Implement builtin_{setjmp/longjmp} on PPC
This implements SJLJ lowering on PPC, making the Clang functions
__builtin_{setjmp/longjmp} functional on PPC platforms. The implementation
strategy is similar to that on X86, with the exception that a branch-and-link
variant is used to get the right jump address. Credit goes to Bill Schmidt for
suggesting the use of the unconditional bcl form (instead of the regular bl
instruction) to limit return-address-cache pollution.
Benchmarking the speed at -O3 of:
static jmp_buf env_sigill;
void foo() {
__builtin_longjmp(env_sigill,1);
}
main() {
...
for (int i = 0; i < c; ++i) {
if (__builtin_setjmp(env_sigill)) {
goto done;
} else {
foo();
}
done:;
}
...
}
vs. the same code using the libc setjmp/longjmp functions on a P7 shows that
this builtin implementation is ~4x faster with Altivec enabled and ~7.25x
faster with Altivec disabled. This comparison is somewhat unfair because the
libc version must also save/restore the VSX registers which we don't yet
support.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@177666 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/lib/Target/PowerPC/PPCInstrInfo.td b/lib/Target/PowerPC/PPCInstrInfo.td
index a05ebad..84cdb1f 100644
--- a/lib/Target/PowerPC/PPCInstrInfo.td
+++ b/lib/Target/PowerPC/PPCInstrInfo.td
@@ -158,6 +158,14 @@
def PPCtc_return : SDNode<"PPCISD::TC_RETURN", SDT_PPCTC_ret,
[SDNPHasChain, SDNPOptInGlue, SDNPVariadic]>;
+def PPCeh_sjlj_setjmp : SDNode<"PPCISD::EH_SJLJ_SETJMP",
+ SDTypeProfile<1, 1, [SDTCisInt<0>,
+ SDTCisPtrTy<1>]>,
+ [SDNPHasChain, SDNPSideEffect]>;
+def PPCeh_sjlj_longjmp : SDNode<"PPCISD::EH_SJLJ_LONGJMP",
+ SDTypeProfile<0, 1, [SDTCisPtrTy<0>]>,
+ [SDNPHasChain, SDNPSideEffect]>;
+
def PPCvcmp : SDNode<"PPCISD::VCMP" , SDT_PPCvcmp, []>;
def PPCvcmp_o : SDNode<"PPCISD::VCMPo", SDT_PPCvcmp, [SDNPOutGlue]>;
@@ -391,6 +399,12 @@
let EncoderMethod = "getMemRIXEncoding";
}
+// A single-register address. This is used with the SjLj
+// pseudo-instructions.
+def memr : Operand<iPTR> {
+ let MIOperandInfo = (ops ptr_rc:$ptrreg);
+}
+
// PowerPC Predicate operand. 20 = (0<<5)|20 = always, CR0 is a dummy reg
// that doesn't matter.
def pred : PredicateOperand<OtherVT, (ops imm, CRRC),
@@ -404,6 +418,10 @@
def xoaddr : ComplexPattern<iPTR, 2, "SelectAddrIdxOnly",[], []>;
def ixaddr : ComplexPattern<iPTR, 2, "SelectAddrImmShift", [], []>; // "std"
+// The address in a single register. This is used with the SjLj
+// pseudo-instructions.
+def addr : ComplexPattern<iPTR, 1, "SelectAddr",[], []>;
+
/// This is just the offset part of iaddr, used for preinc.
def iaddroff : ComplexPattern<iPTR, 1, "SelectAddrImmOffs", [], []>;
def xaddroff : ComplexPattern<iPTR, 1, "SelectAddrIdxOffs", [], []>;
@@ -505,6 +523,14 @@
}
}
+// The direct BCL used by the SjLj setjmp code.
+let isCall = 1, hasCtrlDep = 1, PPC970_Unit = 7 in {
+ let Defs = [LR], Uses = [RM] in {
+ def BCL : BForm_2<16, 20, 31, 0, 1, (outs), (ins condbrtarget:$dst),
+ "bcl 20, 31, $dst">;
+ }
+}
+
// Darwin ABI Calls.
let isCall = 1, PPC970_Unit = 7, Defs = [LR] in {
// Convenient aliases for call instructions
@@ -583,6 +609,23 @@
"ba $dst", BrB,
[]>;
+let hasSideEffects = 1, isBarrier = 1, isCodeGenOnly = 1,
+ usesCustomInserter = 1 in {
+ def EH_SjLj_SetJmp32 : Pseudo<(outs GPRC:$dst), (ins memr:$buf),
+ "#EH_SJLJ_SETJMP32",
+ [(set GPRC:$dst, (PPCeh_sjlj_setjmp addr:$buf))]>,
+ Requires<[In32BitMode]>;
+ let isTerminator = 1 in
+ def EH_SjLj_LongJmp32 : Pseudo<(outs), (ins memr:$buf),
+ "#EH_SJLJ_LONGJMP32",
+ [(PPCeh_sjlj_longjmp addr:$buf)]>,
+ Requires<[In32BitMode]>;
+}
+
+let isBranch = 1, isTerminator = 1, isCodeGenOnly = 1 in {
+ def EH_SjLj_Setup : Pseudo<(outs), (ins directbrtarget:$dst),
+ "#EH_SjLj_Setup\t$dst", []>;
+}
// DCB* instructions.
def DCBA : DCB_Form<758, 0, (outs), (ins memrr:$dst),