similar to: [LLVMdev] Passing llc options to Clang

Hi, I'm working on a project that use llvm openrisc beckend (currently not part of the upstream). Right now I'm looking at a bug where llc generates memory instructions that has out-of-range displacement, for example l.sb 37668(r1), r2 in which 37668 is a 17 bit signed integer, but the instruction only allows 16 bit signed displacement. As a result, after running through the

Hi John Thanks for your suggestions, they all sound reasonable to me. The way I'm thinking right now is to write a MachineFuncionPass that iterate through each MachinBasicBlock, for each MBB, adds up the instructions counts of previous MBBs, that number multiply by 4 should be the offset of that MBB from its MachineFunction. In order to correctly count the instructions, this pass should be

Hi, Is there a way to see what the default argument llc takes from clang/clang++ ? I'm debugging my backend with a very simple c++ program. Running through *clang++ -target myTarget -S simple.cpp -o simple.s * (bug does no show up) gives me different results from *clang++ -target myTarget -S -emit-llvm -o simple.cpp -o simple.ll* *llc simple.ll -o simple.s *(bug shows up) Just trying to

On Sat, Feb 28, 2015 at 7:39 AM, Ziqiang Patrick Huang < ziqiang.huang1001 at gmail.com> wrote: > Hi John > > Thanks for your suggestions, they all sound reasonable to me. The way I'm > thinking right now is to write a MachineFuncionPass that iterate through > each MachinBasicBlock, for each MBB, adds up the instructions counts of > previous MBBs, that number multiply

On 2/27/15 6:30 PM, Ziqiang Patrick Huang wrote: > Hi, all > > Is there a way of getting the basic block offset from its parent > function ? At the LLVM IR level, no. At the code generator layer (MachineFunctionPass layer or the MC layer), probably yes. > > What I'm trying to do is to get an execution count of each basic > blocks, so I need to know the starting

Hi, all Is there a way of getting the basic block offset from its parent function ? What I'm trying to do is to get an execution count of each basic blocks, so I need to know the starting address of each basic blocks. Obviously we can't get the absolute address before linking the program, but the offset relative to parent function should be available so I can take it and get the function

I'm trying to clone a MachineInstr in the AsmPrinter::EmitInstruction, Here is the code: void EsenciaAsmPrinter::EmitInstruction(const MachineInstr *MI) { const MachineFunction *MF = MI->getParent()->getParent(); MachineInstr *CloneMI = MF->CloneMachineInstr(MI); ... ... } The problem is that MF is a const and CloneMachineInstr expects a non-const. Does anybody know if there is

Hello. I'm pretty much new to R and I'm trying to produce some figures. It seems to me, that R has some asynchronous way of plotting figures. When I run this code: #constructs the semivariogram of SC1929 vgm1 <- variogram(SC1929~1,~U+V,puerto.map$att.data) # trying to make new plot dev.set(which=dev.next()) plot(vgm1) title(main="Semivariogram",font.main=4)

Alright, go ahead with it. —Owen > On Nov 17, 2014, at 4:58 AM, Daniil Troshkov <troshkovdanil at gmail.com> wrote: > > Hi! > > I have not found test case. (It is because we have no target using "ExpandInsertToVectorThroughStack"). > But I tested it for target currently not included in llvm trunk. > > This fix correct and trivial, so I'm offering

Hello, I'm trying to switch from Matlab to R-project, and having some difficulties. I make a use of multidimensional matrices. For example, I need to extract mean from one of the dimensions: % we have matrix data of size: 130 x 11 x 350 x 2 data = data - repmat(mean(data,3),[130 1 1 1]); In R project I managed to do that in a very pervarsive way: # mean(data,3) in R base <-

I detected this bug using test case from platform which is not currently supported on llvm targets. (Our team is porting llvm on new target). Creating the test case will take some extra time. I'll try to do it ASAP. Have you any ideas about the test case? (targets using ExpandInsertToVectorThroughStack, etc...) On Wed, Nov 12, 2014 at 8:29 PM, Owen Anderson <resistor at mac.com> wrote:

I agree, but I have no experience with LLVM doc. The job is very simple: add link at page https://github.com/draperlaboratory/fracture/wiki/How-TableGen%27s-DAGISel-Backend-Works into doc page http://llvm.org/docs/TableGen/BackEnds.html But how I can do it? On Tue, Jun 3, 2014 at 4:41 PM, Renato Golin <renato.golin at linaro.org> wrote: > Hi Daniil, > > Those look great! I think

Hi all! There is doc about tglben backends: http://llvm.org/docs/TableGen/BackEnds.html May be should to add link at https://github.com/draperlaboratory/fracture/wiki/How-TableGen%27s-DAGISel-Backend-Works I remember when I first met the LLVM it would be very useful. -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hi Nick! In r207467 you added code(libunwind: DwarfInstructions.hpp): assert(lastReg <= (int)cieInfo.returnAddressRegister && "register range does not contain return address register"); for (int i = 0; i <= lastReg; ++i) { ..... else if (i == (int)cieInfo.returnAddressRegister) There is misprint here: lastReg should be >=

Dear all, I’m not sure if it is O.K. to ask this question here. But where can I find the code for the function ‘C_Cdqrls’ which is called by the R function ‘lsfit‘. Thank you all. Sorry for being naïve if so. -------------------- Ziqiang Zhao 2013-04-23 [[alternative HTML version deleted]]

Hi all! There is a simple c code: double a = 3570.0; double b = 35.7 * 100; int main () { if (b != a) return 1; return 0; } It returns 1 due to: 40abe400 00000000 //a = 3570.0 40abe400 00000001 //b = 35.7 * 100; gcc do the same thing, so I think it's ok but why? For floats: float a = 3570.0; float b = 35.7 * 100; int main () { if (b != a) return 1; return 0; } It returns 0...

I have the Active Directory domain with Windows 2008 R2 domain controller and Samba domain controller on CentOS 7. Samba is 4.3.5 (self-compiled). Forest and domain levels are Windows 2008 R2. After joining Samba to the domain as the domain controller there were no DC=ForestDnsZones and DC=DomainDnsZones records on "OUTBOUND NEIGHBORS". I fixed it with ntdsutil, as it's written here

Hi Owen! The "First store the whole vector" is without uses and will be deleted later. I've attached trivial patch to fix it. I have no commit access so if patch is OK, please, commit it . Danil. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20141112/52592eeb/attachment.html>

Hi all, I¡¯m trying to build R (32-bit) under Windows 7 (64-bit). To do so, I just use ¡±make all recommended¡± as mentioned in the documents and it does work for R 2.15.2. However, for R 3.0.2, I get the following error message, gcc -shared -s -static-libgcc -o tools.dll tmp.def text.o init.o Rmd5.o md5.o signals.o install.o getfmts.o http.o

In the linked bug, the issue is that symbols defined in multiple places (LTO and non-LTO) are provided by the linker plugin (which in GCC's case marks the symbols as private). The error appears to be a difference in which symbol gold or ld picks as canonical. I don't think it's the same as the issue you're seeing (unless you're mixing LTO and non-LTO files). I'd guess