderson  <richard.henderson@linaro.org>

gcc/
	* config/aarch64/aarch64.c (aarch64_emit_atomic_swap): Remove.
	(aarch64_gen_atomic_ldop): Don't call it.
	* config/aarch64/atomics.md (atomic_exchange<ALLI>):
	Use aarch64_reg_or_zero.
	(aarch64_atomic_exchange<ALLI>): Likewise.
	(aarch64_atomic_exchange<ALLI>_lse): Remove split; remove & from
	operand 0; use aarch64_reg_or_zero for input; merge ...
	(@aarch64_atomic_swp<ALLI>): ... this and remove.

(cherry picked from commit 8f5603d363a4e0453d2c38c7103aeb0bdca85c4e)
---
 gcc/config/aarch64/aarch64.c  | 25 ------------------
 gcc/config/aarch64/atomics.md | 49 +++++++++++------------------------
 2 files changed, 15 insertions(+), 59 deletions(-)

diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index afde9e291f9..d08af9d63ca 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -12142,27 +12142,6 @@ aarch64_emit_bic (machine_mode mode, rtx dst, rtx s1, rtx s2, int shift)
   emit_insn (gen (dst, s2, shift_rtx, s1));
 }
 
-/* Emit an atomic swap.  */
-
-static void
-aarch64_emit_atomic_swap (machine_mode mode, rtx dst, rtx value,
-			  rtx mem, rtx model)
-{
-  rtx (*gen) (rtx, rtx, rtx, rtx);
-
-  switch (mode)
-    {
-    case QImode: gen = gen_aarch64_atomic_swpqi; break;
-    case HImode: gen = gen_aarch64_atomic_swphi; break;
-    case SImode: gen = gen_aarch64_atomic_swpsi; break;
-    case DImode: gen = gen_aarch64_atomic_swpdi; break;
-    default:
-      gcc_unreachable ();
-    }
-
-  emit_insn (gen (dst, mem, value, model));
-}
-
 /* Operations supported by aarch64_emit_atomic_load_op.  */
 
 enum aarch64_atomic_load_op_code
@@ -12275,10 +12254,6 @@ aarch64_gen_atomic_ldop (enum rtx_code code, rtx out_data, rtx out_result,
      a SET then emit a swap instruction and finish.  */
   switch (code)
     {
-    case SET:
-      aarch64_emit_atomic_swap (mode, out_data, src, mem, model_rtx);
-      return;
-
     case MINUS:
       /* Negate the value and treat it as a PLUS.  */
       {
diff --git a/gcc/config/aarch64/atomics.md b/gcc/config/aarch64/atomics.md
index 577000fa6a4..f1cc972bae4 100644
--- a/gcc/config/aarch64/atomics.md
+++ b/gcc/config/aarch64/atomics.md
@@ -136,7 +136,7 @@
 (define_expand "atomic_exchange<mode>"
  [(match_operand:ALLI 0 "register_operand" "")
   (match_operand:ALLI 1 "aarch64_sync_memory_operand" "")
-  (match_operand:ALLI 2 "register_operand" "")
+  (match_operand:ALLI 2 "aarch64_reg_or_zero" "")
   (match_operand:SI 3 "const_int_operand" "")]
   ""
   {
@@ -156,10 +156,10 @@
 
 (define_insn_and_split "aarch64_atomic_exchange<mode>"
   [(set (match_operand:ALLI 0 "register_operand" "=&r")		;; output
-    (match_operand:ALLI 1 "aarch64_sync_memory_operand" "+Q")) ;; memory
+    (match_operand:ALLI 1 "aarch64_sync_memory_operand" "+Q"))	;; memory
    (set (match_dup 1)
     (unspec_volatile:ALLI
-      [(match_operand:ALLI 2 "register_operand" "r")	;; input
+      [(match_operand:ALLI 2 "aarch64_reg_or_zero" "rZ")	;; input
        (match_operand:SI 3 "const_int_operand" "")]		;; model
       UNSPECV_ATOMIC_EXCHG))
    (clobber (reg:CC CC_REGNUM))
@@ -175,22 +175,25 @@
   }
 )
 
-(define_insn_and_split "aarch64_atomic_exchange<mode>_lse"
-  [(set (match_operand:ALLI 0 "register_operand" "=&r")
+(define_insn "aarch64_atomic_exchange<mode>_lse"
+  [(set (match_operand:ALLI 0 "register_operand" "=r")
     (match_operand:ALLI 1 "aarch64_sync_memory_operand" "+Q"))
    (set (match_dup 1)
     (unspec_volatile:ALLI
-      [(match_operand:ALLI 2 "register_operand" "r")
+      [(match_operand:ALLI 2 "aarch64_reg_or_zero" "rZ")
        (match_operand:SI 3 "const_int_operand" "")]
       UNSPECV_ATOMIC_EXCHG))]
   "TARGET_LSE"
-  "#"
-  "&& reload_completed"
-  [(const_int 0)]
   {
-    aarch64_gen_atomic_ldop (SET, operands[0], NULL, operands[1],
-			     operands[2], operands[3]);
-    DONE;
+    enum memmodel model = memmodel_from_int (INTVAL (operands[3]));
+    if (is_mm_relaxed (model))
+      return "swp<atomic_sfx>\t%<w>2, %<w>0, %1";
+    else if (is_mm_acquire (model) || is_mm_consume (model))
+      return "swpa<atomic_sfx>\t%<w>2, %<w>0, %1";
+    else if (is_mm_release (model))
+      return "swpl<atomic_sfx>\t%<w>2, %<w>0, %1";
+    else
+      return "swpal<atomic_sfx>\t%<w>2, %<w>0, %1";
   }
 )
 
@@ -582,28 +585,6 @@
 
 ;; ARMv8.1-A LSE instructions.
 
-;; Atomic swap with memory.
-(define_insn "aarch64_atomic_swp<mode>"
- [(set (match_operand:ALLI 0 "register_operand" "+&r")
-   (match_operand:ALLI 1 "aarch64_sync_memory_operand" "+Q"))
-  (set (match_dup 1)
-   (unspec_volatile:ALLI
-    [(match_operand:ALLI 2 "register_operand" "r")
-     (match_operand:SI 3 "const_int_operand" "")]
-    UNSPECV_ATOMIC_SWP))]
-  "TARGET_LSE && reload_completed"
-  {
-    enum memmodel model = memmodel_from_int (INTVAL (operands[3]));
-    if (is_mm_relaxed (model))
-      return "swp<atomic_sfx>\t%<w>2, %<w>0, %1";
-    else if (is_mm_acquire (model) || is_mm_consume (model))
-      return "swpa<atomic_sfx>\t%<w>2, %<w>0, %1";
-    else if (is_mm_release (model))
-      return "swpl<atomic_sfx>\t%<w>2, %<w>0, %1";
-    else
-      return "swpal<atomic_sfx>\t%<w>2, %<w>0, %1";
-  })
-
 ;; Atomic load-op: Load data, operate, store result, keep data.
 
 (define_insn "aarch64_atomic_load<atomic_ldop><mode>"
-- 
2.26.2

From 2d9e1f5e434f9511ca82149bbf9b0d64acacac64 Mon Sep 17 00:00:00 2001
From: Richard Henderson <richard.henderson@linaro.org>
Date: Wed, 31 Oct 2018 09:58:48 +0000
Subject: [PATCH 07/24] aarch64: Improve atomic-op lse generation
To: gcc-patches@gcc.gnu.org

Fix constraints; avoid unnecessary split.  Drop the use of the atomic_op
iterator in favor of the ATOMIC_LDOP iterator; this is simplier and more
logical for ldclr aka bic.

2018-10-31  Richard Henderson  <richard.henderson@linaro.org>

gcc/
	* config/aarch64/aarch64.c (aarch64_emit_bic): Remove.
	(aarch64_atomic_ldop_supported_p): Remove.
	(aarch64_gen_atomic_ldop): Remove.
	* config/aarch64/atomic.md (atomic_<atomic_optab><ALLI>):
	Fully expand LSE operations here.
	(atomic_fetch_<atomic_optab><ALLI>): Likewise.
	(atomic_<atomic_optab>_fetch<ALLI>): Likewise.
	(aarch64_atomic_<ATOMIC_LDOP><ALLI>_lse): Drop atomic_op iterator
	and use ATOMIC_LDOP instead; use register_operand for the input;
	drop the split and emit insns directly.
	(aarch64_atomic_fetch_<ATOMIC_LDOP><ALLI>_lse): Likewise.
	(aarch64_atomic_<atomic_op>_fetch<ALLI>_lse): Remove.
	(@aarch64_atomic_load<ATOMIC_LDOP><ALLI>): Remove.

(cherry picked from commit 7803ec5ee2a547043fb6708a08ddb1361ba91202)
---
 gcc/config/aarch64/aarch64-protos.h |   2 -
 gcc/config/aarch64/aarch64.c        | 247 ----------------------------
 gcc/config/aarch64/atomics.md       | 197 +++++++++++-----------
 gcc/config/aarch64/iterators.md     |   5 +-
 4 files changed, 108 insertions(+), 343 deletions(-)

diff --git a/gcc/config/aarch64/aarch64-protos.h b/gcc/config/aarch64/aarch64-protos.h
index e25f9833af4..e47f2174479 100644
--- a/gcc/config/aarch64/aarch64-protos.h
+++ b/gcc/config/aarch64/aarch64-protos.h
@@ -426,8 +426,6 @@ rtx aarch64_load_tp (rtx);
 void aarch64_expand_compare_and_swap (rtx op[]);
 void aarch64_split_compare_and_swap (rtx op[]);
 
-bool aarch64_atomic_ldop_supported_p (enum rtx_code);
-void aarch64_gen_atomic_ldop (enum rtx_code, rtx, rtx, rtx, rtx, rtx);
 void aarch64_split_atomic_op (enum rtx_code, rtx, rtx, rtx, rtx, rtx, rtx);
 
 bool aarch64_gen_adjusted_ldpstp (rtx *, bool, enum machine_mode, RTX_CODE);
diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index d08af9d63ca..ed3cec30859 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -11983,32 +11983,6 @@ aarch64_expand_compare_and_swap (rtx operands[])
   emit_insn (gen_rtx_SET (bval, x));
 }
 
-/* Test whether the target supports using a atomic load-operate instruction.
-   CODE is the operation and AFTER is TRUE if the data in memory after the
-   operation should be returned and FALSE if the data before the operation
-   should be returned.  Returns FALSE if the operation isn't supported by the
-   architecture.  */
-
-bool
-aarch64_atomic_ldop_supported_p (enum rtx_code code)
-{
-  if (!TARGET_LSE)
-    return false;
-
-  switch (code)
-    {
-    case SET:
-    case AND:
-    case IOR:
-    case XOR:
-    case MINUS:
-    case PLUS:
-      return true;
-    default:
-      return false;
-    }
-}
-
 /* Emit a barrier, that is appropriate for memory model MODEL, at the end of a
    sequence implementing an atomic operation.  */
 
@@ -12123,227 +12097,6 @@ aarch64_split_compare_and_swap (rtx operands[])
     aarch64_emit_post_barrier (model);
 }
 
-/* Emit a BIC instruction.  */
-
-static void
-aarch64_emit_bic (machine_mode mode, rtx dst, rtx s1, rtx s2, int shift)
-{
-  rtx shift_rtx = GEN_INT (shift);
-  rtx (*gen) (rtx, rtx, rtx, rtx);
-
-  switch (mode)
-    {
-    case SImode: gen = gen_and_one_cmpl_lshrsi3; break;
-    case DImode: gen = gen_and_one_cmpl_lshrdi3; break;
-    default:
-      gcc_unreachable ();
-    }
-
-  emit_insn (gen (dst, s2, shift_rtx, s1));
-}
-
-/* Operations supported by aarch64_emit_atomic_load_op.  */
-
-enum aarch64_atomic_load_op_code
-{
-  AARCH64_LDOP_PLUS,	/* A + B  */
-  AARCH64_LDOP_XOR,	/* A ^ B  */
-  AARCH64_LDOP_OR,	/* A | B  */
-  AARCH64_LDOP_BIC	/* A & ~B  */
-};
-
-/* Emit an atomic load-operate.  */
-
-static void
-aarch64_emit_atomic_load_op (enum aarch64_atomic_load_op_code code,
-			     machine_mode mode, rtx dst, rtx src,
-			     rtx mem, rtx model)
-{
-  typedef rtx (*aarch64_atomic_load_op_fn) (rtx, rtx, rtx, rtx);
-  const aarch64_atomic_load_op_fn plus[] =
-  {
-    gen_aarch64_atomic_loadaddqi,
-    gen_aarch64_atomic_loadaddhi,
-    gen_aarch64_atomic_loadaddsi,
-    gen_aarch64_atomic_loadadddi
-  };
-  const aarch64_atomic_load_op_fn eor[] =
-  {
-    gen_aarch64_atomic_loadeorqi,
-    gen_aarch64_atomic_loadeorhi,
-    gen_aarch64_atomic_loadeorsi,
-    gen_aarch64_atomic_loadeordi
-  };
-  const aarch64_atomic_load_op_fn ior[] =
-  {
-    gen_aarch64_atomic_loadsetqi,
-    gen_aarch64_atomic_loadsethi,
-    gen_aarch64_atomic_loadsetsi,
-    gen_aarch64_atomic_loadsetdi
-  };
-  const aarch64_atomic_load_op_fn bic[] =
-  {
-    gen_aarch64_atomic_loadclrqi,
-    gen_aarch64_atomic_loadclrhi,
-    gen_aarch64_atomic_loadclrsi,
-    gen_aarch64_atomic_loadclrdi
-  };
-  aarch64_atomic_load_op_fn gen;
-  int idx = 0;
-
-  switch (mode)
-    {
-    case QImode: idx = 0; break;
-    case HImode: idx = 1; break;
-    case SImode: idx = 2; break;
-    case DImode: idx = 3; break;
-    default:
-      gcc_unreachable ();
-    }
-
-  switch (code)
-    {
-    case AARCH64_LDOP_PLUS: gen = plus[idx]; break;
-    case AARCH64_LDOP_XOR: gen = eor[idx]; break;
-    case AARCH64_LDOP_OR: gen = ior[idx]; break;
-    case AARCH64_LDOP_BIC: gen = bic[idx]; break;
-    default:
-      gcc_unreachable ();
-    }
-
-  emit_insn (gen (dst, mem, src, model));
-}
-
-/* Emit an atomic load+operate.  CODE is the operation.  OUT_DATA is the
-   location to store the data read from memory.  OUT_RESULT is the location to
-   store the result of the operation.  MEM is the memory location to read and
-   modify.  MODEL_RTX is the memory ordering to use.  VALUE is the second
-   operand for the operation.  Either OUT_DATA or OUT_RESULT, but not both, can
-   be NULL.  */
-
-void
-aarch64_gen_atomic_ldop (enum rtx_code code, rtx out_data, rtx out_result,
-			 rtx mem, rtx value, rtx model_rtx)
-{
-  machine_mode mode = GET_MODE (mem);
-  machine_mode wmode = (mode == DImode ? DImode : SImode);
-  const bool short_mode = (mode < SImode);
-  aarch64_atomic_load_op_code ldop_code;
-  rtx src;
-  rtx x;
-
-  if (out_data)
-    out_data = gen_lowpart (mode, out_data);
-
-  if (out_result)
-    out_result = gen_lowpart (mode, out_result);
-
-  /* Make sure the value is in a register, putting it into a destination
-     register if it needs to be manipulated.  */
-  if (!register_operand (value, mode)
-      || code == AND || code == MINUS)
-    {
-      src = out_result ? out_result : out_data;
-      emit_move_insn (src, gen_lowpart (mode, value));
-    }
-  else
-    src = value;
-  gcc_assert (register_operand (src, mode));
-
-  /* Preprocess the data for the operation as necessary.  If the operation is
-     a SET then emit a swap instruction and finish.  */
-  switch (code)
-    {
-    case MINUS:
-      /* Negate the value and treat it as a PLUS.  */
-      {
-	rtx neg_src;
-
-	/* Resize the value if necessary.  */
-	if (short_mode)
-	  src = gen_lowpart (wmode, src);
-
-	neg_src = gen_rtx_NEG (wmode, src);
-	emit_insn (gen_rtx_SET (src, neg_src));
-
-	if (short_mode)
-	  src = gen_lowpart (mode, src);
-      }
-      /* Fall-through.  */
-    case PLUS:
-      ldop_code = AARCH64_LDOP_PLUS;
-      break;
-
-    case IOR:
-      ldop_code = AARCH64_LDOP_OR;
-      break;
-
-    case XOR:
-      ldop_code = AARCH64_LDOP_XOR;
-      break;
-
-    case AND:
-      {
-	rtx not_src;
-
-	/* Resize the value if necessary.  */
-	if (short_mode)
-	  src = gen_lowpart (wmode, src);
-
-	not_src = gen_rtx_NOT (wmode, src);
-	emit_insn (gen_rtx_SET (src, not_src));
-
-	if (short_mode)
-	  src = gen_lowpart (mode, src);
-      }
-      ldop_code = AARCH64_LDOP_BIC;
-      break;
-
-    default:
-      /* The operation can't be done with atomic instructions.  */
-      gcc_unreachable ();
-    }
-
-  aarch64_emit_atomic_load_op (ldop_code, mode, out_data, src, mem, model_rtx);
-
-  /* If necessary, calculate the data in memory after the update by redoing the
-     operation from values in registers.  */
-  if (!out_result)
-    return;
-
-  if (short_mode)
-    {
-      src = gen_lowpart (wmode, src);
-      out_data = gen_lowpart (wmode, out_data);
-      out_result = gen_lowpart (wmode, out_result);
-    }
-
-  x = NULL_RTX;
-
-  switch (code)
-    {
-    case MINUS:
-    case PLUS:
-      x = gen_rtx_PLUS (wmode, out_data, src);
-      break;
-    case IOR:
-      x = gen_rtx_IOR (wmode, out_data, src);
-      break;
-    case XOR:
-      x = gen_rtx_XOR (wmode, out_data, src);
-      break;
-    case AND:
-      aarch64_emit_bic (wmode, out_result, out_data, src, 0);
-      return;
-    default:
-      gcc_unreachable ();
-    }
-
-  emit_set_insn (out_result, x);
-
-  return;
-}
-
 /* Split an atomic operation.  */
 
 void
diff --git a/gcc/config/aarch64/atomics.md b/gcc/config/aarch64/atomics.md
index f1cc972bae4..735407c9fd7 100644
--- a/gcc/config/aarch64/atomics.md
+++ b/gcc/config/aarch64/atomics.md
@@ -207,13 +207,37 @@
     rtx (*gen) (rtx, rtx, rtx);
 
     /* Use an atomic load-operate instruction when possible.  */
-    if (aarch64_atomic_ldop_supported_p (<CODE>))
-      gen = gen_aarch64_atomic_<atomic_optab><mode>_lse;
+    if (TARGET_LSE)
+      {
+	switch (<CODE>)
+	  {
+	  case MINUS:
+	    operands[1] = expand_simple_unop (<MODE>mode, NEG, operands[1],
+					      NULL, 1);
+	    /* fallthru */
+	  case PLUS:
+	    gen = gen_aarch64_atomic_add<mode>_lse;
+	    break;
+	  case IOR:
+	    gen = gen_aarch64_atomic_ior<mode>_lse;
+	    break;
+	  case XOR:
+	    gen = gen_aarch64_atomic_xor<mode>_lse;
+	    break;
+	  case AND:
+	    operands[1] = expand_simple_unop (<MODE>mode, NOT, operands[1],
+					      NULL, 1);
+	    gen = gen_aarch64_atomic_bic<mode>_lse;
+	    break;
+	  default:
+	    gcc_unreachable ();
+	  }
+	operands[1] = force_reg (<MODE>mode, operands[1]);
+      }
     else
       gen = gen_aarch64_atomic_<atomic_optab><mode>;
 
     emit_insn (gen (operands[0], operands[1], operands[2]));
-
     DONE;
   }
 )
@@ -239,22 +263,25 @@
   }
 )
 
-(define_insn_and_split "aarch64_atomic_<atomic_optab><mode>_lse"
+(define_insn "aarch64_atomic_<atomic_ldoptab><mode>_lse"
   [(set (match_operand:ALLI 0 "aarch64_sync_memory_operand" "+Q")
-    (unspec_volatile:ALLI
-      [(atomic_op:ALLI (match_dup 0)
-	(match_operand:ALLI 1 "<atomic_op_operand>" "r<const_atomic>"))
-       (match_operand:SI 2 "const_int_operand")]
-      UNSPECV_ATOMIC_OP))
+	(unspec_volatile:ALLI
+	  [(match_dup 0)
+	   (match_operand:ALLI 1 "register_operand" "r")
+	   (match_operand:SI 2 "const_int_operand")]
+      ATOMIC_LDOP))
    (clobber (match_scratch:ALLI 3 "=&r"))]
   "TARGET_LSE"
-  "#"
-  "&& reload_completed"
-  [(const_int 0)]
   {
-    aarch64_gen_atomic_ldop (<CODE>, operands[3], NULL, operands[0],
-			     operands[1], operands[2]);
-    DONE;
+   enum memmodel model = memmodel_from_int (INTVAL (operands[2]));
+   if (is_mm_relaxed (model))
+     return "ld<atomic_ldop><atomic_sfx>\t%<w>1, %<w>3, %0";
+   else if (is_mm_release (model))
+     return "ld<atomic_ldop>l<atomic_sfx>\t%<w>1, %<w>3, %0";
+   else if (is_mm_acquire (model) || is_mm_consume (model))
+     return "ld<atomic_ldop>a<atomic_sfx>\t%<w>1, %<w>3, %0";
+   else
+     return "ld<atomic_ldop>al<atomic_sfx>\t%<w>1, %<w>3, %0";
   }
 )
 
@@ -280,7 +307,7 @@
   }
 )
 
-;; Load-operate-store, returning the updated memory data.
+;; Load-operate-store, returning the original memory data.
 
 (define_expand "atomic_fetch_<atomic_optab><mode>"
  [(match_operand:ALLI 0 "register_operand" "")
@@ -293,13 +320,37 @@
   rtx (*gen) (rtx, rtx, rtx, rtx);
 
   /* Use an atomic load-operate instruction when possible.  */
-  if (aarch64_atomic_ldop_supported_p (<CODE>))
-    gen = gen_aarch64_atomic_fetch_<atomic_optab><mode>_lse;
+  if (TARGET_LSE)
+    {
+      switch (<CODE>)
+        {
+	case MINUS:
+	  operands[2] = expand_simple_unop (<MODE>mode, NEG, operands[2],
+					    NULL, 1);
+	  /* fallthru */
+	case PLUS:
+	  gen = gen_aarch64_atomic_fetch_add<mode>_lse;
+	  break;
+	case IOR:
+	  gen = gen_aarch64_atomic_fetch_ior<mode>_lse;
+	  break;
+	case XOR:
+	  gen = gen_aarch64_atomic_fetch_xor<mode>_lse;
+	  break;
+	case AND:
+	  operands[2] = expand_simple_unop (<MODE>mode, NOT, operands[2],
+					    NULL, 1);
+	  gen = gen_aarch64_atomic_fetch_bic<mode>_lse;
+	  break;
+	default:
+	  gcc_unreachable ();
+	}
+      operands[2] = force_reg (<MODE>mode, operands[2]);
+    }
   else
     gen = gen_aarch64_atomic_fetch_<atomic_optab><mode>;
 
   emit_insn (gen (operands[0], operands[1], operands[2], operands[3]));
-
   DONE;
 })
 
@@ -326,23 +377,26 @@
   }
 )
 
-(define_insn_and_split "aarch64_atomic_fetch_<atomic_optab><mode>_lse"
-  [(set (match_operand:ALLI 0 "register_operand" "=&r")
-    (match_operand:ALLI 1 "aarch64_sync_memory_operand" "+Q"))
+(define_insn "aarch64_atomic_fetch_<atomic_ldoptab><mode>_lse"
+  [(set (match_operand:ALLI 0 "register_operand" "=r")
+	(match_operand:ALLI 1 "aarch64_sync_memory_operand" "+Q"))
    (set (match_dup 1)
-    (unspec_volatile:ALLI
-      [(atomic_op:ALLI (match_dup 1)
-	(match_operand:ALLI 2 "<atomic_op_operand>" "r<const_atomic>"))
-       (match_operand:SI 3 "const_int_operand")]
-      UNSPECV_ATOMIC_LDOP))]
+	(unspec_volatile:ALLI
+	  [(match_dup 1)
+	   (match_operand:ALLI 2 "register_operand" "r")
+	   (match_operand:SI 3 "const_int_operand")]
+	  ATOMIC_LDOP))]
   "TARGET_LSE"
-  "#"
-  "&& reload_completed"
-  [(const_int 0)]
   {
-    aarch64_gen_atomic_ldop (<CODE>, operands[0], NULL, operands[1],
-			     operands[2], operands[3]);
-    DONE;
+   enum memmodel model = memmodel_from_int (INTVAL (operands[3]));
+   if (is_mm_relaxed (model))
+     return "ld<atomic_ldop><atomic_sfx>\t%<w>2, %<w>0, %1";
+   else if (is_mm_acquire (model) || is_mm_consume (model))
+     return "ld<atomic_ldop>a<atomic_sfx>\t%<w>2, %<w>0, %1";
+   else if (is_mm_release (model))
+     return "ld<atomic_ldop>l<atomic_sfx>\t%<w>2, %<w>0, %1";
+   else
+     return "ld<atomic_ldop>al<atomic_sfx>\t%<w>2, %<w>0, %1";
   }
 )
 
@@ -370,7 +424,7 @@
   }
 )
 
-;; Load-operate-store, returning the original memory data.
+;; Load-operate-store, returning the updated memory data.
 
 (define_expand "atomic_<atomic_optab>_fetch<mode>"
  [(match_operand:ALLI 0 "register_operand" "")
@@ -380,17 +434,23 @@
   (match_operand:SI 3 "const_int_operand")]
  ""
 {
-  rtx (*gen) (rtx, rtx, rtx, rtx);
-  rtx value = operands[2];
-
-  /* Use an atomic load-operate instruction when possible.  */
-  if (aarch64_atomic_ldop_supported_p (<CODE>))
-    gen = gen_aarch64_atomic_<atomic_optab>_fetch<mode>_lse;
+  /* Use an atomic load-operate instruction when possible.  In this case
+     we will re-compute the result from the original mem value. */
+  if (TARGET_LSE)
+    {
+      rtx tmp = gen_reg_rtx (<MODE>mode);
+      operands[2] = force_reg (<MODE>mode, operands[2]);
+      emit_insn (gen_atomic_fetch_<atomic_optab><mode>
+                 (tmp, operands[1], operands[2], operands[3]));
+      tmp = expand_simple_binop (<MODE>mode, <CODE>, tmp, operands[2],
+				 operands[0], 1, OPTAB_WIDEN);
+      emit_move_insn (operands[0], tmp);
+    }
   else
-    gen = gen_aarch64_atomic_<atomic_optab>_fetch<mode>;
-
-  emit_insn (gen (operands[0], operands[1], value, operands[3]));
-
+    {
+      emit_insn (gen_aarch64_atomic_<atomic_optab>_fetch<mode>
+                 (operands[0], operands[1], operands[2], operands[3]));
+    }
   DONE;
 })
 
@@ -417,29 +477,6 @@
   }
 )
 
-(define_insn_and_split "aarch64_atomic_<atomic_optab>_fetch<mode>_lse"
-  [(set (match_operand:ALLI 0 "register_operand" "=&r")
-    (atomic_op:ALLI
-     (match_operand:ALLI 1 "aarch64_sync_memory_operand" "+Q")
-     (match_operand:ALLI 2 "<atomic_op_operand>" "r<const_atomic>")))
-   (set (match_dup 1)
-    (unspec_volatile:ALLI
-      [(match_dup 1)
-       (match_dup 2)
-       (match_operand:SI 3 "const_int_operand")]
-      UNSPECV_ATOMIC_LDOP))
-     (clobber (match_scratch:ALLI 4 "=&r"))]
-  "TARGET_LSE"
-  "#"
-  "&& reload_completed"
-  [(const_int 0)]
-  {
-    aarch64_gen_atomic_ldop (<CODE>, operands[4], operands[0], operands[1],
-			     operands[2], operands[3]);
-    DONE;
-  }
-)
-
 (define_insn_and_split "atomic_nand_fetch<mode>"
   [(set (match_operand:ALLI 0 "register_operand" "=&r")
     (not:ALLI
@@ -582,29 +619,3 @@
       return "dmb\\tish";
   }
 )
-
-;; ARMv8.1-A LSE instructions.
-
-;; Atomic load-op: Load data, operate, store result, keep data.
-
-(define_insn "aarch64_atomic_load<atomic_ldop><mode>"
- [(set (match_operand:ALLI 0 "register_operand" "=r")
-   (match_operand:ALLI 1 "aarch64_sync_memory_operand" "+Q"))
-  (set (match_dup 1)
-   (unspec_volatile:ALLI
-    [(match_dup 1)
-     (match_operand:ALLI 2 "register_operand")
-     (match_operand:SI 3 "const_int_operand")]
-    ATOMIC_LDOP))]
- "TARGET_LSE && reload_completed"
- {
-   enum memmodel model = memmodel_from_int (INTVAL (operands[3]));
-   if (is_mm_relaxed (model))
-     return "ld<atomic_ldop><atomic_sfx>\t%<w>2, %<w>0, %1";
-   else if (is_mm_acquire (model) || is_mm_consume (model))
-     return "ld<atomic_ldop>a<atomic_sfx>\t%<w>2, %<w>0, %1";
-   else if (is_mm_release (model))
-     return "ld<atomic_ldop>l<atomic_sfx>\t%<w>2, %<w>0, %1";
-   else
-     return "ld<atomic_ldop>al<atomic_sfx>\t%<w>2, %<w>0, %1";
- })
diff --git a/gcc/config/aarch64/iterators.md b/gcc/config/aarch64/iterators.md
index 4a39e30da13..fc87fc902a6 100644
--- a/gcc/config/aarch64/iterators.md
+++ b/gcc/config/aarch64/iterators.md
@@ -369,7 +369,6 @@
     UNSPECV_ATOMIC_CAS		; Represent an atomic CAS.
     UNSPECV_ATOMIC_SWP		; Represent an atomic SWP.
     UNSPECV_ATOMIC_OP		; Represent an atomic operation.
-    UNSPECV_ATOMIC_LDOP		; Represent an atomic load-operation
     UNSPECV_ATOMIC_LDOP_OR	; Represent an atomic load-or
     UNSPECV_ATOMIC_LDOP_BIC	; Represent an atomic load-bic
     UNSPECV_ATOMIC_LDOP_XOR	; Represent an atomic load-xor
@@ -1113,6 +1112,10 @@
  [(UNSPECV_ATOMIC_LDOP_OR "set") (UNSPECV_ATOMIC_LDOP_BIC "clr")
   (UNSPECV_ATOMIC_LDOP_XOR "eor") (UNSPECV_ATOMIC_LDOP_PLUS "add")])
 
+(define_int_attr atomic_ldoptab
+ [(UNSPECV_ATOMIC_LDOP_OR "ior") (UNSPECV_ATOMIC_LDOP_BIC "bic")
+  (UNSPECV_ATOMIC_LDOP_XOR "xor") (UNSPECV_ATOMIC_LDOP_PLUS "add")])
+
 ;; -------------------------------------------------------------------
 ;; Int Iterators Attributes.
 ;; -------------------------------------------------------------------
-- 
2.26.2

From 7ab1ff93727b1c32ea3a8dcfc5f068e0c4e1acf8 Mon Sep 17 00:00:00 2001
From: Richard Henderson <richard.henderson@linaro.org>
Date: Wed, 31 Oct 2018 23:11:22 +0000
Subject: [PATCH 08/24] aarch64: Remove early clobber from ATOMIC_LDOP scratch
To: gcc-patches@gcc.gnu.org

2018-10-31  Richard Henderson  <richard.henderson@linaro.org>

gcc/
	* config/aarch64/atomics.md (aarch64_atomic_<ATOMIC_LDOP><ALLI>_lse):
	scratch register need not be early-clobber.  Document the reason
	why we cannot use ST<OP>.

(cherry picked from commit 53de1ea800db54b47290d578c43892799b66c8dc)
---
 gcc/config/aarch64/atomics.md | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/gcc/config/aarch64/atomics.md b/gcc/config/aarch64/atomics.md
index 735407c9fd7..1ef7c20db21 100644
--- a/gcc/config/aarch64/atomics.md
+++ b/gcc/config/aarch64/atomics.md
@@ -263,6 +263,18 @@
   }
 )
 
+;; It is tempting to want to use ST<OP> for relaxed and release
+;; memory models here.  However, that is incompatible with the
+;; C++ memory model for the following case:
+;;
+;;	atomic_fetch_add(ptr, 1, memory_order_relaxed);
+;;	atomic_thread_fence(memory_order_acquire);
+;;
+;; The problem is that the architecture says that ST<OP> (and LD<OP>
+;; insns where the destination is XZR) are not regarded as a read.
+;; However we also implement the acquire memory barrier with DMB LD,
+;; and so the ST<OP> is not blocked by the barrier.
+
 (define_insn "aarch64_atomic_<atomic_ldoptab><mode>_lse"
   [(set (match_operand:ALLI 0 "aarch64_sync_memory_operand" "+Q")
 	(unspec_volatile:ALLI
@@ -270,7 +282,7 @@
 	   (match_operand:ALLI 1 "register_operand" "r")
 	   (match_operand:SI 2 "const_int_operand")]
       ATOMIC_LDOP))
-   (clobber (match_scratch:ALLI 3 "=&r"))]
+   (clobber (match_scratch:ALLI 3 "=r"))]
   "TARGET_LSE"
   {
    enum memmodel model = memmodel_from_int (INTVAL (operands[2]));
-- 
2.26.2

From 3d05e960c18653ce1e4f19fba701645b1a030da7 Mon Sep 17 00:00:00 2001
From: Richard Henderson <richard.henderson@linaro.org>
Date: Thu, 19 Sep 2019 14:36:24 +0000
Subject: [PATCH 09/24] aarch64: Extend %R for integer registers
To: gcc-patches@gcc.gnu.org

2019-09-19  Richard Henderson  <richard.henderson@linaro.org>

gcc/
	* config/aarch64/aarch64.c (aarch64_print_operand): Allow integer
	registers with %R.

(cherry picked from commit e3f15286d1129de2cceee6acd5d5584cb5422db6)
---
 gcc/config/aarch64/aarch64.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index ed3cec30859..ec325813f5a 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -5082,12 +5082,13 @@ aarch64_print_operand (FILE *f, rtx x, int code)
     case 'd':
     case 'q':
       /* Print a scalar FP/SIMD register name.  */
-      if (!REG_P (x) || !FP_REGNUM_P (REGNO (x)))
-	{
-	  output_operand_lossage ("incompatible floating point / vector register operand for '%%%c'", code);
-	  return;
-	}
-      asm_fprintf (f, "%c%d", code, REGNO (x) - V0_REGNUM);
+      if (REG_P (x) && FP_REGNUM_P (REGNO (x)))
+	asm_fprintf (f, "%c%d", code, REGNO (x) - V0_REGNUM);
+      else if (REG_P (x) && GP_REGNUM_P (REGNO (x)))
+	asm_fprintf (f, "%c%d", code, REGNO (x) - R0_REGNUM);
+      else
+	output_operand_lossage ("incompatible register operand for '%%%c'",
+				code);
       break;
 
     case 'S':
-- 
2.26.2

From ed63b93e1aefcea56305ef13e4b4726af0d4f27f Mon Sep 17 00:00:00 2001
From: Richard Henderson <richard.henderson@linaro.org>
Date: Thu, 19 Sep 2019 14:36:29 +0000
Subject: [PATCH 10/24] aarch64: Implement TImode compare-and-swap
To: gcc-patches@gcc.gnu.org

This pattern will only be used with the __sync functions, because
we do not yet have a bare TImode atomic load.

2019-09-19  Richard Henderson  <richard.henderson@linaro.org>

gcc/
	* config/aarch64/aarch64.c (aarch64_gen_compare_reg): Add support
	for NE comparison of TImode values.
	(aarch64_emit_load_exclusive): Add support for TImode.
	(aarch64_emit_store_exclusive): Likewise.
	(aarch64_split_compare_and_swap): Disable strong_zero_p for TImode.
	* config/aarch64/atomics.md (@atomic_compare_and_swap<ALLI_TI>):
	Change iterator from ALLI to ALLI_TI.
	(@atomic_compare_and_swap<JUST_TI>): New.
	(@atomic_compare_and_swap<JUST_TI>_lse): New.
	(aarch64_load_exclusive_pair): New.
	(aarch64_store_exclusive_pair): New.
	* config/aarch64/iterators.md (JUST_TI): New.

(cherry picked from commit 4a2095ebace8534038ce2adf4ae94bfc854066c4)
---
 gcc/config/aarch64/aarch64.c    | 59 ++++++++++++++++++---
 gcc/config/aarch64/atomics.md   | 93 +++++++++++++++++++++++++++++++--
 gcc/config/aarch64/iterators.md |  6 +++
 3 files changed, 145 insertions(+), 13 deletions(-)

diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index ec325813f5a..e86f34edcc6 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -1191,10 +1191,33 @@ emit_set_insn (rtx x, rtx y)
 rtx
 aarch64_gen_compare_reg (RTX_CODE code, rtx x, rtx y)
 {
-  machine_mode mode = SELECT_CC_MODE (code, x, y);
-  rtx cc_reg = gen_rtx_REG (mode, CC_REGNUM);
+  machine_mode cmp_mode = GET_MODE (x);
+  machine_mode cc_mode;
+  rtx cc_reg;
 
-  emit_set_insn (cc_reg, gen_rtx_COMPARE (mode, x, y));
+  if (cmp_mode == TImode)
+    {
+      gcc_assert (code == NE);
+
+      cc_mode = CCmode;
+      cc_reg = gen_rtx_REG (cc_mode, CC_REGNUM);
+
+      rtx x_lo = operand_subword (x, 0, 0, TImode);
+      rtx y_lo = operand_subword (y, 0, 0, TImode);
+      emit_set_insn (cc_reg, gen_rtx_COMPARE (cc_mode, x_lo, y_lo));
+
+      rtx x_hi = operand_subword (x, 1, 0, TImode);
+      rtx y_hi = operand_subword (y, 1, 0, TImode);
+      emit_insn (gen_ccmpdi (cc_reg, cc_reg, x_hi, y_hi,
+			     gen_rtx_EQ (cc_mode, cc_reg, const0_rtx),
+			     GEN_INT (AARCH64_EQ)));
+    }
+  else
+    {
+      cc_mode = SELECT_CC_MODE (code, x, y);
+      cc_reg = gen_rtx_REG (cc_mode, CC_REGNUM);
+      emit_set_insn (cc_reg, gen_rtx_COMPARE (cc_mode, x, y));
+    }
   return cc_reg;
 }
 
@@ -11839,6 +11862,14 @@ aarch64_emit_load_exclusive (machine_mode mode, rtx rval,
 {
   rtx (*gen) (rtx, rtx, rtx);
 
+  if (mode == TImode)
+    {
+      emit_insn (gen_aarch64_load_exclusive_pair
+		 (gen_lowpart (DImode, rval), gen_highpart (DImode, rval),
+		  mem, model_rtx));
+      return;
+    }
+
   switch (mode)
     {
     case QImode: gen = gen_aarch64_load_exclusiveqi; break;
@@ -11856,10 +11887,18 @@ aarch64_emit_load_exclusive (machine_mode mode, rtx rval,
 
 static void
 aarch64_emit_store_exclusive (machine_mode mode, rtx bval,
-			      rtx rval, rtx mem, rtx model_rtx)
+			      rtx mem, rtx rval, rtx model_rtx)
 {
   rtx (*gen) (rtx, rtx, rtx, rtx);
 
+  if (mode == TImode)
+    {
+      emit_insn (gen_aarch64_store_exclusive_pair
+		 (bval, mem, operand_subword (rval, 0, 0, TImode),
+		  operand_subword (rval, 1, 0, TImode), model_rtx));
+      return;
+    }
+
   switch (mode)
     {
     case QImode: gen = gen_aarch64_store_exclusiveqi; break;
@@ -11870,7 +11909,7 @@ aarch64_emit_store_exclusive (machine_mode mode, rtx bval,
       gcc_unreachable ();
     }
 
-  emit_insn (gen (bval, rval, mem, model_rtx));
+  emit_insn (gen (bval, mem, rval, model_rtx));
 }
 
 /* Mark the previous jump instruction as unlikely.  */
@@ -11898,7 +11937,8 @@ aarch64_expand_compare_and_swap (rtx operands[])
     gen_aarch64_compare_and_swapqi,
     gen_aarch64_compare_and_swaphi,
     gen_aarch64_compare_and_swapsi,
-    gen_aarch64_compare_and_swapdi
+    gen_aarch64_compare_and_swapdi,
+    gen_aarch64_compare_and_swapti
   };
   typedef rtx (*gen_lse_fn) (rtx, rtx, rtx, rtx);
   const gen_lse_fn atomic_cas[] =
@@ -11906,7 +11946,8 @@ aarch64_expand_compare_and_swap (rtx operands[])
     gen_aarch64_compare_and_swapqi_lse,
     gen_aarch64_compare_and_swaphi_lse,
     gen_aarch64_compare_and_swapsi_lse,
-    gen_aarch64_compare_and_swapdi_lse
+    gen_aarch64_compare_and_swapdi_lse,
+    gen_aarch64_compare_and_swapti_lse
   };
 
   bval = operands[0];
@@ -11939,6 +11980,7 @@ aarch64_expand_compare_and_swap (rtx operands[])
     case HImode: idx = 1; break;
     case SImode: idx = 2; break;
     case DImode: idx = 3; break;
+    case TImode: idx = 4; break;
     default:
       gcc_unreachable ();
     }
@@ -11965,6 +12007,7 @@ aarch64_expand_compare_and_swap (rtx operands[])
 	case HImode: code = CODE_FOR_aarch64_compare_and_swaphi; break;
 	case SImode: code = CODE_FOR_aarch64_compare_and_swapsi; break;
 	case DImode: code = CODE_FOR_aarch64_compare_and_swapdi; break;
+	case TImode: code = CODE_FOR_aarch64_compare_and_swapti; break;
 	default:
 	  gcc_unreachable ();
 	}
@@ -12033,7 +12076,7 @@ aarch64_split_compare_and_swap (rtx operands[])
 	CBNZ	scratch, .label1
     .label2:
 	CMP	rval, 0.  */
-  bool strong_zero_p = !is_weak && oldval == const0_rtx;
+  bool strong_zero_p = !is_weak && oldval == const0_rtx && mode != TImode;
 
   label1 = NULL;
   if (!is_weak)
diff --git a/gcc/config/aarch64/atomics.md b/gcc/config/aarch64/atomics.md
index 1ef7c20db21..316c84699d0 100644
--- a/gcc/config/aarch64/atomics.md
+++ b/gcc/config/aarch64/atomics.md
@@ -22,10 +22,10 @@
 
 (define_expand "atomic_compare_and_swap<mode>"
   [(match_operand:SI 0 "register_operand" "")			;; bool out
-   (match_operand:ALLI 1 "register_operand" "")			;; val out
-   (match_operand:ALLI 2 "aarch64_sync_memory_operand" "")	;; memory
-   (match_operand:ALLI 3 "nonmemory_operand" "")		;; expected
-   (match_operand:ALLI 4 "aarch64_reg_or_zero" "")		;; desired
+   (match_operand:ALLI_TI 1 "register_operand" "")		;; val out
+   (match_operand:ALLI_TI 2 "aarch64_sync_memory_operand" "")	;; memory
+   (match_operand:ALLI_TI 3 "nonmemory_operand" "")		;; expected
+   (match_operand:ALLI_TI 4 "aarch64_reg_or_zero" "")		;; desired
    (match_operand:SI 5 "const_int_operand")			;; is_weak
    (match_operand:SI 6 "const_int_operand")			;; mod_s
    (match_operand:SI 7 "const_int_operand")]			;; mod_f
@@ -88,9 +88,33 @@
   }
 )
 
+(define_insn_and_split "aarch64_compare_and_swap<mode>"
+  [(set (reg:CC CC_REGNUM)					;; bool out
+    (unspec_volatile:CC [(const_int 0)] UNSPECV_ATOMIC_CMPSW))
+   (set (match_operand:JUST_TI 0 "register_operand" "=&r")	;; val out
+    (match_operand:JUST_TI 1 "aarch64_sync_memory_operand" "+Q")) ;; memory
+   (set (match_dup 1)
+    (unspec_volatile:JUST_TI
+      [(match_operand:JUST_TI 2 "aarch64_reg_or_zero" "rZ")	;; expect
+       (match_operand:JUST_TI 3 "aarch64_reg_or_zero" "rZ")	;; desired
+       (match_operand:SI 4 "const_int_operand")			;; is_weak
+       (match_operand:SI 5 "const_int_operand")			;; mod_s
+       (match_operand:SI 6 "const_int_operand")]		;; mod_f
+      UNSPECV_ATOMIC_CMPSW))
+   (clobber (match_scratch:SI 7 "=&r"))]
+  ""
+  "#"
+  "&& reload_completed"
+  [(const_int 0)]
+  {
+    aarch64_split_compare_and_swap (operands);
+    DONE;
+  }
+)
+
 (define_insn "aarch64_compare_and_swap<mode>_lse"
   [(set (match_operand:SI 0 "register_operand" "+r")		;; val out
-     (zero_extend:SI
+    (zero_extend:SI
      (match_operand:SHORT 1 "aarch64_sync_memory_operand" "+Q"))) ;; memory
     (set (match_dup 1)
      (unspec_volatile:SHORT
@@ -133,6 +157,28 @@
     return "casal<atomic_sfx>\t%<w>0, %<w>2, %1";
 })
 
+(define_insn "aarch64_compare_and_swap<mode>_lse"
+  [(set (match_operand:JUST_TI 0 "register_operand" "+r")	;; val out
+    (match_operand:JUST_TI 1 "aarch64_sync_memory_operand" "+Q")) ;; memory
+   (set (match_dup 1)
+    (unspec_volatile:JUST_TI
+      [(match_dup 0)						;; expect
+       (match_operand:JUST_TI 2 "register_operand" "r")		;; desired
+       (match_operand:SI 3 "const_int_operand")]		;; mod_s
+      UNSPECV_ATOMIC_CMPSW))]
+  "TARGET_LSE"
+{
+  enum memmodel model = memmodel_from_int (INTVAL (operands[3]));
+  if (is_mm_relaxed (model))
+    return "casp\t%0, %R0, %2, %R2, %1";
+  else if (is_mm_acquire (model) || is_mm_consume (model))
+    return "caspa\t%0, %R0, %2, %R2, %1";
+  else if (is_mm_release (model))
+    return "caspl\t%0, %R0, %2, %R2, %1";
+  else
+    return "caspal\t%0, %R0, %2, %R2, %1";
+})
+
 (define_expand "atomic_exchange<mode>"
  [(match_operand:ALLI 0 "register_operand" "")
   (match_operand:ALLI 1 "aarch64_sync_memory_operand" "")
@@ -578,6 +624,24 @@
   }
 )
 
+(define_insn "aarch64_load_exclusive_pair"
+  [(set (match_operand:DI 0 "register_operand" "=r")
+	(unspec_volatile:DI
+	  [(match_operand:TI 2 "aarch64_sync_memory_operand" "Q")
+	   (match_operand:SI 3 "const_int_operand")]
+	  UNSPECV_LX))
+   (set (match_operand:DI 1 "register_operand" "=r")
+	(unspec_volatile:DI [(match_dup 2) (match_dup 3)] UNSPECV_LX))]
+  ""
+  {
+    enum memmodel model = memmodel_from_int (INTVAL (operands[3]));
+    if (is_mm_relaxed (model) || is_mm_consume (model) || is_mm_release (model))
+      return "ldxp\t%0, %1, %2";
+    else
+      return "ldaxp\t%0, %1, %2";
+  }
+)
+
 (define_insn "aarch64_store_exclusive<mode>"
   [(set (match_operand:SI 0 "register_operand" "=&r")
     (unspec_volatile:SI [(const_int 0)] UNSPECV_SX))
@@ -596,6 +660,25 @@
   }
 )
 
+(define_insn "aarch64_store_exclusive_pair"
+  [(set (match_operand:SI 0 "register_operand" "=&r")
+	(unspec_volatile:SI [(const_int 0)] UNSPECV_SX))
+   (set (match_operand:TI 1 "aarch64_sync_memory_operand" "=Q")
+	(unspec_volatile:TI
+	  [(match_operand:DI 2 "aarch64_reg_or_zero" "rZ")
+	   (match_operand:DI 3 "aarch64_reg_or_zero" "rZ")
+	   (match_operand:SI 4 "const_int_operand")]
+	  UNSPECV_SX))]
+  ""
+  {
+    enum memmodel model = memmodel_from_int (INTVAL (operands[3]));
+    if (is_mm_relaxed (model) || is_mm_consume (model) || is_mm_acquire (model))
+      return "stxp\t%w0, %x2, %x3, %1";
+    else
+      return "stlxp\t%w0, %x2, %x3, %1";
+  }
+)
+
 (define_expand "mem_thread_fence"
   [(match_operand:SI 0 "const_int_operand" "")]
   ""
diff --git a/gcc/config/aarch64/iterators.md b/gcc/config/aarch64/iterators.md
index fc87fc902a6..813f53690b7 100644
--- a/gcc/config/aarch64/iterators.md
+++ b/gcc/config/aarch64/iterators.md
@@ -29,12 +29,18 @@
 ;; Iterator for HI, SI, DI, some instructions can only work on these modes.
 (define_mode_iterator GPI_I16 [(HI "AARCH64_ISA_F16") SI DI])
 
+;; "Iterator" for just TI -- features like @pattern only work with iterators.
+(define_mode_iterator JUST_TI [TI])
+
 ;; Iterator for QI and HI modes
 (define_mode_iterator SHORT [QI HI])
 
 ;; Iterator for all integer modes (up to 64-bit)
 (define_mode_iterator ALLI [QI HI SI DI])
 
+;; Iterator for all integer modes (up to 128-bit)
+(define_mode_iterator ALLI_TI [QI HI SI DI TI])
+
 ;; Iterator for all integer modes that can be extended (up to 64-bit)
 (define_mode_iterator ALLX [QI HI SI])
 
-- 
2.26.2

From a10b63386c2e87e7712ee2d1705a6af74ced8fec Mon Sep 17 00:00:00 2001
From: Wilco Dijkstra <wdijkstr@arm.com>
Date: Fri, 17 Jan 2020 13:17:21 +0000
Subject: [PATCH 11/24] Fix shrinkwrapping interactions with atomics (PR92692)
To: gcc-patches@gcc.gnu.org

The separate shrinkwrapping pass may insert stores in the middle
of atomics loops which can cause issues on some implementations.
Avoid this by delaying splitting atomics patterns until after
prolog/epilog generation.

2020-01-17  Wilco Dijkstra  <wdijkstr@arm.com>

gcc/
	PR target/92692
	* config/aarch64/aarch64.c (aarch64_split_compare_and_swap)
	Add assert to ensure prolog has been emitted.
	(aarch64_split_atomic_op): Likewise.
	* config/aarch64/atomics.md (aarch64_compare_and_swap<mode>)
	Use epilogue_completed rather than reload_completed.
	(aarch64_atomic_exchange<mode>): Likewise.
	(aarch64_atomic_<atomic_optab><mode>): Likewise.
	(atomic_nand<mode>): Likewise.
	(aarch64_atomic_fetch_<atomic_optab><mode>): Likewise.
	(atomic_fetch_nand<mode>): Likewise.
	(aarch64_atomic_<atomic_optab>_fetch<mode>): Likewise.
	(atomic_nand_fetch<mode>): Likewise.

(cherry picked from commit e5e07b68187b9aa334519746c45b8cffc5eb7e5c)
---
 gcc/config/aarch64/atomics.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gcc/config/aarch64/atomics.md b/gcc/config/aarch64/atomics.md
index 316c84699d0..f32f5528f1a 100644
--- a/gcc/config/aarch64/atomics.md
+++ b/gcc/config/aarch64/atomics.md
@@ -104,7 +104,7 @@
    (clobber (match_scratch:SI 7 "=&r"))]
   ""
   "#"
-  "&& reload_completed"
+  "&& epilogue_completed"
   [(const_int 0)]
   {
     aarch64_split_compare_and_swap (operands);
-- 
2.26.2

From fd8684c86bebd765649e7e4b70f151c87a1f13da Mon Sep 17 00:00:00 2001
From: Kyrylo Tkachov <kyrylo.tkachov@arm.com>
Date: Wed, 1 Apr 2020 11:54:14 +0100
Subject: [PATCH 12/24] aarch64: Tidy aarch64_split_compare_and_swap
To: gcc-patches@gcc.gnu.org

2019-09-19  Richard Henderson  <richard.henderson@linaro.org>

* config/aarch64/aarch64 (aarch64_split_compare_and_swap): Disable
	strong_zero_p for aarch64_track_speculation; unify some code paths;
	use aarch64_gen_compare_reg instead of open-coding.

(cherry picked from commit b7e560deb37e38fb224a0cf108e15df4a717167a)
---
 gcc/config/aarch64/aarch64.c | 40 ++++++++++++------------------------
 1 file changed, 13 insertions(+), 27 deletions(-)

diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index e86f34edcc6..53de92e65fc 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -12049,13 +12049,11 @@ aarch64_emit_post_barrier (enum memmodel model)
 void
 aarch64_split_compare_and_swap (rtx operands[])
 {
-  rtx rval, mem, oldval, newval, scratch;
+  rtx rval, mem, oldval, newval, scratch, x, model_rtx;
   machine_mode mode;
   bool is_weak;
   rtx_code_label *label1, *label2;
-  rtx x, cond;
   enum memmodel model;
-  rtx model_rtx;
 
   rval = operands[0];
   mem = operands[1];
@@ -12076,7 +12074,7 @@ aarch64_split_compare_and_swap (rtx operands[])
 	CBNZ	scratch, .label1
     .label2:
 	CMP	rval, 0.  */
-  bool strong_zero_p = !is_weak && oldval == const0_rtx && mode != TImode;
+  bool strong_zero_p = (!is_weak && oldval == const0_rtx && mode != TImode);
 
   label1 = NULL;
   if (!is_weak)
@@ -12089,26 +12087,20 @@ aarch64_split_compare_and_swap (rtx operands[])
   /* The initial load can be relaxed for a __sync operation since a final
      barrier will be emitted to stop code hoisting.  */
   if (is_mm_sync (model))
-    aarch64_emit_load_exclusive (mode, rval, mem,
-				 GEN_INT (MEMMODEL_RELAXED));
+    aarch64_emit_load_exclusive (mode, rval, mem, GEN_INT (MEMMODEL_RELAXED));
   else
     aarch64_emit_load_exclusive (mode, rval, mem, model_rtx);
 
   if (strong_zero_p)
-    {
-      x = gen_rtx_NE (VOIDmode, rval, const0_rtx);
-      x = gen_rtx_IF_THEN_ELSE (VOIDmode, x,
-				gen_rtx_LABEL_REF (Pmode, label2), pc_rtx);
-      aarch64_emit_unlikely_jump (gen_rtx_SET (pc_rtx, x));
-    }
+    x = gen_rtx_NE (VOIDmode, rval, const0_rtx);
   else
     {
-      cond = aarch64_gen_compare_reg_maybe_ze (NE, rval, oldval, mode);
-      x = gen_rtx_NE (VOIDmode, cond, const0_rtx);
-      x = gen_rtx_IF_THEN_ELSE (VOIDmode, x,
-				gen_rtx_LABEL_REF (Pmode, label2), pc_rtx);
-      aarch64_emit_unlikely_jump (gen_rtx_SET (pc_rtx, x));
+      rtx cc_reg = aarch64_gen_compare_reg_maybe_ze (NE, rval, oldval, mode);
+      x = gen_rtx_NE (VOIDmode, cc_reg, const0_rtx);
     }
+  x = gen_rtx_IF_THEN_ELSE (VOIDmode, x,
+			    gen_rtx_LABEL_REF (Pmode, label2), pc_rtx);
+  aarch64_emit_unlikely_jump (gen_rtx_SET (pc_rtx, x));
 
   aarch64_emit_store_exclusive (mode, scratch, mem, newval, model_rtx);
 
@@ -12120,22 +12112,16 @@ aarch64_split_compare_and_swap (rtx operands[])
       aarch64_emit_unlikely_jump (gen_rtx_SET (pc_rtx, x));
     }
   else
-    {
-      cond = gen_rtx_REG (CCmode, CC_REGNUM);
-      x = gen_rtx_COMPARE (CCmode, scratch, const0_rtx);
-      emit_insn (gen_rtx_SET (cond, x));
-    }
+    aarch64_gen_compare_reg (NE, scratch, const0_rtx);
 
   emit_label (label2);
+
   /* If we used a CBNZ in the exchange loop emit an explicit compare with RVAL
      to set the condition flags.  If this is not used it will be removed by
      later passes.  */
   if (strong_zero_p)
-    {
-      cond = gen_rtx_REG (CCmode, CC_REGNUM);
-      x = gen_rtx_COMPARE (CCmode, rval, const0_rtx);
-      emit_insn (gen_rtx_SET (cond, x));
-    }
+    aarch64_gen_compare_reg (NE, rval, const0_rtx);
+
   /* Emit any final barrier needed for a __sync operation.  */
   if (is_mm_sync (model))
     aarch64_emit_post_barrier (model);
-- 
2.26.2

From a77275f87d9e4c33a00610c5ccfba48c6eeffe55 Mon Sep 17 00:00:00 2001
From: Richard Henderson <richard.henderson@linaro.org>
Date: Thu, 19 Sep 2019 14:36:38 +0000
Subject: [PATCH 13/24] aarch64: Add out-of-line functions for LSE atomics
To: gcc-patches@gcc.gnu.org

This is the libgcc part of the interface -- providing the functions.
Rationale is provided at the top of libgcc/config/aarch64/lse.S.

2019-09-19  Richard Henderson  <richard.henderson@linaro.org>

gcc/
	* config/aarch64/lse-init.c: New file.
	* config/aarch64/lse.S: New file.
	* config/aarch64/t-lse: New file.
	* config.host: Add t-lse to all aarch64 tuples.

(cherry picked from commit 33befddcb849235353dc263db1c7d07dc15c9faa)
---
 libgcc/config.host               |   3 +
 libgcc/config/aarch64/lse-init.c |  45 ++++++
 libgcc/config/aarch64/lse.S      | 235 +++++++++++++++++++++++++++++++
 libgcc/config/aarch64/t-lse      |  44 ++++++
 4 files changed, 327 insertions(+)
 create mode 100644 libgcc/config/aarch64/lse-init.c
 create mode 100644 libgcc/config/aarch64/lse.S
 create mode 100644 libgcc/config/aarch64/t-lse

diff --git a/libgcc/config.host b/libgcc/config.host
index b8e23766695..b937cfb1763 100644
--- a/libgcc/config.host
+++ b/libgcc/config.host
@@ -334,12 +334,14 @@ aarch64*-*-elf | aarch64*-*-rtems*)
 	extra_parts="$extra_parts crtbegin.o crtend.o crti.o crtn.o"
 	extra_parts="$extra_parts crtfastmath.o"
 	tmake_file="${tmake_file} ${cpu_type}/t-aarch64"
+	tmake_file="${tmake_file} ${cpu_type}/t-lse t-slibgcc-libgcc"
 	tmake_file="${tmake_file} ${cpu_type}/t-softfp t-softfp t-crtfm"
 	md_unwind_header=aarch64/aarch64-unwind.h
 	;;
 aarch64*-*-freebsd*)
 	extra_parts="$extra_parts crtfastmath.o"
 	tmake_file="${tmake_file} ${cpu_type}/t-aarch64"
+	tmake_file="${tmake_file} ${cpu_type}/t-lse t-slibgcc-libgcc"
 	tmake_file="${tmake_file} ${cpu_type}/t-softfp t-softfp t-crtfm"
 	md_unwind_header=aarch64/freebsd-unwind.h
 	;;
@@ -347,6 +349,7 @@ aarch64*-*-linux*)
 	extra_parts="$extra_parts crtfastmath.o"
 	md_unwind_header=aarch64/linux-unwind.h
 	tmake_file="${tmake_file} ${cpu_type}/t-aarch64"
+	tmake_file="${tmake_file} ${cpu_type}/t-lse t-slibgcc-libgcc"
 	tmake_file="${tmake_file} ${cpu_type}/t-softfp t-softfp t-crtfm"
 	;;
 alpha*-*-linux*)
diff --git a/libgcc/config/aarch64/lse-init.c b/libgcc/config/aarch64/lse-init.c
new file mode 100644
index 00000000000..33d29147479
--- /dev/null
+++ b/libgcc/config/aarch64/lse-init.c
@@ -0,0 +1,45 @@
+/* Out-of-line LSE atomics for AArch64 architecture, Init.
+   Copyright (C) 2019 Free Software Foundation, Inc.
+   Contributed by Linaro Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free
+Software Foundation; either version 3, or (at your option) any later
+version.
+
+GCC is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+for more details.
+
+Under Section 7 of GPL version 3, you are granted additional
+permissions described in the GCC Runtime Library Exception, version
+3.1, as published by the Free Software Foundation.
+
+You should have received a copy of the GNU General Public License and
+a copy of the GCC Runtime Library Exception along with this program;
+see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
+<http://www.gnu.org/licenses/>.  */
+
+/* Define the symbol gating the LSE implementations.  */
+_Bool __aarch64_have_lse_atomics
+  __attribute__((visibility("hidden"), nocommon));
+
+/* Disable initialization of __aarch64_have_lse_atomics during bootstrap.  */
+#ifndef inhibit_libc
+# include <sys/auxv.h>
+
+/* Disable initialization if the system headers are too old.  */
+# if defined(AT_HWCAP) && defined(HWCAP_ATOMICS)
+
+static void __attribute__((constructor))
+init_have_lse_atomics (void)
+{
+  unsigned long hwcap = getauxval (AT_HWCAP);
+  __aarch64_have_lse_atomics = (hwcap & HWCAP_ATOMICS) != 0;
+}
+
+# endif /* HWCAP */
+#endif /* inhibit_libc */
diff --git a/libgcc/config/aarch64/lse.S b/libgcc/config/aarch64/lse.S
new file mode 100644
index 00000000000..a5f6673596c
--- /dev/null
+++ b/libgcc/config/aarch64/lse.S
@@ -0,0 +1,235 @@
+/* Out-of-line LSE atomics for AArch64 architecture.
+   Copyright (C) 2019 Free Software Foundation, Inc.
+   Contributed by Linaro Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free
+Software Foundation; either version 3, or (at your option) any later
+version.
+
+GCC is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+for more details.
+
+Under Section 7 of GPL version 3, you are granted additional
+permissions described in the GCC Runtime Library Exception, version
+3.1, as published by the Free Software Foundation.
+
+You should have received a copy of the GNU General Public License and
+a copy of the GCC Runtime Library Exception along with this program;
+see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
+<http://www.gnu.org/licenses/>.  */
+
+/*
+ * The problem that we are trying to solve is operating system deployment
+ * of ARMv8.1-Atomics, also known as Large System Exensions (LSE).
+ *
+ * There are a number of potential solutions for this problem which have
+ * been proposed and rejected for various reasons.  To recap:
+ *
+ * (1) Multiple builds.  The dynamic linker will examine /lib64/atomics/
+ * if HWCAP_ATOMICS is set, allowing entire libraries to be overwritten.
+ * However, not all Linux distributions are happy with multiple builds,
+ * and anyway it has no effect on main applications.
+ *
+ * (2) IFUNC.  We could put these functions into libgcc_s.so, and have
+ * a single copy of each function for all DSOs.  However, ARM is concerned
+ * that the branch-to-indirect-branch that is implied by using a PLT,
+ * as required by IFUNC, is too much overhead for smaller cpus.
+ *
+ * (3) Statically predicted direct branches.  This is the approach that
+ * is taken here.  These functions are linked into every DSO that uses them.
+ * All of the symbols are hidden, so that the functions are called via a
+ * direct branch.  The choice of LSE vs non-LSE is done via one byte load
+ * followed by a well-predicted direct branch.  The functions are compiled
+ * separately to minimize code size.
+ */
+
+/* Tell the assembler to accept LSE instructions.  */
+	.arch armv8-a+lse
+
+/* Declare the symbol gating the LSE implementations.  */
+	.hidden	__aarch64_have_lse_atomics
+
+/* Turn size and memory model defines into mnemonic fragments.  */
+#if SIZE == 1
+# define S     b
+# define UXT   uxtb
+#elif SIZE == 2
+# define S     h
+# define UXT   uxth
+#elif SIZE == 4 || SIZE == 8 || SIZE == 16
+# define S
+# define UXT   mov
+#else
+# error
+#endif
+
+#if MODEL == 1
+# define SUFF  _relax
+# define A
+# define L
+#elif MODEL == 2
+# define SUFF  _acq
+# define A     a
+# define L
+#elif MODEL == 3
+# define SUFF  _rel
+# define A
+# define L     l
+#elif MODEL == 4
+# define SUFF  _acq_rel
+# define A     a
+# define L     l
+#else
+# error
+#endif
+
+/* Concatenate symbols.  */
+#define glue2_(A, B)		A ## B
+#define glue2(A, B)		glue2_(A, B)
+#define glue3_(A, B, C)		A ## B ## C
+#define glue3(A, B, C)		glue3_(A, B, C)
+#define glue4_(A, B, C, D)	A ## B ## C ## D
+#define glue4(A, B, C, D)	glue4_(A, B, C, D)
+
+/* Select the size of a register, given a regno.  */
+#define x(N)			glue2(x, N)
+#define w(N)			glue2(w, N)
+#if SIZE < 8
+# define s(N)			w(N)
+#else
+# define s(N)			x(N)
+#endif
+
+#define NAME(BASE)		glue4(__aarch64_, BASE, SIZE, SUFF)
+#define LDXR			glue4(ld, A, xr, S)
+#define STXR			glue4(st, L, xr, S)
+
+/* Temporary registers used.  Other than these, only the return value
+   register (x0) and the flags are modified.  */
+#define tmp0	16
+#define tmp1	17
+#define tmp2	15
+
+/* Start and end a function.  */
+.macro	STARTFN name
+	.text
+	.balign	16
+	.globl	\name
+	.hidden	\name
+	.type	\name, %function
+	.cfi_startproc
+\name:
+.endm
+
+.macro	ENDFN name
+	.cfi_endproc
+	.size	\name, . - \name
+.endm
+
+/* Branch to LABEL if LSE is disabled.  */
+.macro	JUMP_IF_NOT_LSE label
+	adrp	x(tmp0), __aarch64_have_lse_atomics
+	ldrb	w(tmp0), [x(tmp0), :lo12:__aarch64_have_lse_atomics]
+	cbz	w(tmp0), \label
+.endm
+
+#ifdef L_cas
+
+STARTFN	NAME(cas)
+	JUMP_IF_NOT_LSE	8f
+
+#if SIZE < 16
+#define CAS	glue4(cas, A, L, S)
+
+	CAS		s(0), s(1), [x2]
+	ret
+
+8:	UXT		s(tmp0), s(0)
+0:	LDXR		s(0), [x2]
+	cmp		s(0), s(tmp0)
+	bne		1f
+	STXR		w(tmp1), s(1), [x2]
+	cbnz		w(tmp1), 0b
+1:	ret
+
+#else
+#define LDXP	glue3(ld, A, xp)
+#define STXP	glue3(st, L, xp)
+#define CASP	glue3(casp, A, L)
+
+	CASP		x0, x1, x2, x3, [x4]
+	ret
+
+8:	mov		x(tmp0), x0
+	mov		x(tmp1), x1
+0:	LDXP		x0, x1, [x4]
+	cmp		x0, x(tmp0)
+	ccmp		x1, x(tmp1), #0, eq
+	bne		1f
+	STXP		w(tmp2), x(tmp0), x(tmp1), [x4]
+	cbnz		w(tmp2), 0b
+1:	ret
+
+#endif
+
+ENDFN	NAME(cas)
+#endif
+
+#ifdef L_swp
+#define SWP	glue4(swp, A, L, S)
+
+STARTFN	NAME(swp)
+	JUMP_IF_NOT_LSE	8f
+
+	SWP		s(0), s(0), [x1]
+	ret
+
+8:	mov		s(tmp0), s(0)
+0:	LDXR		s(0), [x1]
+	STXR		w(tmp1), s(tmp0), [x1]
+	cbnz		w(tmp1), 0b
+	ret
+
+ENDFN	NAME(swp)
+#endif
+
+#if defined(L_ldadd) || defined(L_ldclr) \
+    || defined(L_ldeor) || defined(L_ldset)
+
+#ifdef L_ldadd
+#define LDNM	ldadd
+#define OP	add
+#elif defined(L_ldclr)
+#define LDNM	ldclr
+#define OP	bic
+#elif defined(L_ldeor)
+#define LDNM	ldeor
+#define OP	eor
+#elif defined(L_ldset)
+#define LDNM	ldset
+#define OP	orr
+#else
+#error
+#endif
+#define LDOP	glue4(LDNM, A, L, S)
+
+STARTFN	NAME(LDNM)
+	JUMP_IF_NOT_LSE	8f
+
+	LDOP		s(0), s(0), [x1]
+	ret
+
+8:	mov		s(tmp0), s(0)
+0:	LDXR		s(0), [x1]
+	OP		s(tmp1), s(0), s(tmp0)
+	STXR		w(tmp1), s(tmp1), [x1]
+	cbnz		w(tmp1), 0b
+	ret
+
+ENDFN	NAME(LDNM)
+#endif
diff --git a/libgcc/config/aarch64/t-lse b/libgcc/config/aarch64/t-lse
new file mode 100644
index 00000000000..fe3868dacbf
--- /dev/null
+++ b/libgcc/config/aarch64/t-lse
@@ -0,0 +1,44 @@
+# Out-of-line LSE atomics for AArch64 architecture.
+# Copyright (C) 2019 Free Software Foundation, Inc.
+# Contributed by Linaro Ltd.
+#
+# This file is part of GCC.
+#
+# GCC is free software; you can redistribute it and/or modify it
+# under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3, or (at your option)
+# any later version.
+#
+# GCC is distributed in the hope that it will be useful, but
+# WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+# General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with GCC; see the file COPYING3.  If not see
+# <http://www.gnu.org/licenses/>.
+
+# Compare-and-swap has 5 sizes and 4 memory models.
+S0 := $(foreach s, 1 2 4 8 16, $(addsuffix _$(s), cas))
+O0 := $(foreach m, 1 2 3 4, $(addsuffix _$(m)$(objext), $(S0)))
+
+# Swap, Load-and-operate have 4 sizes and 4 memory models
+S1 := $(foreach s, 1 2 4 8, $(addsuffix _$(s), swp ldadd ldclr ldeor ldset))
+O1 := $(foreach m, 1 2 3 4, $(addsuffix _$(m)$(objext), $(S1)))
+
+LSE_OBJS := $(O0) $(O1)
+
+libgcc-objects += $(LSE_OBJS) lse-init$(objext)
+
+empty      =
+space      = $(empty) $(empty)
+PAT_SPLIT  = $(subst _,$(space),$(*F))
+PAT_BASE   = $(word 1,$(PAT_SPLIT))
+PAT_N      = $(word 2,$(PAT_SPLIT))
+PAT_M      = $(word 3,$(PAT_SPLIT))
+
+lse-init$(objext): $(srcdir)/config/aarch64/lse-init.c
+	$(gcc_compile) -c $<
+
+$(LSE_OBJS): $(srcdir)/config/aarch64/lse.S
+	$(gcc_compile) -DL_$(PAT_BASE) -DSIZE=$(PAT_N) -DMODEL=$(PAT_M) -c $<
-- 
2.26.2

From 96202f3214509cdc52a10d032d392d797ab93330 Mon Sep 17 00:00:00 2001
From: Richard Henderson <richard.henderson@linaro.org>
Date: Wed, 31 Oct 2018 23:12:14 +0000
Subject: [PATCH 14/24] Add visibility to libfunc constructors
To: gcc-patches@gcc.gnu.org

2018-10-31  Richard Henderson  <richard.henderson@linaro.org>

gcc/
	* optabs-libfuncs.c (build_libfunc_function_visibility):
	New, split out from...
	(build_libfunc_function): ... here.
	(init_one_libfunc_visibility): New, split out from ...
	(init_one_libfunc): ... here.

(cherry picked from commit a3ace685830da611697d0b9721ca675f3ae13766)
---
 gcc/optabs-libfuncs.c | 26 ++++++++++++++++++++------
 gcc/optabs-libfuncs.h |  2 ++
 2 files changed, 22 insertions(+), 6 deletions(-)

diff --git a/gcc/optabs-libfuncs.c b/gcc/optabs-libfuncs.c
index a3a32b7652c..1cbda947fb7 100644
--- a/gcc/optabs-libfuncs.c
+++ b/gcc/optabs-libfuncs.c
@@ -726,10 +726,10 @@ struct libfunc_decl_hasher : ggc_ptr_hash<tree_node>
 /* A table of previously-created libfuncs, hashed by name.  */
 static GTY (()) hash_table<libfunc_decl_hasher> *libfunc_decls;
 
-/* Build a decl for a libfunc named NAME.  */
+/* Build a decl for a libfunc named NAME with visibility VIS.  */
 
 tree
-build_libfunc_function (const char *name)
+build_libfunc_function_visibility (const char *name, symbol_visibility vis)
 {
   /* ??? We don't have any type information; pretend this is "int foo ()".  */
   tree decl = build_decl (UNKNOWN_LOCATION, FUNCTION_DECL,
@@ -738,7 +738,7 @@ build_libfunc_function (const char *name)
   DECL_EXTERNAL (decl) = 1;
   TREE_PUBLIC (decl) = 1;
   DECL_ARTIFICIAL (decl) = 1;
-  DECL_VISIBILITY (decl) = VISIBILITY_DEFAULT;
+  DECL_VISIBILITY (decl) = vis;
   DECL_VISIBILITY_SPECIFIED (decl) = 1;
   gcc_assert (DECL_ASSEMBLER_NAME (decl));
 
@@ -749,11 +749,19 @@ build_libfunc_function (const char *name)
   return decl;
 }
 
+/* Build a decl for a libfunc named NAME.  */
+
+tree
+build_libfunc_function (const char *name)
+{
+  return build_libfunc_function_visibility (name, VISIBILITY_DEFAULT);
+}
+
 /* Return a libfunc for NAME, creating one if we don't already have one.
-   The returned rtx is a SYMBOL_REF.  */
+   The decl is given visibility VIS.  The returned rtx is a SYMBOL_REF.  */
 
 rtx
-init_one_libfunc (const char *name)
+init_one_libfunc_visibility (const char *name, symbol_visibility vis)
 {
   tree id, decl;
   hashval_t hash;
@@ -770,12 +778,18 @@ init_one_libfunc (const char *name)
     {
       /* Create a new decl, so that it can be passed to
 	 targetm.encode_section_info.  */
-      decl = build_libfunc_function (name);
+      decl = build_libfunc_function_visibility (name, vis);
       *slot = decl;
     }
   return XEXP (DECL_RTL (decl), 0);
 }
 
+rtx
+init_one_libfunc (const char *name)
+{
+  return init_one_libfunc_visibility (name, VISIBILITY_DEFAULT);
+}
+
 /* Adjust the assembler name of libfunc NAME to ASMSPEC.  */
 
 rtx
diff --git a/gcc/optabs-libfuncs.h b/gcc/optabs-libfuncs.h
index a271d2dee5b..af06d6100a8 100644
--- a/gcc/optabs-libfuncs.h
+++ b/gcc/optabs-libfuncs.h
@@ -63,7 +63,9 @@ void gen_satfract_conv_libfunc (convert_optab, const char *,
 void gen_satfractuns_conv_libfunc (convert_optab, const char *,
 				   machine_mode, machine_mode);
 
+tree build_libfunc_function_visibility (const char *, symbol_visibility);
 tree build_libfunc_function (const char *);
+rtx init_one_libfunc_visibility (const char *, symbol_visibility);
 rtx init_one_libfunc (const char *);
 rtx set_user_assembler_libfunc (const char *, const char *);
 
-- 
2.26.2

From ab5bf64b93983d41d5619e0f2024b76913b812db Mon Sep 17 00:00:00 2001
From: Andrew Pinski <apinski@cavium.com>
Date: Sun, 6 Aug 2017 18:10:58 +0000
Subject: [PATCH 15/24] atomic_cmp_exchange_zero_reg_1.c: Pass
 -march=armv8-a+nolse, skip if -mcpu= is passed.
To: gcc-patches@gcc.gnu.org

2017-08-06  Andrew Pinski  <apinski@cavium.com>

gcc/testsuite/
        * gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c: Pass
        -march=armv8-a+nolse, skip if -mcpu= is passed.
        * gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c: Likewise.

(cherry picked from commit 905964740f674a784224620d1339676448aaada6)
---
 .../gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c        | 3 ++-
 .../gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c     | 3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c b/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c
index 15606b68990..f2a21ddf2e1 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c
@@ -1,5 +1,6 @@
 /* { dg-do compile } */
-/* { dg-options "-O2" } */
+/* { dg-options "-O2 -march=armv8-a+nolse" } */
+/* { dg-skip-if "" { *-*-* } { "-mcpu=*" } { "" } } */
 
 int
 foo (int *a)
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c b/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c
index b14a7c29437..8d2ae67dfbe 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c
@@ -1,5 +1,6 @@
 /* { dg-do compile } */
-/* { dg-options "-O2" } */
+/* { dg-options "-O2 -march=armv8-a+nolse" } */
+/* { dg-skip-if "" { *-*-* } { "-mcpu=*" } { "" } } */
 
 int
 foo (int *a)
-- 
2.26.2

From 4547522bebc79ece2ebc505116a81f47c423743a Mon Sep 17 00:00:00 2001
From: Richard Henderson <richard.henderson@linaro.org>
Date: Thu, 19 Sep 2019 14:36:43 +0000
Subject: [PATCH 16/24] aarch64: Implement -moutline-atomics
To: gcc-patches@gcc.gnu.org

2019-09-19  Richard Henderson  <richard.henderson@linaro.org>

gcc/
	* config/aarch64/aarch64.opt (-moutline-atomics): New.
	* config/aarch64/aarch64.c (aarch64_atomic_ool_func): New.
	(aarch64_ool_cas_names, aarch64_ool_swp_names): New.
	(aarch64_ool_ldadd_names, aarch64_ool_ldset_names): New.
	(aarch64_ool_ldclr_names, aarch64_ool_ldeor_names): New.
	(aarch64_expand_compare_and_swap): Honor TARGET_OUTLINE_ATOMICS.
	* config/aarch64/atomics.md (atomic_exchange<ALLI>): Likewise.
	(atomic_<atomic_op><ALLI>): Likewise.
	(atomic_fetch_<atomic_op><ALLI>): Likewise.
	(atomic_<atomic_op>_fetch<ALLI>): Likewise.
	* doc/invoke.texi: Document -moutline-atomics.

gcc/testsuite/
	* gcc.target/aarch64/atomic-op-acq_rel.c: Use -mno-outline-atomics.
	* gcc.target/aarch64/atomic-comp-swap-release-acquire.c: Likewise.
	* gcc.target/aarch64/atomic-op-acquire.c: Likewise.
	* gcc.target/aarch64/atomic-op-char.c: Likewise.
	* gcc.target/aarch64/atomic-op-consume.c: Likewise.
	* gcc.target/aarch64/atomic-op-imm.c: Likewise.
	* gcc.target/aarch64/atomic-op-int.c: Likewise.
	* gcc.target/aarch64/atomic-op-long.c: Likewise.
	* gcc.target/aarch64/atomic-op-relaxed.c: Likewise.
	* gcc.target/aarch64/atomic-op-release.c: Likewise.
	* gcc.target/aarch64/atomic-op-seq_cst.c: Likewise.
	* gcc.target/aarch64/atomic-op-short.c: Likewise.
	* gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c: Likewise.
	* gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c: Likewise.
	* gcc.target/aarch64/sync-comp-swap.c: Likewise.
	* gcc.target/aarch64/sync-op-acquire.c: Likewise.
	* gcc.target/aarch64/sync-op-full.c: Likewise.

(cherry picked from commit 3950b229a5ed6710f30241c2ddc3c74909bf4740)
---
 gcc/config/aarch64/aarch64-protos.h           | 13 +++
 gcc/config/aarch64/aarch64.c                  | 87 +++++++++++++++++
 gcc/config/aarch64/aarch64.opt                |  4 +
 gcc/config/aarch64/atomics.md                 | 94 +++++++++++++++++--
 gcc/doc/invoke.texi                           | 15 ++-
 .../atomic-comp-swap-release-acquire.c        |  2 +-
 .../gcc.target/aarch64/atomic-op-acq_rel.c    |  2 +-
 .../gcc.target/aarch64/atomic-op-acquire.c    |  2 +-
 .../gcc.target/aarch64/atomic-op-char.c       |  2 +-
 .../gcc.target/aarch64/atomic-op-consume.c    |  2 +-
 .../gcc.target/aarch64/atomic-op-imm.c        |  2 +-
 .../gcc.target/aarch64/atomic-op-int.c        |  2 +-
 .../gcc.target/aarch64/atomic-op-long.c       |  2 +-
 .../gcc.target/aarch64/atomic-op-relaxed.c    |  2 +-
 .../gcc.target/aarch64/atomic-op-release.c    |  2 +-
 .../gcc.target/aarch64/atomic-op-seq_cst.c    |  2 +-
 .../gcc.target/aarch64/atomic-op-short.c      |  2 +-
 .../aarch64/atomic_cmp_exchange_zero_reg_1.c  |  2 +-
 .../atomic_cmp_exchange_zero_strong_1.c       |  2 +-
 .../gcc.target/aarch64/sync-comp-swap.c       |  2 +-
 .../gcc.target/aarch64/sync-op-acquire.c      |  2 +-
 .../gcc.target/aarch64/sync-op-full.c         |  2 +-
 22 files changed, 221 insertions(+), 26 deletions(-)

diff --git a/gcc/config/aarch64/aarch64-protos.h b/gcc/config/aarch64/aarch64-protos.h
index e47f2174479..e5ab894ddc7 100644
--- a/gcc/config/aarch64/aarch64-protos.h
+++ b/gcc/config/aarch64/aarch64-protos.h
@@ -472,4 +472,17 @@ std::string aarch64_get_extension_string_for_isa_flags (unsigned long,
 
 rtl_opt_pass *make_pass_fma_steering (gcc::context *ctxt);
 
+struct atomic_ool_names
+{
+    const char *str[5][4];
+};
+
+rtx aarch64_atomic_ool_func(machine_mode mode, rtx model_rtx,
+			    const atomic_ool_names *names);
+extern const atomic_ool_names aarch64_ool_swp_names;
+extern const atomic_ool_names aarch64_ool_ldadd_names;
+extern const atomic_ool_names aarch64_ool_ldset_names;
+extern const atomic_ool_names aarch64_ool_ldclr_names;
+extern const atomic_ool_names aarch64_ool_ldeor_names;
+
 #endif /* GCC_AARCH64_PROTOS_H */
diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index 53de92e65fc..5ccb13b46fe 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -11923,6 +11923,82 @@ aarch64_emit_unlikely_jump (rtx insn)
   add_int_reg_note (jump, REG_BR_PROB, very_unlikely);
 }
 
+/* We store the names of the various atomic helpers in a 5x4 array.
+   Return the libcall function given MODE, MODEL and NAMES.  */
+
+rtx
+aarch64_atomic_ool_func(machine_mode mode, rtx model_rtx,
+			const atomic_ool_names *names)
+{
+  memmodel model = memmodel_base (INTVAL (model_rtx));
+  int mode_idx, model_idx;
+
+  switch (mode)
+    {
+    case QImode:
+      mode_idx = 0;
+      break;
+    case HImode:
+      mode_idx = 1;
+      break;
+    case SImode:
+      mode_idx = 2;
+      break;
+    case DImode:
+      mode_idx = 3;
+      break;
+    case TImode:
+      mode_idx = 4;
+      break;
+    default:
+      gcc_unreachable ();
+    }
+
+  switch (model)
+    {
+    case MEMMODEL_RELAXED:
+      model_idx = 0;
+      break;
+    case MEMMODEL_CONSUME:
+    case MEMMODEL_ACQUIRE:
+      model_idx = 1;
+      break;
+    case MEMMODEL_RELEASE:
+      model_idx = 2;
+      break;
+    case MEMMODEL_ACQ_REL:
+    case MEMMODEL_SEQ_CST:
+      model_idx = 3;
+      break;
+    default:
+      gcc_unreachable ();
+    }
+
+  return init_one_libfunc_visibility (names->str[mode_idx][model_idx],
+				      VISIBILITY_HIDDEN);
+}
+
+#define DEF0(B, N) \
+  { "__aarch64_" #B #N "_relax", \
+    "__aarch64_" #B #N "_acq", \
+    "__aarch64_" #B #N "_rel", \
+    "__aarch64_" #B #N "_acq_rel" }
+
+#define DEF4(B)  DEF0(B, 1), DEF0(B, 2), DEF0(B, 4), DEF0(B, 8), \
+		 { NULL, NULL, NULL, NULL }
+#define DEF5(B)  DEF0(B, 1), DEF0(B, 2), DEF0(B, 4), DEF0(B, 8), DEF0(B, 16)
+
+static const atomic_ool_names aarch64_ool_cas_names = { { DEF5(cas) } };
+const atomic_ool_names aarch64_ool_swp_names = { { DEF4(swp) } };
+const atomic_ool_names aarch64_ool_ldadd_names = { { DEF4(ldadd) } };
+const atomic_ool_names aarch64_ool_ldset_names = { { DEF4(ldset) } };
+const atomic_ool_names aarch64_ool_ldclr_names = { { DEF4(ldclr) } };
+const atomic_ool_names aarch64_ool_ldeor_names = { { DEF4(ldeor) } };
+
+#undef DEF0
+#undef DEF4
+#undef DEF5
+
 /* Expand a compare and swap pattern.  */
 
 void
@@ -11997,6 +12073,17 @@ aarch64_expand_compare_and_swap (rtx operands[])
       emit_insn (atomic_cas[idx] (rval, mem, newval, mod_s));
       cc_reg = aarch64_gen_compare_reg_maybe_ze (NE, rval, oldval, mode);
     }
+  else if (TARGET_OUTLINE_ATOMICS)
+    {
+      /* Oldval must satisfy compare afterward.  */
+      if (!aarch64_plus_operand (oldval, mode))
+	oldval = force_reg (mode, oldval);
+      rtx func = aarch64_atomic_ool_func (mode, mod_s, &aarch64_ool_cas_names);
+      rval = emit_library_call_value (func, NULL_RTX, LCT_NORMAL, r_mode,
+				      3, oldval, mode, newval, mode,
+				      XEXP (mem, 0), Pmode);
+      cc_reg = aarch64_gen_compare_reg_maybe_ze (NE, rval, oldval, mode);
+    }
   else
     {
       /* The oldval predicate varies by mode.  Test it and force to reg.  */
diff --git a/gcc/config/aarch64/aarch64.opt b/gcc/config/aarch64/aarch64.opt
index 942a7d558f2..edfb1b92be1 100644
--- a/gcc/config/aarch64/aarch64.opt
+++ b/gcc/config/aarch64/aarch64.opt
@@ -188,3 +188,7 @@ single precision and to 32 bits for double precision.
 mverbose-cost-dump
 Common Undocumented Var(flag_aarch64_verbose_cost)
 Enables verbose cost model dumping in the debug dump files.
+
+moutline-atomics
+Target Report Mask(OUTLINE_ATOMICS) Save
+Generate local calls to out-of-line atomic operations.
diff --git a/gcc/config/aarch64/atomics.md b/gcc/config/aarch64/atomics.md
index f32f5528f1a..23333a803ec 100644
--- a/gcc/config/aarch64/atomics.md
+++ b/gcc/config/aarch64/atomics.md
@@ -186,16 +186,27 @@
   (match_operand:SI 3 "const_int_operand" "")]
   ""
   {
-    rtx (*gen) (rtx, rtx, rtx, rtx);
-
     /* Use an atomic SWP when available.  */
     if (TARGET_LSE)
-      gen = gen_aarch64_atomic_exchange<mode>_lse;
+      {
+	emit_insn (gen_aarch64_atomic_exchange<mode>_lse
+		   (operands[0], operands[1], operands[2], operands[3]));
+      }
+    else if (TARGET_OUTLINE_ATOMICS)
+      {
+	machine_mode mode = <MODE>mode;
+	rtx func = aarch64_atomic_ool_func (mode, operands[3],
+					    &aarch64_ool_swp_names);
+	rtx rval = emit_library_call_value (func, operands[0], LCT_NORMAL,
+					    mode, 2, operands[2], mode,
+					    XEXP (operands[1], 0), Pmode);
+        emit_move_insn (operands[0], rval);
+      }
     else
-      gen = gen_aarch64_atomic_exchange<mode>;
-
-    emit_insn (gen (operands[0], operands[1], operands[2], operands[3]));
-
+      {
+	emit_insn (gen_aarch64_atomic_exchange<mode>
+		   (operands[0], operands[1], operands[2], operands[3]));
+      }
     DONE;
   }
 )
@@ -280,6 +291,39 @@
 	  }
 	operands[1] = force_reg (<MODE>mode, operands[1]);
       }
+    else if (TARGET_OUTLINE_ATOMICS)
+      {
+        const atomic_ool_names *names;
+	switch (<CODE>)
+	  {
+	  case MINUS:
+	    operands[1] = expand_simple_unop (<MODE>mode, NEG, operands[1],
+					      NULL, 1);
+	    /* fallthru */
+	  case PLUS:
+	    names = &aarch64_ool_ldadd_names;
+	    break;
+	  case IOR:
+	    names = &aarch64_ool_ldset_names;
+	    break;
+	  case XOR:
+	    names = &aarch64_ool_ldeor_names;
+	    break;
+	  case AND:
+	    operands[1] = expand_simple_unop (<MODE>mode, NOT, operands[1],
+					      NULL, 1);
+	    names = &aarch64_ool_ldclr_names;
+	    break;
+	  default:
+	    gcc_unreachable ();
+	  }
+        machine_mode mode = <MODE>mode;
+	rtx func = aarch64_atomic_ool_func (mode, operands[2], names);
+	emit_library_call_value (func, NULL_RTX, LCT_NORMAL, mode, 2,
+				 operands[1], mode,
+				 XEXP (operands[0], 0), Pmode);
+        DONE;
+      }
     else
       gen = gen_aarch64_atomic_<atomic_optab><mode>;
 
@@ -405,6 +449,40 @@
 	}
       operands[2] = force_reg (<MODE>mode, operands[2]);
     }
+  else if (TARGET_OUTLINE_ATOMICS)
+    {
+      const atomic_ool_names *names;
+      switch (<CODE>)
+	{
+	case MINUS:
+	  operands[2] = expand_simple_unop (<MODE>mode, NEG, operands[2],
+					    NULL, 1);
+	  /* fallthru */
+	case PLUS:
+	  names = &aarch64_ool_ldadd_names;
+	  break;
+	case IOR:
+	  names = &aarch64_ool_ldset_names;
+	  break;
+	case XOR:
+	  names = &aarch64_ool_ldeor_names;
+	  break;
+	case AND:
+	  operands[2] = expand_simple_unop (<MODE>mode, NOT, operands[2],
+					    NULL, 1);
+	  names = &aarch64_ool_ldclr_names;
+	  break;
+	default:
+	  gcc_unreachable ();
+	}
+      machine_mode mode = <MODE>mode;
+      rtx func = aarch64_atomic_ool_func (mode, operands[3], names);
+      rtx rval = emit_library_call_value (func, operands[0], LCT_NORMAL, mode,
+					  2, operands[2], mode,
+					  XEXP (operands[1], 0), Pmode);
+      emit_move_insn (operands[0], rval);
+      DONE;
+    }
   else
     gen = gen_aarch64_atomic_fetch_<atomic_optab><mode>;
 
@@ -494,7 +572,7 @@
 {
   /* Use an atomic load-operate instruction when possible.  In this case
      we will re-compute the result from the original mem value. */
-  if (TARGET_LSE)
+  if (TARGET_LSE || TARGET_OUTLINE_ATOMICS)
     {
       rtx tmp = gen_reg_rtx (<MODE>mode);
       operands[2] = force_reg (<MODE>mode, operands[2]);
diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
index d941c7b1ed4..71b8e92cf90 100644
--- a/gcc/doc/invoke.texi
+++ b/gcc/doc/invoke.texi
@@ -1042,7 +1042,7 @@ See RS/6000 and PowerPC Options.
 -mfloat128  -mno-float128  -mfloat128-hardware  -mno-float128-hardware @gol
 -mgnu-attribute  -mno-gnu-attribute @gol
 -mstack-protector-guard=@var{guard} -mstack-protector-guard-reg=@var{reg} @gol
--mstack-protector-guard-offset=@var{offset} @gol
+-mstack-protector-guard-offset=@var{offset} -moutline-atomics @gol
 -mlra  -mno-lra}
 
 @emph{RX Options}
@@ -13955,6 +13955,19 @@ This option only has an effect if @option{-ffast-math} or
 precision of division results to about 16 bits for
 single precision and to 32 bits for double precision.
 
+@item -moutline-atomics
+@itemx -mno-outline-atomics
+Enable or disable calls to out-of-line helpers to implement atomic operations.
+These helpers will, at runtime, determine if the LSE instructions from
+ARMv8.1-A can be used; if not, they will use the load/store-exclusive
+instructions that are present in the base ARMv8.0 ISA.
+
+This option is only applicable when compiling for the base ARMv8.0
+instruction set.  If using a later revision, e.g. @option{-march=armv8.1-a}
+or @option{-march=armv8-a+lse}, the ARMv8.1-Atomics instructions will be
+used directly.  The same applies when using @option{-mcpu=} when the
+selected cpu supports the @samp{lse} feature.
+
 @item -march=@var{name}
 @opindex march
 Specify the name of the target architecture and, optionally, one or
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic-comp-swap-release-acquire.c b/gcc/testsuite/gcc.target/aarch64/atomic-comp-swap-release-acquire.c
index 49ca5d0d09c..a828a72aa75 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic-comp-swap-release-acquire.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic-comp-swap-release-acquire.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2 -fno-ipa-icf" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -fno-ipa-icf -mno-outline-atomics" } */
 
 #include "atomic-comp-swap-release-acquire.x"
 
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic-op-acq_rel.c b/gcc/testsuite/gcc.target/aarch64/atomic-op-acq_rel.c
index 74f26348e42..6823ce381b2 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic-op-acq_rel.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic-op-acq_rel.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 #include "atomic-op-acq_rel.x"
 
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic-op-acquire.c b/gcc/testsuite/gcc.target/aarch64/atomic-op-acquire.c
index 66c1b1efe20..87937de378a 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic-op-acquire.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic-op-acquire.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 #include "atomic-op-acquire.x"
 
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic-op-char.c b/gcc/testsuite/gcc.target/aarch64/atomic-op-char.c
index c09d0434ecf..60955e57da3 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic-op-char.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic-op-char.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 #include "atomic-op-char.x"
 
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic-op-consume.c b/gcc/testsuite/gcc.target/aarch64/atomic-op-consume.c
index 5783ab84f5c..16cb11aeeaf 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic-op-consume.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic-op-consume.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 #include "atomic-op-consume.x"
 
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic-op-imm.c b/gcc/testsuite/gcc.target/aarch64/atomic-op-imm.c
index 18b8f0b04e9..bcab4e481e3 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic-op-imm.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic-op-imm.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 int v = 0;
 
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic-op-int.c b/gcc/testsuite/gcc.target/aarch64/atomic-op-int.c
index 8520f0839ba..040e4a8d168 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic-op-int.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic-op-int.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 #include "atomic-op-int.x"
 
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic-op-long.c b/gcc/testsuite/gcc.target/aarch64/atomic-op-long.c
index d011f8c5ce2..fc88b92cd3e 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic-op-long.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic-op-long.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 long v = 0;
 
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic-op-relaxed.c b/gcc/testsuite/gcc.target/aarch64/atomic-op-relaxed.c
index ed96bfdb978..503d62b0280 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic-op-relaxed.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic-op-relaxed.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 #include "atomic-op-relaxed.x"
 
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic-op-release.c b/gcc/testsuite/gcc.target/aarch64/atomic-op-release.c
index fc4be17de89..efe14aea7e4 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic-op-release.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic-op-release.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 #include "atomic-op-release.x"
 
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic-op-seq_cst.c b/gcc/testsuite/gcc.target/aarch64/atomic-op-seq_cst.c
index 613000fe490..09973bf82ba 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic-op-seq_cst.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic-op-seq_cst.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 #include "atomic-op-seq_cst.x"
 
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic-op-short.c b/gcc/testsuite/gcc.target/aarch64/atomic-op-short.c
index e82c8118ece..e1dcebb0f89 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic-op-short.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic-op-short.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 #include "atomic-op-short.x"
 
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c b/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c
index f2a21ddf2e1..29246979bfb 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_reg_1.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-O2 -march=armv8-a+nolse" } */
+/* { dg-options "-O2 -march=armv8-a+nolse -mno-outline-atomics" } */
 /* { dg-skip-if "" { *-*-* } { "-mcpu=*" } { "" } } */
 
 int
diff --git a/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c b/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c
index 8d2ae67dfbe..6daf9b08f5a 100644
--- a/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c
+++ b/gcc/testsuite/gcc.target/aarch64/atomic_cmp_exchange_zero_strong_1.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-O2 -march=armv8-a+nolse" } */
+/* { dg-options "-O2 -march=armv8-a+nolse -mno-outline-atomics" } */
 /* { dg-skip-if "" { *-*-* } { "-mcpu=*" } { "" } } */
 
 int
diff --git a/gcc/testsuite/gcc.target/aarch64/sync-comp-swap.c b/gcc/testsuite/gcc.target/aarch64/sync-comp-swap.c
index e571b2f13b3..f56415f3354 100644
--- a/gcc/testsuite/gcc.target/aarch64/sync-comp-swap.c
+++ b/gcc/testsuite/gcc.target/aarch64/sync-comp-swap.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2 -fno-ipa-icf" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -fno-ipa-icf -mno-outline-atomics" } */
 
 #include "sync-comp-swap.x"
 
diff --git a/gcc/testsuite/gcc.target/aarch64/sync-op-acquire.c b/gcc/testsuite/gcc.target/aarch64/sync-op-acquire.c
index 357bf1be3b2..39b3144aa36 100644
--- a/gcc/testsuite/gcc.target/aarch64/sync-op-acquire.c
+++ b/gcc/testsuite/gcc.target/aarch64/sync-op-acquire.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 #include "sync-op-acquire.x"
 
diff --git a/gcc/testsuite/gcc.target/aarch64/sync-op-full.c b/gcc/testsuite/gcc.target/aarch64/sync-op-full.c
index c6ba1629965..6b8b2043f40 100644
--- a/gcc/testsuite/gcc.target/aarch64/sync-op-full.c
+++ b/gcc/testsuite/gcc.target/aarch64/sync-op-full.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-march=armv8-a+nolse -O2" } */
+/* { dg-options "-march=armv8-a+nolse -O2 -mno-outline-atomics" } */
 
 #include "sync-op-full.x"
 
-- 
2.26.2

From 6da4552680e8269b677c08770797042f70ab8fdc Mon Sep 17 00:00:00 2001
From: Richard Henderson <richard.henderson@linaro.org>
Date: Wed, 25 Sep 2019 21:48:41 +0000
Subject: [PATCH 17/24] aarch64: Fix store-exclusive in load-operate LSE
 helpers
To: gcc-patches@gcc.gnu.org

2019-09-25  Richard Henderson  <richard.henderson@linaro.org>

gcc/
	PR target/91834
	* config/aarch64/lse.S (LDNM): Ensure STXR output does not
	overlap the inputs.

(cherry picked from commit 88a51d68c4aaa61adb36a9cad6f25ef214bde853)
---
 libgcc/config/aarch64/lse.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/libgcc/config/aarch64/lse.S b/libgcc/config/aarch64/lse.S
index a5f6673596c..c7979382ad7 100644
--- a/libgcc/config/aarch64/lse.S
+++ b/libgcc/config/aarch64/lse.S
@@ -227,8 +227,8 @@ STARTFN	NAME(LDNM)
 8:	mov		s(tmp0), s(0)
 0:	LDXR		s(0), [x1]
 	OP		s(tmp1), s(0), s(tmp0)
-	STXR		w(tmp1), s(tmp1), [x1]
-	cbnz		w(tmp1), 0b
+	STXR		w(tmp2), s(tmp1), [x1]
+	cbnz		w(tmp2), 0b
 	ret
 
 ENDFN	NAME(LDNM)
-- 
2.26.2

From a5f0d552ba6d5ce02b1c2f056116a0f8532b9667 Mon Sep 17 00:00:00 2001
From: Richard Henderson <richard.henderson@linaro.org>
Date: Wed, 25 Sep 2019 22:51:55 +0000
Subject: [PATCH 18/24] aarch64: Configure for sys/auxv.h in libgcc for
 lse-init.c
To: gcc-patches@gcc.gnu.org

2019-09-25  Richard Henderson  <richard.henderson@linaro.org>

gcc/
	PR target/91833
	* config/aarch64/lse-init.c: Include auto-target.h.  Disable
	initialization if !HAVE_SYS_AUXV_H.
	* configure.ac (AC_CHECK_HEADERS): Add sys/auxv.h.
	* config.in, configure: Rebuild.

(cherry picked from commit 58d169ba9ffca04d77314f525af9efd93881a86b)
---
 libgcc/config.in                 |  8 ++++++++
 libgcc/config/aarch64/lse-init.c |  4 +++-
 libgcc/configure                 | 26 +++++++++++++++++++-------
 libgcc/configure.ac              |  2 +-
 4 files changed, 31 insertions(+), 9 deletions(-)
 mode change 100644 => 100755 libgcc/configure

diff --git a/libgcc/config.in b/libgcc/config.in
index 7de22ee0a72..89d4aba3133 100644
--- a/libgcc/config.in
+++ b/libgcc/config.in
@@ -33,6 +33,9 @@
 /* Define to 1 if you have the <string.h> header file. */
 #undef HAVE_STRING_H
 
+/* Define to 1 if you have the <sys/auxv.h> header file. */
+#undef HAVE_SYS_AUXV_H
+
 /* Define to 1 if you have the <sys/stat.h> header file. */
 #undef HAVE_SYS_STAT_H
 
@@ -72,6 +75,11 @@
 /* Define to 1 if the target use emutls for thread-local storage. */
 #undef USE_EMUTLS
 
+/* Enable large inode numbers on Mac OS X 10.5.  */
+#ifndef _DARWIN_USE_64_BIT_INODE
+# define _DARWIN_USE_64_BIT_INODE 1
+#endif
+
 /* Number of bits in a file offset, on hosts where this is settable. */
 #undef _FILE_OFFSET_BITS
 
diff --git a/libgcc/config/aarch64/lse-init.c b/libgcc/config/aarch64/lse-init.c
index 33d29147479..1a8f4c55213 100644
--- a/libgcc/config/aarch64/lse-init.c
+++ b/libgcc/config/aarch64/lse-init.c
@@ -23,12 +23,14 @@ a copy of the GCC Runtime Library Exception along with this program;
 see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
 <http://www.gnu.org/licenses/>.  */
 
+#include "auto-target.h"
+
 /* Define the symbol gating the LSE implementations.  */
 _Bool __aarch64_have_lse_atomics
   __attribute__((visibility("hidden"), nocommon));
 
 /* Disable initialization of __aarch64_have_lse_atomics during bootstrap.  */
-#ifndef inhibit_libc
+#if !defined(inhibit_libc) && defined(HAVE_SYS_AUXV_H)
 # include <sys/auxv.h>
 
 /* Disable initialization if the system headers are too old.  */
diff --git a/libgcc/configure b/libgcc/configure
old mode 100644
new mode 100755
index 441601a1f76..59e15a7f33f
--- a/libgcc/configure
+++ b/libgcc/configure
@@ -640,6 +640,7 @@ infodir
 docdir
 oldincludedir
 includedir
+runstatedir
 localstatedir
 sharedstatedir
 sysconfdir
@@ -727,6 +728,7 @@ datadir='${datarootdir}'
 sysconfdir='${prefix}/etc'
 sharedstatedir='${prefix}/com'
 localstatedir='${prefix}/var'
+runstatedir='${localstatedir}/run'
 includedir='${prefix}/include'
 oldincludedir='/usr/include'
 docdir='${datarootdir}/doc/${PACKAGE_TARNAME}'
@@ -978,6 +980,15 @@ do
   | -silent | --silent | --silen | --sile | --sil)
     silent=yes ;;
 
+  -runstatedir | --runstatedir | --runstatedi | --runstated \
+  | --runstate | --runstat | --runsta | --runst | --runs \
+  | --run | --ru | --r)
+    ac_prev=runstatedir ;;
+  -runstatedir=* | --runstatedir=* | --runstatedi=* | --runstated=* \
+  | --runstate=* | --runstat=* | --runsta=* | --runst=* | --runs=* \
+  | --run=* | --ru=* | --r=*)
+    runstatedir=$ac_optarg ;;
+
   -sbindir | --sbindir | --sbindi | --sbind | --sbin | --sbi | --sb)
     ac_prev=sbindir ;;
   -sbindir=* | --sbindir=* | --sbindi=* | --sbind=* | --sbin=* \
@@ -1115,7 +1126,7 @@ fi
 for ac_var in	exec_prefix prefix bindir sbindir libexecdir datarootdir \
 		datadir sysconfdir sharedstatedir localstatedir includedir \
 		oldincludedir docdir infodir htmldir dvidir pdfdir psdir \
-		libdir localedir mandir
+		libdir localedir mandir runstatedir
 do
   eval ac_val=\$$ac_var
   # Remove trailing slashes.
@@ -1270,6 +1281,7 @@ Fine tuning of the installation directories:
   --sysconfdir=DIR        read-only single-machine data [PREFIX/etc]
   --sharedstatedir=DIR    modifiable architecture-independent data [PREFIX/com]
   --localstatedir=DIR     modifiable single-machine data [PREFIX/var]
+  --runstatedir=DIR       modifiable per-process data [LOCALSTATEDIR/run]
   --libdir=DIR            object code libraries [EPREFIX/lib]
   --includedir=DIR        C header files [PREFIX/include]
   --oldincludedir=DIR     C header files for non-gcc [/usr/include]
@@ -4088,7 +4100,7 @@ else
     We can't simply define LARGE_OFF_T to be 9223372036854775807,
     since some C++ compilers masquerading as C compilers
     incorrectly reject 9223372036854775807.  */
-#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
+#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
   int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
 		       && LARGE_OFF_T % 2147483647 == 1)
 		      ? 1 : -1];
@@ -4134,7 +4146,7 @@ else
     We can't simply define LARGE_OFF_T to be 9223372036854775807,
     since some C++ compilers masquerading as C compilers
     incorrectly reject 9223372036854775807.  */
-#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
+#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
   int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
 		       && LARGE_OFF_T % 2147483647 == 1)
 		      ? 1 : -1];
@@ -4158,7 +4170,7 @@ rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
     We can't simply define LARGE_OFF_T to be 9223372036854775807,
     since some C++ compilers masquerading as C compilers
     incorrectly reject 9223372036854775807.  */
-#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
+#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
   int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
 		       && LARGE_OFF_T % 2147483647 == 1)
 		      ? 1 : -1];
@@ -4203,7 +4215,7 @@ else
     We can't simply define LARGE_OFF_T to be 9223372036854775807,
     since some C++ compilers masquerading as C compilers
     incorrectly reject 9223372036854775807.  */
-#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
+#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
   int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
 		       && LARGE_OFF_T % 2147483647 == 1)
 		      ? 1 : -1];
@@ -4227,7 +4239,7 @@ rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
     We can't simply define LARGE_OFF_T to be 9223372036854775807,
     since some C++ compilers masquerading as C compilers
     incorrectly reject 9223372036854775807.  */
-#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
+#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
   int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
 		       && LARGE_OFF_T % 2147483647 == 1)
 		      ? 1 : -1];
@@ -4339,7 +4351,7 @@ as_fn_arith $ac_cv_sizeof_long_double \* 8 && long_double_type_size=$as_val
 
 for ac_header in inttypes.h stdint.h stdlib.h ftw.h \
 	unistd.h sys/stat.h sys/types.h \
-	string.h strings.h memory.h
+	string.h strings.h memory.h sys/auxv.h
 do :
   as_ac_Header=`$as_echo "ac_cv_header_$ac_header" | $as_tr_sh`
 ac_fn_c_check_header_preproc "$LINENO" "$ac_header" "$as_ac_Header"
diff --git a/libgcc/configure.ac b/libgcc/configure.ac
index 99b8e15562f..fdd25f295b1 100644
--- a/libgcc/configure.ac
+++ b/libgcc/configure.ac
@@ -202,7 +202,7 @@ AC_SUBST(long_double_type_size)
 
 AC_CHECK_HEADERS(inttypes.h stdint.h stdlib.h ftw.h \
 	unistd.h sys/stat.h sys/types.h \
-	string.h strings.h memory.h)
+	string.h strings.h memory.h sys/auxv.h)
 AC_HEADER_STDC
 
 # Check for decimal float support.
-- 
2.26.2

From b5727f45769817cb8b7628abf5898ef840983425 Mon Sep 17 00:00:00 2001
From: Jakub Jelinek <jakub@redhat.com>
Date: Tue, 31 Mar 2020 11:08:22 +0200
Subject: [PATCH 19/24] aarch64: Fix up aarch64_compare_and_swaphi pattern
 [PR94368]
To: gcc-patches@gcc.gnu.org

The following testcase ICEs in final_scan_insn_1.  The problem is in the
@aarch64_compare_and_swaphi define_insn_and_split, since 9 it uses
aarch64_plushi_operand predicate for the "expected value" operand, which
allows either 0..0xfff constants or 0x1000..0xf000 constants (i.e. HImode
values which when zero extended are either 0..0xfff or (0..0xfff) << 12).
The problem is that RA doesn't care about predicates, it honors just
constraints and the used constraint on the operand is n, which means any
HImode CONST_SCALAR_INT.  In the testcase LRA thus propagates the -1
value into the insn.
This is a define_insn_and_split which requires mandatory split.
But during split2 pass, we check the predicate (and don't check
constraints), which fails and thus we don't split it and during final ICE
because the mandatory splitting didn't happen.

The following patch fixes it by adding a matching constraint to the
predicate and using it.

2020-03-31  Jakub Jelinek  <jakub@redhat.com>

gcc/
	PR target/94368
	* config/aarch64/constraints.md (Uph): New constraint.
	* config/aarch64/atomics.md (cas_short_expected_imm): New mode attr.
	(@aarch64_compare_and_swap<mode>): Use it instead of n in operand 2's
	constraint.

gcc/testsuite/
	* gcc.dg/pr94368.c: New test.

(cherry picked from commit a27c534794dbe3530acae3427d2c58f937f1b050)
---
 gcc/config/aarch64/atomics.md     |  5 ++++-
 gcc/config/aarch64/constraints.md |  7 +++++++
 gcc/testsuite/gcc.dg/pr94368.c    | 25 +++++++++++++++++++++++++
 3 files changed, 36 insertions(+), 1 deletion(-)
 create mode 100644 gcc/testsuite/gcc.dg/pr94368.c

diff --git a/gcc/config/aarch64/atomics.md b/gcc/config/aarch64/atomics.md
index 23333a803ec..05ed4cee75b 100644
--- a/gcc/config/aarch64/atomics.md
+++ b/gcc/config/aarch64/atomics.md
@@ -38,6 +38,8 @@
 
 (define_mode_attr cas_short_expected_pred
   [(QI "aarch64_reg_or_imm") (HI "aarch64_plushi_operand")])
+(define_mode_attr cas_short_expected_imm
+  [(QI "n") (HI "Uph")])
 
 (define_insn_and_split "aarch64_compare_and_swap<mode>"
   [(set (reg:CC CC_REGNUM)					;; bool out
@@ -47,7 +49,8 @@
       (match_operand:SHORT 1 "aarch64_sync_memory_operand" "+Q"))) ;; memory
    (set (match_dup 1)
     (unspec_volatile:SHORT
-      [(match_operand:SHORT 2 "<cas_short_expected_pred>" "rn")	;; expected
+      [(match_operand:SHORT 2 "<cas_short_expected_pred>"
+			      "r<cas_short_expected_imm>")	;; expected
        (match_operand:SHORT 3 "aarch64_reg_or_zero" "rZ")	;; desired
        (match_operand:SI 4 "const_int_operand")			;; is_weak
        (match_operand:SI 5 "const_int_operand")			;; mod_s
diff --git a/gcc/config/aarch64/constraints.md b/gcc/config/aarch64/constraints.md
index 9b3c7339c54..8dfd9026752 100644
--- a/gcc/config/aarch64/constraints.md
+++ b/gcc/config/aarch64/constraints.md
@@ -142,6 +142,13 @@
   (and (match_code "const_int")
        (match_test "(unsigned) exact_log2 (ival) <= 4")))
 
+(define_constraint "Uph"
+  "@internal
+  A constraint that matches HImode integers zero extendable to
+  SImode plus_operand."
+  (and (match_code "const_int")
+       (match_test "aarch64_plushi_immediate (op, VOIDmode)")))
+
 (define_memory_constraint "Q"
  "A memory address which uses a single base register with no offset."
  (and (match_code "mem")
diff --git a/gcc/testsuite/gcc.dg/pr94368.c b/gcc/testsuite/gcc.dg/pr94368.c
new file mode 100644
index 00000000000..1267b822098
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/pr94368.c
@@ -0,0 +1,25 @@
+/* PR target/94368 */
+/* { dg-do compile { target fpic } } */
+/* { dg-options "-fpic -O1 -fcommon" } */
+
+int b, c, d, e, f, h;
+short g;
+int foo (int) __attribute__ ((__const__));
+
+void
+bar (void)
+{
+  while (1)
+    {
+      while (1)
+	{
+	  __atomic_load_n (&e, 0);
+	  if (foo (2))
+	    __sync_val_compare_and_swap (&c, 0, f);
+	  b = 1;
+	  if (h == e)
+	    break;
+	}
+      __sync_val_compare_and_swap (&g, -1, f);
+    }
+}
-- 
2.26.2

From e98ee125c439b20329010b8a2386a252e9ca20cf Mon Sep 17 00:00:00 2001
From: Jakub Jelinek <jakub@redhat.com>
Date: Wed, 15 Apr 2020 11:01:19 +0200
Subject: [PATCH 20/24] aarch64: Fix bootstrap with old binutils [PR93053]
To: gcc-patches@gcc.gnu.org

As reported in the PR, GCC 10 (and also 9.3.1 but not 9.3.0) fails to build
when using older binutils which lack LSE support, because those instructions
are used in libgcc.
Thanks to Kyrylo's hint, the following patches (hopefully) allow it to build
even with older binutils by using .inst directive if LSE support isn't
available in the assembler.

2020-04-15  Jakub Jelinek  <jakub@redhat.com>

gcc/
	PR target/93053
	* configure.ac (LIBGCC_CHECK_AS_LSE): Add HAVE_AS_LSE checking.
	* config/aarch64/lse.S: Include auto-target.h, if HAVE_AS_LSE
	is not defined, use just .arch armv8-a.
	(B, M, N, OPN): Define.
	(COMMENT): New .macro.
	(CAS, CASP, SWP, LDOP): Use .inst directive if HAVE_AS_LSE is not
	defined.  Otherwise, move the operands right after the glue? and
	comment out operands where the macros are used.
	* configure: Regenerated.
	* config.in: Regenerated.

(cherry picked from commit 5b2f76e36d861c881c6770b4f47c1fae6c0c8965)
---
 libgcc/config.in            |  8 ++---
 libgcc/config/aarch64/lse.S | 57 ++++++++++++++++++++++++++++-----
 libgcc/configure            | 64 ++++++++++++++++++++++++++-----------
 libgcc/configure.ac         | 19 +++++++++++
 4 files changed, 117 insertions(+), 31 deletions(-)

diff --git a/libgcc/config.in b/libgcc/config.in
index 89d4aba3133..9009790625f 100644
--- a/libgcc/config.in
+++ b/libgcc/config.in
@@ -1,5 +1,8 @@
 /* config.in.  Generated from configure.ac by autoheader.  */
 
+/* Define to 1 if the assembler supports LSE. */
+#undef HAVE_AS_LSE
+
 /* Define to 1 if the target assembler supports thread-local storage. */
 #undef HAVE_CC_TLS
 
@@ -75,11 +78,6 @@
 /* Define to 1 if the target use emutls for thread-local storage. */
 #undef USE_EMUTLS
 
-/* Enable large inode numbers on Mac OS X 10.5.  */
-#ifndef _DARWIN_USE_64_BIT_INODE
-# define _DARWIN_USE_64_BIT_INODE 1
-#endif
-
 /* Number of bits in a file offset, on hosts where this is settable. */
 #undef _FILE_OFFSET_BITS
 
diff --git a/libgcc/config/aarch64/lse.S b/libgcc/config/aarch64/lse.S
index c7979382ad7..f7f1c19587b 100644
--- a/libgcc/config/aarch64/lse.S
+++ b/libgcc/config/aarch64/lse.S
@@ -48,8 +48,14 @@ see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
  * separately to minimize code size.
  */
 
+#include "auto-target.h"
+
 /* Tell the assembler to accept LSE instructions.  */
+#ifdef HAVE_AS_LSE
 	.arch armv8-a+lse
+#else
+	.arch armv8-a
+#endif
 
 /* Declare the symbol gating the LSE implementations.  */
 	.hidden	__aarch64_have_lse_atomics
@@ -58,12 +64,19 @@ see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
 #if SIZE == 1
 # define S     b
 # define UXT   uxtb
+# define B     0x00000000
 #elif SIZE == 2
 # define S     h
 # define UXT   uxth
+# define B     0x40000000
 #elif SIZE == 4 || SIZE == 8 || SIZE == 16
 # define S
 # define UXT   mov
+# if SIZE == 4
+#  define B    0x80000000
+# elif SIZE == 8
+#  define B    0xc0000000
+# endif
 #else
 # error
 #endif
@@ -72,18 +85,26 @@ see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
 # define SUFF  _relax
 # define A
 # define L
+# define M     0x000000
+# define N     0x000000
 #elif MODEL == 2
 # define SUFF  _acq
 # define A     a
 # define L
+# define M     0x400000
+# define N     0x800000
 #elif MODEL == 3
 # define SUFF  _rel
 # define A
 # define L     l
+# define M     0x008000
+# define N     0x400000
 #elif MODEL == 4
 # define SUFF  _acq_rel
 # define A     a
 # define L     l
+# define M     0x408000
+# define N     0xc00000
 #else
 # error
 #endif
@@ -144,9 +165,13 @@ STARTFN	NAME(cas)
 	JUMP_IF_NOT_LSE	8f
 
 #if SIZE < 16
-#define CAS	glue4(cas, A, L, S)
+#ifdef HAVE_AS_LSE
+# define CAS	glue4(cas, A, L, S)	s(0), s(1), [x2]
+#else
+# define CAS	.inst 0x08a07c41 + B + M
+#endif
 
-	CAS		s(0), s(1), [x2]
+	CAS		/* s(0), s(1), [x2] */
 	ret
 
 8:	UXT		s(tmp0), s(0)
@@ -160,9 +185,13 @@ STARTFN	NAME(cas)
 #else
 #define LDXP	glue3(ld, A, xp)
 #define STXP	glue3(st, L, xp)
-#define CASP	glue3(casp, A, L)
+#ifdef HAVE_AS_LSE
+# define CASP	glue3(casp, A, L)	x0, x1, x2, x3, [x4]
+#else
+# define CASP	.inst 0x48207c82 + M
+#endif
 
-	CASP		x0, x1, x2, x3, [x4]
+	CASP		/* x0, x1, x2, x3, [x4] */
 	ret
 
 8:	mov		x(tmp0), x0
@@ -181,12 +210,16 @@ ENDFN	NAME(cas)
 #endif
 
 #ifdef L_swp
-#define SWP	glue4(swp, A, L, S)
+#ifdef HAVE_AS_LSE
+# define SWP	glue4(swp, A, L, S)	s(0), s(0), [x1]
+#else
+# define SWP	.inst 0x38208020 + B + N
+#endif
 
 STARTFN	NAME(swp)
 	JUMP_IF_NOT_LSE	8f
 
-	SWP		s(0), s(0), [x1]
+	SWP		/* s(0), s(0), [x1] */
 	ret
 
 8:	mov		s(tmp0), s(0)
@@ -204,24 +237,32 @@ ENDFN	NAME(swp)
 #ifdef L_ldadd
 #define LDNM	ldadd
 #define OP	add
+#define OPN	0x0000
 #elif defined(L_ldclr)
 #define LDNM	ldclr
 #define OP	bic
+#define OPN	0x1000
 #elif defined(L_ldeor)
 #define LDNM	ldeor
 #define OP	eor
+#define OPN	0x2000
 #elif defined(L_ldset)
 #define LDNM	ldset
 #define OP	orr
+#define OPN	0x3000
 #else
 #error
 #endif
-#define LDOP	glue4(LDNM, A, L, S)
+#ifdef HAVE_AS_LSE
+# define LDOP	glue4(LDNM, A, L, S)	s(0), s(0), [x1]
+#else
+# define LDOP	.inst 0x38200020 + OPN + B + N
+#endif
 
 STARTFN	NAME(LDNM)
 	JUMP_IF_NOT_LSE	8f
 
-	LDOP		s(0), s(0), [x1]
+	LDOP		/* s(0), s(0), [x1] */
 	ret
 
 8:	mov		s(tmp0), s(0)
diff --git a/libgcc/configure b/libgcc/configure
index 59e15a7f33f..8b6f38eeddd 100755
--- a/libgcc/configure
+++ b/libgcc/configure
@@ -640,7 +640,6 @@ infodir
 docdir
 oldincludedir
 includedir
-runstatedir
 localstatedir
 sharedstatedir
 sysconfdir
@@ -728,7 +727,6 @@ datadir='${datarootdir}'
 sysconfdir='${prefix}/etc'
 sharedstatedir='${prefix}/com'
 localstatedir='${prefix}/var'
-runstatedir='${localstatedir}/run'
 includedir='${prefix}/include'
 oldincludedir='/usr/include'
 docdir='${datarootdir}/doc/${PACKAGE_TARNAME}'
@@ -980,15 +978,6 @@ do
   | -silent | --silent | --silen | --sile | --sil)
     silent=yes ;;
 
-  -runstatedir | --runstatedir | --runstatedi | --runstated \
-  | --runstate | --runstat | --runsta | --runst | --runs \
-  | --run | --ru | --r)
-    ac_prev=runstatedir ;;
-  -runstatedir=* | --runstatedir=* | --runstatedi=* | --runstated=* \
-  | --runstate=* | --runstat=* | --runsta=* | --runst=* | --runs=* \
-  | --run=* | --ru=* | --r=*)
-    runstatedir=$ac_optarg ;;
-
   -sbindir | --sbindir | --sbindi | --sbind | --sbin | --sbi | --sb)
     ac_prev=sbindir ;;
   -sbindir=* | --sbindir=* | --sbindi=* | --sbind=* | --sbin=* \
@@ -1126,7 +1115,7 @@ fi
 for ac_var in	exec_prefix prefix bindir sbindir libexecdir datarootdir \
 		datadir sysconfdir sharedstatedir localstatedir includedir \
 		oldincludedir docdir infodir htmldir dvidir pdfdir psdir \
-		libdir localedir mandir runstatedir
+		libdir localedir mandir
 do
   eval ac_val=\$$ac_var
   # Remove trailing slashes.
@@ -1281,7 +1270,6 @@ Fine tuning of the installation directories:
   --sysconfdir=DIR        read-only single-machine data [PREFIX/etc]
   --sharedstatedir=DIR    modifiable architecture-independent data [PREFIX/com]
   --localstatedir=DIR     modifiable single-machine data [PREFIX/var]
-  --runstatedir=DIR       modifiable per-process data [LOCALSTATEDIR/run]
   --libdir=DIR            object code libraries [EPREFIX/lib]
   --includedir=DIR        C header files [PREFIX/include]
   --oldincludedir=DIR     C header files for non-gcc [/usr/include]
@@ -4100,7 +4088,7 @@ else
     We can't simply define LARGE_OFF_T to be 9223372036854775807,
     since some C++ compilers masquerading as C compilers
     incorrectly reject 9223372036854775807.  */
-#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
+#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
   int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
 		       && LARGE_OFF_T % 2147483647 == 1)
 		      ? 1 : -1];
@@ -4146,7 +4134,7 @@ else
     We can't simply define LARGE_OFF_T to be 9223372036854775807,
     since some C++ compilers masquerading as C compilers
     incorrectly reject 9223372036854775807.  */
-#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
+#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
   int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
 		       && LARGE_OFF_T % 2147483647 == 1)
 		      ? 1 : -1];
@@ -4170,7 +4158,7 @@ rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
     We can't simply define LARGE_OFF_T to be 9223372036854775807,
     since some C++ compilers masquerading as C compilers
     incorrectly reject 9223372036854775807.  */
-#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
+#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
   int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
 		       && LARGE_OFF_T % 2147483647 == 1)
 		      ? 1 : -1];
@@ -4215,7 +4203,7 @@ else
     We can't simply define LARGE_OFF_T to be 9223372036854775807,
     since some C++ compilers masquerading as C compilers
     incorrectly reject 9223372036854775807.  */
-#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
+#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
   int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
 		       && LARGE_OFF_T % 2147483647 == 1)
 		      ? 1 : -1];
@@ -4239,7 +4227,7 @@ rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
     We can't simply define LARGE_OFF_T to be 9223372036854775807,
     since some C++ compilers masquerading as C compilers
     incorrectly reject 9223372036854775807.  */
-#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
+#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
   int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
 		       && LARGE_OFF_T % 2147483647 == 1)
 		      ? 1 : -1];
@@ -5221,6 +5209,46 @@ if test "$enable_tls $gcc_cv_use_emutls" = "yes yes"; then
 fi
 
 
+
+
+case "${target}" in
+aarch64*-*-*)
+  { $as_echo "$as_me:${as_lineno-$LINENO}: checking if the assembler supports LSE" >&5
+$as_echo_n "checking if the assembler supports LSE... " >&6; }
+if test "${libgcc_cv_as_lse+set}" = set; then :
+  $as_echo_n "(cached) " >&6
+else
+
+    cat confdefs.h - <<_ACEOF >conftest.$ac_ext
+/* end confdefs.h.  */
+
+int
+main ()
+{
+			asm(".arch armv8-a+lse\n\tcas w0, w1, [x2]");
+
+  ;
+  return 0;
+}
+_ACEOF
+if ac_fn_c_try_compile "$LINENO"; then :
+  libgcc_cv_as_lse=yes
+else
+  libgcc_cv_as_lse=no
+fi
+rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
+
+fi
+{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $libgcc_cv_as_lse" >&5
+$as_echo "$libgcc_cv_as_lse" >&6; }
+  if test x$libgcc_cv_as_lse = xyes; then
+
+$as_echo "#define HAVE_AS_LSE 1" >>confdefs.h
+
+  fi
+  ;;
+esac
+
 { $as_echo "$as_me:${as_lineno-$LINENO}: checking for init priority support" >&5
 $as_echo_n "checking for init priority support... " >&6; }
 if test "${libgcc_cv_init_priority+set}" = set; then :
diff --git a/libgcc/configure.ac b/libgcc/configure.ac
index fdd25f295b1..708ed19f8e4 100644
--- a/libgcc/configure.ac
+++ b/libgcc/configure.ac
@@ -503,6 +503,25 @@ if test "$enable_tls $gcc_cv_use_emutls" = "yes yes"; then
 fi
 AC_SUBST(set_use_emutls)
 
+dnl Check if as supports LSE instructions.
+AC_DEFUN([LIBGCC_CHECK_AS_LSE], [
+case "${target}" in
+aarch64*-*-*)
+  AC_CACHE_CHECK([if the assembler supports LSE], libgcc_cv_as_lse, [
+    AC_TRY_COMPILE([],
+changequote(,)dnl
+			asm(".arch armv8-a+lse\n\tcas w0, w1, [x2]");
+changequote([,])dnl
+		       ,
+		   [libgcc_cv_as_lse=yes], [libgcc_cv_as_lse=no])
+  ])
+  if test x$libgcc_cv_as_lse = xyes; then
+    AC_DEFINE(HAVE_AS_LSE, 1, [Define to 1 if the assembler supports LSE.])
+  fi
+  ;;
+esac])
+LIBGCC_CHECK_AS_LSE
+
 AC_CACHE_CHECK(for init priority support, libgcc_cv_init_priority, [
 AC_COMPILE_IFELSE([AC_LANG_PROGRAM(,
   [[void ip (void) __attribute__ ((constructor (1)));]])],
-- 
2.26.2

From 37c58b17785af416b1d78ddd40ec3b1d394a584c Mon Sep 17 00:00:00 2001
From: Jakub Jelinek <jakub@redhat.com>
Date: Thu, 2 Apr 2020 12:57:11 +0200
Subject: [PATCH 21/24] aarch64: Fix ICE due to
 aarch64_gen_compare_reg_maybe_ze [PR94435]
To: gcc-patches@gcc.gnu.org

The following testcase ICEs, because aarch64_gen_compare_reg_maybe_ze emits
invalid RTL.
For y_mode [QH]Imode it expects y to be of that mode (or CONST_INT that fits
into that mode) and x being SImode; for non-CONST_INT y it zero extends y
into SImode and compares that against x, for CONST_INT y it zero extends y
into SImode.  The problem is that when the zero extended constant isn't
usable directly, it forces it into a REG, but with y_mode mode, and then
compares against y.  That is wrong, because it should force it into a SImode
REG and compare that way.

2020-04-02  Jakub Jelinek  <jakub@redhat.com>

gcc/
	PR target/94435
	* config/aarch64/aarch64.c (aarch64_gen_compare_reg_maybe_ze): For
	y_mode E_[QH]Imode and y being a CONST_INT, change y_mode to SImode.

gcc/testsuite/
	* gcc.target/aarch64/pr94435.c: New test.

(cherry picked from commit df562b12d90699c20923f91df48eed08ebcb572e)
---
 gcc/config/aarch64/aarch64.c               |  5 ++++-
 gcc/testsuite/gcc.target/aarch64/pr94435.c | 25 ++++++++++++++++++++++
 2 files changed, 29 insertions(+), 1 deletion(-)
 create mode 100644 gcc/testsuite/gcc.target/aarch64/pr94435.c

diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index 5ccb13b46fe..a0685a5ad41 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -1230,7 +1230,10 @@ aarch64_gen_compare_reg_maybe_ze (RTX_CODE code, rtx x, rtx y,
   if (y_mode == QImode || y_mode == HImode)
     {
       if (CONST_INT_P (y))
-	y = GEN_INT (INTVAL (y) & GET_MODE_MASK (y_mode));
+	{
+	  y = GEN_INT (INTVAL (y) & GET_MODE_MASK (y_mode));
+	  y_mode = SImode;
+	}
       else
 	{
 	  rtx t, cc_reg;
diff --git a/gcc/testsuite/gcc.target/aarch64/pr94435.c b/gcc/testsuite/gcc.target/aarch64/pr94435.c
new file mode 100644
index 00000000000..5713c14d5f9
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/pr94435.c
@@ -0,0 +1,25 @@
+/* PR target/94435 */
+/* { dg-do compile } */
+/* { dg-options "-march=armv8-a+nolse -moutline-atomics" } */
+
+int b, c, d, e, f, h;
+short g;
+int foo (int) __attribute__ ((__const__));
+
+void
+bar (void)
+{
+  while (1)
+    {
+      while (1)
+	{
+	  __atomic_load_n (&e, 0);
+	  if (foo (2))
+	    __sync_val_compare_and_swap (&c, 0, f);
+	  b = 1;
+	  if (h == e)
+	    break;
+	}
+      __sync_val_compare_and_swap (&g, -1, f);
+    }
+}
-- 
2.26.2

From 57927b2e7e2aa9c7ae703a110cc86dfc40e7659c Mon Sep 17 00:00:00 2001
From: Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
Date: Wed, 21 Aug 2019 18:34:43 +0000
Subject: [PATCH 22/24] re PR target/90724 (ICE with
 __sync_bool_compare_and_swap with -march=armv8.2-a+sve)
To: gcc-patches@gcc.gnu.org

2019-08-21  Prathamesh Kulkarni  <prathamesh.kulkarni@linaro.org>

gcc/
	PR target/90724
	* config/aarch64/aarch64.c (aarch64_gen_compare_reg_maybe_ze): Force y
	in reg if it fails aarch64_plus_operand predicate.

(cherry picked from commit 846f78d414101dbd33ff9c370d379bae73ae0efa)
---
 gcc/config/aarch64/aarch64.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index a0685a5ad41..9535d688ee5 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -1248,6 +1248,9 @@ aarch64_gen_compare_reg_maybe_ze (RTX_CODE code, rtx x, rtx y,
 	}
     }
 
+  if (!aarch64_plus_operand (y, y_mode))
+    y = force_reg (y_mode, y);
+
   return aarch64_gen_compare_reg (code, x, y);
 }
 
-- 
2.26.2

From d019077cc8d3c18bc510aeaf34c6c4532913f378 Mon Sep 17 00:00:00 2001
From: Andre Vieira <andre.simoesdiasvieira@arm.com>
Date: Tue, 28 Apr 2020 13:25:43 +0100
Subject: [PATCH 23/24] aarch64: Fix for PR target/94814
To: gcc-patches@gcc.gnu.org

Backport of PR target/94518: Fix memmodel index in aarch64_store_exclusive_pair

2020-04-28  Andre Vieira  <andre.simoesdiasvieira@arm.com>

PR target/94814
	Backport from gcc-9.
	2020-04-07  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>

PR target/94518
	2019-09-23  Richard Sandiford  <richard.sandiford@arm.com>

gcc/
	* config/aarch64/atomics.md (aarch64_store_exclusive_pair): Fix
	memmodel index.

(cherry picked from commit 3a30d2558b3a199fe346479e6140cddae7fba5ed)
---
 gcc/config/aarch64/atomics.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gcc/config/aarch64/atomics.md b/gcc/config/aarch64/atomics.md
index 05ed4cee75b..1fdc3092a33 100644
--- a/gcc/config/aarch64/atomics.md
+++ b/gcc/config/aarch64/atomics.md
@@ -752,7 +752,7 @@
 	  UNSPECV_SX))]
   ""
   {
-    enum memmodel model = memmodel_from_int (INTVAL (operands[3]));
+    enum memmodel model = memmodel_from_int (INTVAL (operands[4]));
     if (is_mm_relaxed (model) || is_mm_consume (model) || is_mm_acquire (model))
       return "stxp\t%w0, %x2, %x3, %1";
     else
-- 
2.26.2

From 7c216ba945cb92bd79fbe01b35e16bd1e3cd854d Mon Sep 17 00:00:00 2001
From: Andre Vieira <andre.simoesdiasvieira@arm.com>
Date: Wed, 29 Apr 2020 15:42:27 +0100
Subject: [PATCH 24/24] aarch64: Force TImode values into even registers
To: gcc-patches@gcc.gnu.org

The LSE CASP instruction requires values to be placed in even
register pairs.  A solution involving two additional register
classes was rejected in favor of the much simpler solution of
simply requiring all TImode values to be aligned.

gcc/ChangeLog:
2020-04-29  Andre Vieira  <andre.simoesdiasvieira@arm.com>

Backport from mainline.
	2018-10-31  Richard Henderson  <richard.henderson@linaro.org>

* config/aarch64/aarch64.c (aarch64_hard_regno_mode_ok): Force
	16-byte modes held in GP registers to use an even regno.

(cherry picked from commit 563cc649beaf11d707c422e5f4e9e5cdacb818c3)
---
 gcc/config/aarch64/aarch64.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index 9535d688ee5..643ba7e8153 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -1084,10 +1084,14 @@ aarch64_hard_regno_mode_ok (unsigned regno, machine_mode mode)
   if (regno == FRAME_POINTER_REGNUM || regno == ARG_POINTER_REGNUM)
     return mode == Pmode;
 
-  if (GP_REGNUM_P (regno) && ! aarch64_vect_struct_mode_p (mode))
-    return 1;
-
-  if (FP_REGNUM_P (regno))
+  if (GP_REGNUM_P (regno))
+    {
+      if (GET_MODE_SIZE (mode) <= 8)
+	return true;
+      if (GET_MODE_SIZE (mode) <= 16)
+	return (regno & 1) == 0;
+    }
+  else if (FP_REGNUM_P (regno))
     {
       if (aarch64_vect_struct_mode_p (mode))
 	return
-- 
2.26.2

commit 1266778548e20de82983b6446f3cb685068cfb1e
Author: Kyrylo Tkachov <kyrylo.tkachov@arm.com>
Date:   Wed May 6 16:20:38 2020 +0100

[AArch64] Use __getauxval instead of getauxval in LSE detection code in libgcc
    
    This version of the fix uses __getauxval instead of getauxval.
    The whole thing is guarded simply on __gnu_linux__.
    __getauxval was introduced in 2.16 but the aarch64 port was added in 2.17 so in practice I expect all aarch64 glibcs to support __getauxval.
    
    Bootstrapped and tested on aarch64-none-linux-gnu.
    Also tested on aarch64-none-elf.
    
    2020-05-06  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>
    
            * config/aarch64/lse-init.c (init_have_lse_atomics): Use __getauxval
            instead of getauxval.
            (AT_HWCAP): Define.
            (HWCAP_ATOMICS): Define.
            Guard detection on __gnu_linux__.

diff --git a/libgcc/config/aarch64/lse-init.c b/libgcc/config/aarch64/lse-init.c
index 74acef25cce..00e9ab8cd1c 100644
--- a/libgcc/config/aarch64/lse-init.c
+++ b/libgcc/config/aarch64/lse-init.c
@@ -29,19 +29,20 @@ see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
 _Bool __aarch64_have_lse_atomics
   __attribute__((visibility("hidden"), nocommon));
 
-/* Disable initialization of __aarch64_have_lse_atomics during bootstrap.  */
-#if !defined(inhibit_libc) && defined(HAVE_SYS_AUXV_H)
-# include <sys/auxv.h>
+/* Gate availability of __getauxval on glibc.  All AArch64-supporting glibc
+   versions support it.  */
+#ifdef __gnu_linux__
 
-/* Disable initialization if the system headers are too old.  */
-# if defined(AT_HWCAP) && defined(HWCAP_ATOMICS)
+# define AT_HWCAP	16
+# define HWCAP_ATOMICS	(1 << 8)
+
+unsigned long int __getauxval (unsigned long int);
 
 static void __attribute__((constructor))
 init_have_lse_atomics (void)
 {
-  unsigned long hwcap = getauxval (AT_HWCAP);
+  unsigned long hwcap = __getauxval (AT_HWCAP);
   __aarch64_have_lse_atomics = (hwcap & HWCAP_ATOMICS) != 0;
 }
 
-# endif /* HWCAP */
-#endif /* inhibit_libc */
+#endif /* __gnu_linux__  */

Places

File gcc7-aarch64-moutline-atomics.patch of Package cross-arm-gcc7

Places