similar to: [LLVMdev] Split MachineBasicBlocks

Does the error function *have* to be auto-generated in your pass? Perhaps the original code should use invokes, and your pass insert the error check with a jne to the "unwind" block. On Mon, Aug 3, 2009 at 5:18 AM, Artjom Kochtchi<artjom.kochtchi at googlemail.com> wrote: > > > Dan Gohman-2 wrote: >> >> You don't need a ModulePass to modify a CFG. A

On Jul 8, 2009, at 10:16 AM, Artjom Kochtchi wrote: > > Hi, > > I'm currently trying to modify LLVM to include runtime checks into X86 > binaries. I've looked into some of the possibilities during the phases > happening in LLVM and have the impression that inserting runtime > checks > during selection would be great, since lots of optimizations are > already

Hi, I'm currently trying to modify LLVM to include runtime checks into X86 binaries. I've looked into some of the possibilities during the phases happening in LLVM and have the impression that inserting runtime checks during selection would be great, since lots of optimizations are already done and I can work directly with X86 instructions. I've read through the documentation for

Greetinigs, I am extending llc to include runtime checks for calls (in X86). So a call 'call target' is altered to look like this: [some check] jne error_function call target I've done this by implementing a MachineFunctionPass that is instantiated and added to the PassManager in X86TargetMachine::addPreRegAlloc. In order to create the jne-instruction I need some BasicBlock

On Aug 2, 2009, at 7:09 AM, Artjom Kochtchi wrote: > > Greetinigs, > > I am extending llc to include runtime checks for calls (in X86). So > a call > 'call target' is altered to look like this: > > [some check] > jne error_function > call target > > I've done this by implementing a MachineFunctionPass that is > instantiated > and added

I have a MachineFunctionPass plugged into llc during LLVMTargetMachine::addPreRegAlloc. In this Pass I need to extend calls (i. e. CALL32m, CALL32r) iff they call function within the program. CALL32m has, I think, ten different possibilities for the four operands giving the target address. At the moment I have excluded calls that give the displacement as GlobalAddress or JumpTableIndex

Kenneth Uildriks wrote: > > Does the error function *have* to be auto-generated in your pass? > Perhaps the original code should use invokes, and your pass insert the > error check with a jne to the "unwind" block. > If I understand correctly, unwind is some kind of exception handling in LLVM IR? I'm not sure if this is the right thing. First, I include runtime

On Aug 6, 2009, at 5:45 AM, Artjom Kochtchi wrote: > So I was searching for a possibility to include an additional > (Machine)BasicBlock (or rather a MachineFunction?) somewhere in the > program > that contains the program termination and that I can jump to on check > failure. Can't I do that during a Pass in llc? If you just have a magic block of instructions you want to

Sorry to ask again, but I still can't get it right. The following code compiles and runs, but produces no instructions: Ops.push_back(DAG.getRegister(X86::EAX, MVT::i32)); Ops.push_back(DAG.getConstant(1, MVT::i32)); DAG.getNode(ISD::ADD, DAG.getVTList(MVT::i32), &Ops[0], Ops.size()); I reckon that has something to do with the fact that I am not using the Chain object. But as soon

Thanks to your help I've actually made some progress... Especially the SelectionDAGNodes.h was a good hint. But there are still some things that I can't figure out: // 'mov eax, 41' Chain = DAG.getCopyToReg(Chain, DAG.getRegister(X86::EAX, MVT::i32), DAG.getConstant(41, MVT::i32), InFlag); InFlag = Chain.getValue(1); // 'inc eax' SDValue eaxVal =

On Jun 29, 2009, at 9:00 AM, Artjom Kochtchi wrote: > > Sorry to ask again, but I still can't get it right. > > The following code compiles and runs, but produces no instructions: > Ops.push_back(DAG.getRegister(X86::EAX, MVT::i32)); > Ops.push_back(DAG.getConstant(1, MVT::i32)); > DAG.getNode(ISD::ADD, DAG.getVTList(MVT::i32), &Ops[0], Ops.size()); To read the

Greetings, I am rather new to LLVM, so please excuse my limited knowledge about it. Currently I am trying to modify the X86TargetLowering::LowerCALL method by inserting additional instructions before the call. As far as I understand, nodes are created by calling the getNode method on the DAG. If, for example, I insert the following code Ops.push_back(Chain); Chain = DAG.getNode(ISD::TRAP,

Hi all, I am writing a pass subclassing the MachineFunctionPass and doing the instrumentation in runOnMachineFunction(). But I do not know how to insert a function call. In my work, I write a special function for a program to be compiled and I need to insert this function after certain instructions whose destination register is esp. In X86InstrInfo.cpp, there are two kinds of call may be

Hi, For some schedulers like Trace Scheduling it is necessary to clone basic blocks. Instinctively I would think the "Machine Instruction Scheduler" would be the right place to do so. Is it possible to clone MachineBasicBlocks in the "Machine Instruction Scheduler" pass? Any snares? Or is it to much effort to implement it there and there is a better place for such things?

Hello, I was looking at how the memory for MachineBasicBlocks is managed and something seems strange. On one hand, a MachineFunction uses a recycling allocator to create and destroy MBBs without costly heap allocation (the same is done for MachineInstrs, MachineOperands, etc.). On the other hand, each MBB also has a number of STL vectors within it. For example, for Predecessors, Successors,

I want to be able to do two things with LLVM (both just before code emission): 1. Insert a MachineBasicBlock just before a MachineBasicBlock. There is a function called AddPredecessor(). However, the comment says that it does not update the actual CFG. I want to redirect all CFG edges that are incoming to this MachineBasicBlock to the new one I create, and add just one outgoing edge (no branch)

Dan Gohman-2 wrote: > > You don't need a ModulePass to modify a CFG. A MachineFunctionPass > is fine for this. > Well, okay. But I need to insert some BasicBlock that contains the error handling (that the runtime check jumpts to on failure). Just putting another MachineBasicBlock at the beginning of some arbitrary MachineFunction doesn't seem to be the right way to go, does

On Jun 26, 2009, at 4:49 AM, Artjom K. wrote: > > Thank you for your help. > > I think I managed to create the instruction I wanted: > > // mov eax, 41 > Chain = DAG.getCopyToReg(Chain, DAG.getRegister(X86::EAX, MVT::i32), > DAG.getConstant(41, MVT::i32), InFlag); > InFlag = Chain.getValue(1); > > I don't understand though what InFlag is for. As I read the

Thank you for your help. I think I managed to create the instruction I wanted: // mov eax, 41 Chain = DAG.getCopyToReg(Chain, DAG.getRegister(X86::EAX, MVT::i32), DAG.getConstant(41, MVT::i32), InFlag); InFlag = Chain.getValue(1); I don't understand though what InFlag is for. As I read the code, it even remains uninitialized when first passed to some node creation method.

On Jul 1, 2009, at 2:22 PMPDT, Dan Gohman wrote: >> Ops.push_back(DAG.getConstant(1, MVT::i32)); >> Chain = DAG.getNode(ISD::ADD, DAG.getVTList(MVT::Other, MVT::i32), >> &Ops[0], Ops.size()); >> >> Isn't that the way how it is supposed to work? > > ADD does not use a chain, so there's no chain operand, or > MVT::Other result for it in an ADD