Current, indirect function call prevents tail-call optimization on AArch64.
This patch adapt the fix for PR arm/19599 to AArch64.
Is it ok for next stage 1?
Thanks.
-- Jiong
gcc/
* config/aarch64/predicates.md (aarch64_call_insn_operand): New
predicate.
* config/aarch64/constraints.md ("Ucs", "Usf"): New constraints.
* config/aarch64/aarch64.md (*sibcall_insn, *sibcall_value_insn):
Adjust for
tailcalling through registers.
* config/aarch64/aarch64.h (enum reg_class): New caller save
register class.
(REG_CLASS_NAMES): Likewise.
(REG_CLASS_CONTENTS): Likewise.
* config/aarch64/aarch64.c (aarch64_function_ok_for_sibcall): Allow
tailcalling
without decls.
gcc/testsuite
*gcc.target/aarch64/tail-indirect-call.c: New test.
diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index 901ad3d..fd93554 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -1168,16 +1168,7 @@ aarch64_expand_mov_immediate (rtx dest, rtx imm)
static bool
aarch64_function_ok_for_sibcall (tree decl, tree exp ATTRIBUTE_UNUSED)
{
- /* Indirect calls are not currently supported. */
- if (decl == NULL)
- return false;
-
- /* Cannot tail-call to long-calls, since these are outside of the
- range of a branch instruction (we could handle this if we added
- support for indirect tail-calls. */
- if (aarch64_decl_is_long_call_p (decl))
- return false;
-
+ /* Currently, always true. */
return true;
}
@@ -4255,6 +4246,7 @@ aarch64_class_max_nregs (reg_class_t regclass, enum machine_mode mode)
switch (regclass)
{
case CORE_REGS:
+ case CALLER_SAVE_REGS:
case POINTER_REGS:
case GENERAL_REGS:
case ALL_REGS:
diff --git a/gcc/config/aarch64/aarch64.h b/gcc/config/aarch64/aarch64.h
index 13c424c..6911206 100644
--- a/gcc/config/aarch64/aarch64.h
+++ b/gcc/config/aarch64/aarch64.h
@@ -407,6 +407,7 @@ enum reg_class
{
NO_REGS,
CORE_REGS,
+ CALLER_SAVE_REGS,
GENERAL_REGS,
STACK_REG,
POINTER_REGS,
@@ -422,6 +423,7 @@ enum reg_class
{ \
"NO_REGS", \
"CORE_REGS", \
+ "CALLER_SAVE_REGS", \
"GENERAL_REGS", \
"STACK_REG", \
"POINTER_REGS", \
@@ -434,6 +436,7 @@ enum reg_class
{ \
{ 0x00000000, 0x00000000, 0x00000000 }, /* NO_REGS */ \
{ 0x7fffffff, 0x00000000, 0x00000003 }, /* CORE_REGS */ \
+ { 0x0007ffff, 0x00000000, 0x00000000 }, /* CALLER_SAVE_REGS */ \
{ 0x7fffffff, 0x00000000, 0x00000003 }, /* GENERAL_REGS */ \
{ 0x80000000, 0x00000000, 0x00000000 }, /* STACK_REG */ \
{ 0xffffffff, 0x00000000, 0x00000003 }, /* POINTER_REGS */ \
diff --git a/gcc/config/aarch64/aarch64.md b/gcc/config/aarch64/aarch64.md
index 99a6ac8..f30b444 100644
--- a/gcc/config/aarch64/aarch64.md
+++ b/gcc/config/aarch64/aarch64.md
@@ -513,6 +513,10 @@
(use (match_operand 2 "" ""))])]
""
{
+ if (!REG_P (XEXP (operands[0], 0))
+ && (GET_CODE (XEXP (operands[0], 0)) != SYMBOL_REF))
+ XEXP (operands[0], 0) = force_reg (Pmode, XEXP (operands[0], 0));
+
if (operands[2] == NULL_RTX)
operands[2] = const0_rtx;
}
@@ -526,30 +530,37 @@
(use (match_operand 3 "" ""))])]
""
{
+ if (!REG_P (XEXP (operands[1], 0))
+ && (GET_CODE (XEXP (operands[1], 0)) != SYMBOL_REF))
+ XEXP (operands[1], 0) = force_reg (Pmode, XEXP (operands[1], 0));
+
if (operands[3] == NULL_RTX)
operands[3] = const0_rtx;
}
)
(define_insn "*sibcall_insn"
- [(call (mem:DI (match_operand:DI 0 "" "X"))
+ [(call (mem:DI (match_operand:DI 0 "aarch64_call_insn_operand" "Ucs,Usf"))
(match_operand 1 "" ""))
(return)
(use (match_operand 2 "" ""))]
- "GET_CODE (operands[0]) == SYMBOL_REF"
- "b\\t%a0"
+ "SIBLING_CALL_P (insn)"
+ "@
+ br\\t%0
+ b\\t%a0"
[(set_attr "type" "branch")]
-
)
(define_insn "*sibcall_value_insn"
[(set (match_operand 0 "" "")
- (call (mem:DI (match_operand 1 "" "X"))
+ (call (mem:DI (match_operand 1 "aarch64_call_insn_operand" "Ucs,Usf"))
(match_operand 2 "" "")))
(return)
(use (match_operand 3 "" ""))]
- "GET_CODE (operands[1]) == SYMBOL_REF"
- "b\\t%a1"
+ "SIBLING_CALL_P (insn)"
+ "@
+ br\\t%1
+ b\\t%a1"
[(set_attr "type" "branch")]
)
diff --git a/gcc/config/aarch64/constraints.md b/gcc/config/aarch64/constraints.md
index 12ab570..244f97d 100644
--- a/gcc/config/aarch64/constraints.md
+++ b/gcc/config/aarch64/constraints.md
@@ -92,6 +92,14 @@
(and (match_code "const_int")
(match_test "(unsigned HOST_WIDE_INT) ival < 64")))
+(define_register_constraint "Ucs" "CALLER_SAVE_REGS"
+ "@internal The caller save registers. Useful for sibcalls.")
+
+(define_constraint "Usf"
+ "@internal Usf is a symbol reference."
+ (match_code "symbol_ref")
+)
+
(define_constraint "UsM"
"@internal
A constraint that matches the immediate constant -1."
diff --git a/gcc/config/aarch64/predicates.md b/gcc/config/aarch64/predicates.md
index c8e27d8..2702a3c 100644
--- a/gcc/config/aarch64/predicates.md
+++ b/gcc/config/aarch64/predicates.md
@@ -26,6 +26,10 @@
&& GET_MODE_CLASS (GET_MODE (op)) == MODE_CC"))))
)
+(define_predicate "aarch64_call_insn_operand"
+ (ior (match_code "symbol_ref")
+ (match_operand 0 "register_operand")))
+
(define_predicate "aarch64_simd_register"
(and (match_code "reg")
(ior (match_test "REGNO_REG_CLASS (REGNO (op)) == FP_LO_REGS")
diff --git a/gcc/testsuite/gcc.target/aarch64/tail-indirect-call.c b/gcc/testsuite/gcc.target/aarch64/tail-indirect-call.c
new file mode 100644
index 0000000..057ae3e
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/tail-indirect-call.c
@@ -0,0 +1,8 @@
+/* { dg-do compile } */
+/* { dg-options "-O2" } */
+
+typedef void FP(int);
+/* { dg-final { scan-assembler "br" } } */
+/* { dg-final { scan-assembler-not "blr" } } */
+void f1(FP fp, int n) { (fp)(n); }
+void f2(int n, FP fp) { (fp)(n); }