similar to: [LLVMdev] How to change the linker of clang

That's what I used to do but each Ubuntu update would revert the symlink. That's when I found about -fuse-ld=gold, I'd expect it to work with other linkers as well. On Fri, Jun 12, 2015 at 11:52 AM, Nathan Wilson <nwilson20 at gmail.com> wrote: > > > On Thu, Jun 11, 2015 at 8:31 PM, Eric Lu <eirc.lew at gmail.com> wrote: > >> 1) Build clang with

reg2mem does not eliminate phi nodes the way codegen does, it just converts ssa values to non-ssa values. It's not the same thing. On Tue, Apr 21, 2015 at 10:14 PM, Eric Lu <eirc.lew at gmail.com> wrote: > By the way. When I remove these phi nodes with -reg2mem, some new load > operations will be inserted, but when I try cache load operations with: > visitFunction >

Hi, Daniel I want to profile load/store operations, in order to reduce the overhead of profiling, I try to instrument the optimized llvm ir, which has phi nodes. BTW, when the value of some load/store operations may have multi-source, then the load will be translated into phi nodes, and all phi nodes are placed in the front of BB. Sometimes, the position is not where the load happens, is there

Best Regards! Eric Lew On 周三, 4月 22, 2015 at 12:11 下午, David Blaikie < dblaikie at gmail.com [dblaikie at gmail.com] > wrote: On Tue, Apr 21, 2015 at 8:34 PM, Eric Lu <eirc.lew at gmail.com> wrote: > Hi, all > > I want to instrument load/store operations in LLVM IR. And I find the LLVM > IR generated with flag -o3 is much more efficient than -o0, so I try to >

Hi, If I want to know where the stack variables are declared? For example, whether it is declared within a loop or not? Like variables a[100] and temp. int a[100]; for( int i = 0; i < N; i++){ int temp; } Can this be done in LLVM IR? Or should be implemented in Clang. Thanks! Eric -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hi, Ben gcc version is: 4.7.0 clang version is: 3.2 The command options are: 1) clang generated object files: clang -O3 -c *.c -o x.o link: clang *.o -lm -o mpeg2enc 2) gcc generated object files: gcc -O3 -c *.c -o x.o link: gcc *.o -lm -o mpeg2enc Yes, maybe it is the config introduced problems, I build it with enable debug options. Thanks! Eric On Thu, Jun 27, 2013 at 7:49 PM,

Hi, all When compiling a program with -g -O0, and if we have a PC, then with addr2line, we can get the line number of the instruction. My quesions are: what is the result of Phi node instruction, can we get the similar results ? -- Best Regards! Eric Lew -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hi Eric,  I was considering machine instructions to get scope information. And variable declaration does not correspond to machine instruction, hence the problem i.e. no scope associated with it. If 'i' is initialized in the 'if-scope' then we get 'variable i' mapped to correct scope as corresponding machine instruction is generated for this. This is not a problem as we

Hi, all I want to instrument load/store operations in LLVM IR. And I find the LLVM IR generated with flag -o3 is much more efficient than -o0, so I try to instrument on these optimized LLVM IR, but, some load operations are changed to Phi loads. Then, I try the -reg2mem flag to remove these phi nodes. The problems are: 1) I use visitLoadInst to collect these load operations, and it seems I

It seems there is no -enable-iv-rewrite now in llvm3.2, and it suggest -enable-load-pre, but it still does not work. So, how to active the transform? -- View this message in context: http://llvm.1065342.n5.nabble.com/indvars-issues-tp4646p58587.html Sent from the LLVM - Dev mailing list archive at Nabble.com.

On Thu, Jun 27, 2013 at 6:04 PM, Eric Lu <eirc.lew at gmail.com> wrote: > Hi, Ben > gcc version is: 4.7.0 > clang version is: 3.2 > > The command options are: > 1) clang > generated object files: clang -O3 -c *.c -o x.o > link: clang *.o -lm -o mpeg2enc > > 2) gcc > generated object files: gcc -O3 -c *.c -o x.o > link: gcc *.o -lm -o mpeg2enc >

Thank your reply. Pankaj. Actually, I have done it very similar to yours. But I think for my demand, it is better to implement in Front End. Maybe I will re-implement it later in clang. ---------- Forwarded message ---------- From: Pankaj Gode [via LLVM] <ml-node+s1065342n59345h22 at n5.nabble.com> Date: Mon, Jul 15, 2013 at 2:35 PM Subject: Re: Regarding scope information for variable

On 04/12/2019 06:31 PM, Tom Stellard via llvm-dev wrote: > On 04/12/2019 04:28 PM, AiChi via llvm-dev wrote: >> Hi, >> >> I'm currently working on one of my team's project to build LLVM full clang toolchain (Clang, libcxx, libcxxabi) on a CentOS machine. >> >> Previously we compiled our codebase with llvm-toolset-7/clang++, which by default takes stdlibc++

I have the same demand. Have you resolved this problems? if so, would you share me the solution? Best Regards. Eric -- View this message in context: http://llvm.1065342.n5.nabble.com/Regarding-scope-information-for-variable-declaration-tp47707p59268.html Sent from the LLVM - Dev mailing list archive at Nabble.com.

Hi All,   I have question regarding lexical scope information.   If I have .c source with scope as below. void func() {    //-- some code here..    { //parent scope       if() //high pass       {          int i;          for( i =0; i < FRAM_I; i++)          {          }       }       if() //low pass       {          int i;          for( i =0; i < FRAM_J; i++)          {          }       }

Hello, all When I compile mpeg2enc of Mediabench, it seems clang and llvm take longer time than gcc. I meausre the time as follow: gcc -O3 8.190s clang -O3 20.253s will this means clang and llvm are slower than gcc in compiling c code. Eric -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hi, Currently, I can build lld/llvm on Fedora 17 but fail to build it on iMac. The commands I use as follows. Anyone can help me? Or is there the lld mail list I can post the question? Linux/Fedora can be built by the following command cmake -DCMAKE_CXX_COMPILER=clang++ -DCMAKE_C_COMPILER=clang -DCMAKE_CXX_FLAGS=-std=c++11 -DCMAKE_BUILD_TYPE=Debug -G "Unix Makefiles" ../src/ make iMac

Greetings! I have been building llvm-3.6.x, 3.7.1 and 3.7.2 with (glibc-2.12.1, binutils-2.24, gcc-4.9.2) almost same set if CMake flags. However while building LLVM-3.8.0 using same CMake flags I am observing projects/libcxx/include/cmath errors... ...'::signbit' has not been declared ...'::fpclassify' has not been declared ...'::isfinite' has not been declared ...

2015-03-29 3:33 GMT-05:00 Ed Schouten <ed at 80386.nl>: > I am cross-compiling compiler-rt for my specific use-case. That said, > I'm cross-compiling from x86-64 to x86-64, but for a different ABI. > This is what I use: > > > https://github.com/NuxiNL/cloudlibc/wiki/Building-runtime-libraries#user-content-compiler-rt > > -DCMAKE_CROSSCOMPILING=True > > As

The options is cmake -DCMAKE_BUILD_TYPE="Debug" -DCMAKE_INSTALL_PREFIX=~software/llvmsvn -DCMAKE_EXPORT_COMPILE_COMMANDS=ON /home/xxx/lsoftware/llvm-3.2-svn $make $make install On Fri, Jun 28, 2013 at 12:37 PM, Eric Christopher <echristo at gmail.com>wrote: > On Thu, Jun 27, 2013 at 6:04 PM, Eric Lu <eirc.lew at gmail.com> wrote: > > Hi, Ben > > gcc