search for: llvm31

...o use inst_iterator in the *template <> struct GraphTraits<DFG<Function*> >*) to MachineInstrIterator.h to iterate over machine instructions in a machine function. But the build is throwing the following error: llvm[2]: Compiling MC_DFG.cpp for Debug+Asserts build /home/abhi/work/llvm31/llvm/include/llvm/Support/GraphWriter.h: In member function ‘void llvm::GraphWriter<GraphType>::writeNodes() [with GraphType = llvm::MCDFGraph<llvm::MachineFunction*>]’: /home/abhi/work/llvm31/llvm/include/llvm/Support/GraphWriter.h:100: instantiated from ‘void llvm::GraphWriter<Grap...

...emplate > <> struct GraphTraits<DFG<Function*> >*) to MachineInstrIterator.h to > iterate over machine instructions in a machine function. But the build is > throwing the following error: > > llvm[2]: Compiling MC_DFG.cpp for Debug+Asserts build > /home/abhi/work/llvm31/llvm/include/llvm/Support/GraphWriter.h: In member > function ‘void llvm::GraphWriter<GraphType>::writeNodes() [with GraphType = > llvm::MCDFGraph<llvm::MachineFunction*>]’: > /home/abhi/work/llvm31/llvm/include/llvm/Support/GraphWriter.h:100: > instantiated from ‘void llvm:...

Edwin, thanks, it starts making sense inline comments... Török Edwin wrote: > On 2009-09-06 19:57, Ioannis Nousias wrote: > >> Edwin, >> >> thank you for your effort, but I'm not sure I understand. >> Are you describing a graph traversal problem? Is the data model stored >> in a predecessor/successor fashion, which requires you to 'walk' the

On 2009-09-06 19:57, Ioannis Nousias wrote: > Edwin, > > thank you for your effort, but I'm not sure I understand. > Are you describing a graph traversal problem? Is the data model stored > in a predecessor/successor fashion, which requires you to 'walk' the > graph in order to visit all nodes? (and what happens when you have > disjointed DFGs?). Sorry for the

...le with the Perl script attached to the Bug (5000 is a good size parameter to clearly see the regression): perl create.pl example.cpp 5000 Generate LLVM IR code for the C++ file (the regression is in the backend functionality, so we don't want to measure frontend compilation) llvm31/clang++ example.cpp -c -emit-llvm -o example.ir-31 llvm35/clang++ example.cpp -c -emit-llvm -o example.ir-35 Now you can run the actual performance measurements using llc: llvm31/llc -cppgen=program -o=5000.asm example.ir-31 llvm35/llc -cppgen=program -o=5000.asm examp...

...LLVM with debug symbols as: ./configure --enable-profiling --enable-assertions --enable-expensive-checks --enable-debug-runtime --enable-debug-symbols make ENABLE_OPTIMIZED=0 But now if I use 'BYTECODE_LIBRARY=1' in my llvm pass Makefile then I get the following warning: /data/ahmad/llvm/llvm31/Makefile.rules:1273: Bytecode libraries require LLVM capable compiler but none is available **** Although I put clang in llvm/tools. This warning doesn't appear when I compile llvm in non-debug mode. Cheers, Ahmad -------------- next part -------------- An HTML attachment was scrubbed... URL:...

On Oct 25, 2011, at 6:09 AM, Marcello Maggioni wrote: > Hi, I'm trying to compile the latest clang/llvm SVN versions and I get > this error on multiple systems : Linking, not compiling, but still. I am getting a similar error when building this morning. > Undefined symbols for architecture x86_64: > "clang::Sema::checkPseudoObjectRValue(clang::Expr*)", referenced

Hi, I'm trying to compile the latest clang/llvm SVN versions and I get this error on multiple systems : (Compiling with gcc): llvm[4]: Compiling cc1_main.cpp for Debug+Asserts build llvm[4]: Compiling cc1as_main.cpp for Debug+Asserts build llvm[4]: Compiling driver.cpp for Debug+Asserts build llvm[4]: Linking Debug+Asserts executable clang