diff options
Diffstat (limited to 'llvm/lib')
| -rw-r--r-- | llvm/lib/Target/AArch64/AArch64AsmPrinter.cpp | 1 | ||||
| -rw-r--r-- | llvm/lib/Target/AArch64/AArch64FrameLowering.cpp | 3 | ||||
| -rw-r--r-- | llvm/lib/Target/AArch64/AArch64InstrInfo.td | 13 | ||||
| -rw-r--r-- | llvm/lib/Target/AArch64/AArch64RegisterInfo.td | 6 |
4 files changed, 21 insertions, 2 deletions
diff --git a/llvm/lib/Target/AArch64/AArch64AsmPrinter.cpp b/llvm/lib/Target/AArch64/AArch64AsmPrinter.cpp index 469c9c27ec4..23b6a65555a 100644 --- a/llvm/lib/Target/AArch64/AArch64AsmPrinter.cpp +++ b/llvm/lib/Target/AArch64/AArch64AsmPrinter.cpp @@ -591,6 +591,7 @@ void AArch64AsmPrinter::EmitInstruction(const MachineInstr *MI) { // attributes (isCall, isReturn, etc.). We lower them to the real // instruction here. case AArch64::TCRETURNri: + case AArch64::TCRETURNriBTI: case AArch64::TCRETURNriALL: { MCInst TmpInst; TmpInst.setOpcode(AArch64::BR); diff --git a/llvm/lib/Target/AArch64/AArch64FrameLowering.cpp b/llvm/lib/Target/AArch64/AArch64FrameLowering.cpp index f2d7ea9ba3e..b0451ca2edb 100644 --- a/llvm/lib/Target/AArch64/AArch64FrameLowering.cpp +++ b/llvm/lib/Target/AArch64/AArch64FrameLowering.cpp @@ -927,7 +927,8 @@ void AArch64FrameLowering::emitEpilogue(MachineFunction &MF, DL = MBBI->getDebugLoc(); unsigned RetOpcode = MBBI->getOpcode(); IsTailCallReturn = RetOpcode == AArch64::TCRETURNdi || - RetOpcode == AArch64::TCRETURNri; + RetOpcode == AArch64::TCRETURNri || + RetOpcode == AArch64::TCRETURNriBTI; } int NumBytes = MFI.getStackSize(); const AArch64FunctionInfo *AFI = MF.getInfo<AArch64FunctionInfo>(); diff --git a/llvm/lib/Target/AArch64/AArch64InstrInfo.td b/llvm/lib/Target/AArch64/AArch64InstrInfo.td index 3531d1e1622..1d9e3d0b812 100644 --- a/llvm/lib/Target/AArch64/AArch64InstrInfo.td +++ b/llvm/lib/Target/AArch64/AArch64InstrInfo.td @@ -360,6 +360,9 @@ let RecomputePerFunction = 1 in { def NotForCodeSize : Predicate<"!MF->getFunction().optForSize()">; // Avoid generating STRQro if it is slow, unless we're optimizing for code size. def UseSTRQro : Predicate<"!Subtarget->isSTRQroSlow() || MF->getFunction().optForSize()">; + + def UseBTI : Predicate<[{ MF->getFunction().hasFnAttribute("branch-target-enforcement") }]>; + def NotUseBTI : Predicate<[{ !MF->getFunction().hasFnAttribute("branch-target-enforcement") }]>; } include "AArch64InstrFormats.td" @@ -6641,10 +6644,18 @@ let isCall = 1, isTerminator = 1, isReturn = 1, isBarrier = 1, Uses = [SP] in { // some verifier checks for outlined functions. def TCRETURNriALL : Pseudo<(outs), (ins GPR64:$dst, i32imm:$FPDiff), []>, Sched<[WriteBrReg]>; + // Indirect tail-call limited to only use registers (x16 and x17) which are + // allowed to tail-call a "BTI c" instruction. + def TCRETURNriBTI : Pseudo<(outs), (ins rtcGPR64:$dst, i32imm:$FPDiff), []>, + Sched<[WriteBrReg]>; } def : Pat<(AArch64tcret tcGPR64:$dst, (i32 timm:$FPDiff)), - (TCRETURNri tcGPR64:$dst, imm:$FPDiff)>; + (TCRETURNri tcGPR64:$dst, imm:$FPDiff)>, + Requires<[NotUseBTI]>; +def : Pat<(AArch64tcret rtcGPR64:$dst, (i32 timm:$FPDiff)), + (TCRETURNriBTI rtcGPR64:$dst, imm:$FPDiff)>, + Requires<[UseBTI]>; def : Pat<(AArch64tcret tglobaladdr:$dst, (i32 timm:$FPDiff)), (TCRETURNdi texternalsym:$dst, imm:$FPDiff)>; def : Pat<(AArch64tcret texternalsym:$dst, (i32 timm:$FPDiff)), diff --git a/llvm/lib/Target/AArch64/AArch64RegisterInfo.td b/llvm/lib/Target/AArch64/AArch64RegisterInfo.td index bbf401b474c..d3710cea068 100644 --- a/llvm/lib/Target/AArch64/AArch64RegisterInfo.td +++ b/llvm/lib/Target/AArch64/AArch64RegisterInfo.td @@ -200,6 +200,12 @@ def tcGPR64 : RegisterClass<"AArch64", [i64], 64, (sub GPR64common, X19, X20, X2 X22, X23, X24, X25, X26, X27, X28, FP, LR)>; +// Restricted set of tail call registers, for use when branch target +// enforcement is enabled. These are the only registers which can be used to +// indirectly branch (not call) to the "BTI c" instruction at the start of a +// BTI-protected function. +def rtcGPR64 : RegisterClass<"AArch64", [i64], 64, (add X16, X17)>; + // GPR register classes for post increment amount of vector load/store that // has alternate printing when Rm=31 and prints a constant immediate value // equal to the total number of bytes transferred. |
