similar to: How to invoke the print method in MachineFunctionPass

On Jul 11, 2007, at 11:39 AM, David Greene wrote: > On Wednesday 11 July 2007 12:41, Tanya M. Lattner wrote: > >> I think the coalescer should be flexible enough to be run >> independent of >> the register allocator. For example, you may want to expose the >> copies >> induced by transforming out of SSA to the scheduler. If the >> scheduler is

On Wednesday 11 July 2007 15:07, Christopher Lamb wrote: > Could it be possible for there to be a harness type interface that > would allow coalescers that support both modes to be hooked into the > pass registration, and those that depend on the allocator not be > registered as passes? I have a patch for this kind of thing attached. Please take a look and let me know if it looks

Hi Jakob, I've got a good test case that I'm working on at the moment. I noticed something odd though: Is '0' a valid register unit? I'm seeing a LiveInterval with li->reg == 0 show up, which previously wasn't valid. We have a few checks around the place to disregard the '0' physreg - could these trigger on interaction with a '0' interval? That could

Hello, I have a minimal testcase which crashes RegisterCoalescer in my out-of-tree target. It only crashes in Debug builds of llc---not in Release builds. Also, interesting to note that the x86 backend lowers this same testcase successfully. I did a quick search of bugs.llvm.org and found no matches. This implies that the problem is in my backend and/or how my backend interacts with

LLVM can model some quite complicated register banks now, and we even use registers to model some encoding constraints. For example, a few ARM instructions like strexd have two register operands that must be an aligned pair of consecutive GPR registers (like r0, r1). This constraint is modeled with the GPRPair register class containing R0_R1, R2_R3, ... pseudo-registers. Sometimes ISAs also

I am not sure if my problem is similar to: http://lists.cs.uiuc.edu/pipermail/llvmdev/2003-December/000715.html It seems that the constructor of the static global pass object isn't called: lib/CodeGen/DependenceAnalyzer.cpp: static RegisterAnalysis<DependenceAnalyzer> X("depana", "Dependence Analysis"); I traced into struct RegisterAnalysis ctor, but my pass

The code in collectRanges() does: // Collect ranges for register units. These live ranges are computed on // demand, so just skip any that haven't been computed yet. if (TargetRegisterInfo::isPhysicalRegister(Reg)) { for (MCRegUnitIterator Units(Reg, &TRI); Units.isValid(); ++Units) if (LiveInterval *LI = LIS.getCachedRegUnit(*Units))

Andy, Jacob, I have ported Hexagon MI scheduler to use the new scheduler infrastructure, but one of my tests triggers an assert in LiveInterval update. On the surface it does not make much sense to me, so I wonder if this is something you readily recognize, before I try to prop it open... The assert is: lib/CodeGen/LiveInterval.cpp:266: llvm::LiveRange*

Here it is :  Starting program: /home/vlj/llvmbin/bin/llc -march=r600 -mcpu=cayman /home/vlj/shader [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib64/libthread_db.so.1". llc: /home/vlj/llvm/include/llvm/ADT/SmallVector.h:143: const T& llvm::SmallVectorTemplateCommon<T, <template-parameter-1-2> >::operator[](unsigned int) const [with T

On Aug 13, 2012, at 8:34 AM, Sergei Larin <slarin at codeaurora.org> wrote: > Andy, > > Yes, this is what Arnold has suggested also, and from this point it looks > like it should work, but it will require parsing the bundle every time we > care to know whether this is a real use or a conditional def. This might > become awkward... but I guess I should provide a better

Ok, just to clarify, RegUnits, as far I understand, are Physical registers or alias to Physical registers. They exist because some instructions use physical registers directly rather than virtual register. It's right? And why this RegUnits should be present in the Interference Graph? I thought were only the Live Intervals would be the nodes of the graph. Sorry about the trouble to

Hi, I'm trying to use the VLIWPacketizerList to schedule instructions for the R600 target, and I'm running into this assertion failure: ScheduleDAGInstrs.cpp:558: Cannot schedule terminators or labels! I think I might not be using the VLIWPacketizerList class correctly. I've attached my code to this email. Can anyone spot what I'm doing wrong? Also, I had to add a LiveIntervals

On Tuesday 04 May 2010 05:45:36 Josef Eisl wrote: > >> - As far as I understand it, register allocators are implemented as > >> MachineFunctionPasses. Does a MachineFunction object contain all > >> information needed for a (classic) allocator? > > > > It has the instructions, operands and dependencies among them. There's > > a

It's done at the MachineFunctionPass level as far as I can tell ( http://llvm.org/doxygen/classllvm_1_1LiveIntervals.html). I'd like something at the FunctionPass level. On Tue, Aug 19, 2008 at 6:11 PM, Evan Cheng <evan.cheng at apple.com> wrote: > I am not sure what you mean. Live interval analysis is done at the > function level. > > Evan > > On Aug 19, 2008,

Jakob Stoklund Olesen <stoklund at 2pi.dk> writes: >> A MachineOperand has both a getReg() and a getSubReg() interface. >> For a physical register operand, is getReg() guaranteed to be the >> "most super" register with getSubReg() providing the specific >> subregister information for the operand? If so then for my current >> purposes it seems I

David Greene wrote: > On Saturday 01 May 2010 08:34:50 Josef Eisl wrote: >> Hello, >> >> I want learn more about register allocation and do some analysis for a >> current research project. After reading some papers (eg. Chaitin, >> Briggs) I think its time to get my hands dirty :). > > Welcome! > >> First I plan to (re)implement some of the classic

Hello to all. I'm trying to implement a simple register allocator using graph colouring (I know, everyone has already done that :-)) and I'm also using LLVM 3.4 from master branch. The algorithm I'm using is based on the one described on the "Modern Compiler Implementation in C". My implementation is totally experimental and doesn't aim to be fast, eficient or even

On Aug 30, 2012, at 1:20 PM, Arnold Schwaighofer <arnolds at codeaurora.org> wrote: > The code in collectRanges() does: > > // Collect ranges for register units. These live ranges are computed on > // demand, so just skip any that haven't been computed yet. > if (TargetRegisterInfo::isPhysicalRegister(Reg)) { > for (MCRegUnitIterator Units(Reg,

Here are some recent stats of the fast vs local vs linear scan at O0 on "opt -std-compile-opts" processed bitcode files. The fast regalloc is still certainly faster at codegen than local with such bitcode files. Let me know if the link doesn't work: https://spreadsheets.google.com/a/google.com/ccc?key=0At5EJFcCBf-wdDgtd2FoZjU4bFBzcFBtT25rQkgzMEE&hl=en Misc stuff: I ran into an

On Sep 17, 2013, at 5:15 AM, Leandro Santiago <leandrosansilva at gmail.com> wrote: > Hello to all. I'm trying to implement a simple register allocator using graph colouring (I know, everyone has already done that :-)) and I'm also using LLVM 3.4 from master branch. > > The algorithm I'm using is based on the one described on the "Modern Compiler Implementation in