similar to: [LLVMdev] llvm LTO SPEC2000 results?

Thanks Nick. I've been pursuing Gao's technique but can't seem to get opt to remove obviously dead code from even the following trivial example: int mult(int a, int b){ return a*b; } int main(void){ return 0; } While mult is never called it still is not removed. I just can't seem to get opt to understand it's seeing the whole program so it can remove this

Hi All , We are in process of exploring the LTO and found that internalize pass is the replacement for whole program optimisation (-fwhole-program in gcc) in clang and in the below case define i32 @test() #0 { entry: ret i32 0 } define i32 @main() #0 { entry: %retval = alloca i32, align 4 store i32 0, i32* %retval, align 4 %call = call i32 @test() ret i32 %call } ***

Using relative offsets applies to more than just vtables. It would do wonders for constant strings too. -- Sean Silva On Mon, Feb 29, 2016 at 1:53 PM, Peter Collingbourne via cfe-dev < cfe-dev at lists.llvm.org> wrote: > Hi all, > > I'd like to make a proposal to implement the new vtable ABI described in > PR26723, which I'll call the relative ABI. That bug gives more

Dear guys, I am writing some scripts to allow me to compile the programs in SPEC2000 using llvm-gcc. I have been successfull with almost all of them, but need some help with 254.gap. I am producing a .bc file using llvm-gcc, and then a .s using llc. Then I use gcc to produce an executable. In this last phase, I am getting: /usr/bin/ld: Undefined symbols: _SyLibname _SyMemory _SyTime

On Fri, May 27, 2016 at 3:43 AM, Umesh Kalappa via llvm-dev < llvm-dev at lists.llvm.org> wrote: > Hi All , > > We are in process of exploring the LTO and found that internalize > pass is the replacement for whole program optimisation > (-fwhole-program in gcc) in clang and in the below case > > define i32 @test() #0 { > > entry: > > ret i32 0 >

> On 2016-May-27, at 05:55, Teresa Johnson via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > > > On Fri, May 27, 2016 at 3:43 AM, Umesh Kalappa via llvm-dev <llvm-dev at lists.llvm.org> wrote: > Hi All , > > We are in process of exploring the LTO and found that internalize > pass is the replacement for whole program optimisation > (-fwhole-program

Hi all, I'd like to make a proposal to implement the new vtable ABI described in PR26723, which I'll call the relative ABI. That bug gives more details and justification for that ABI. The user interface for the new ABI would be that -fwhole-program-vtables would take an optional value indicating which aspects of the program have whole-program scope. For example, the existing

Please send any comments. As mentioned at the end I will follow up with some patches as soon as they are cleaned up and I create some test cases. RFC: Safe Whole Program Devirtualization Enablement =================================================== High Level Summary ------------------ The goal of the changes described in this RFC is to support aggressive Whole Program Devirtualization without

> On May 27, 2016, at 9:57 AM, Duncan P. N. Exon Smith via llvm-dev <llvm-dev at lists.llvm.org> wrote: > >> >> On 2016-May-27, at 05:55, Teresa Johnson via llvm-dev <llvm-dev at lists.llvm.org> wrote: >> >> >> >> On Fri, May 27, 2016 at 3:43 AM, Umesh Kalappa via llvm-dev <llvm-dev at lists.llvm.org> wrote: >> Hi All ,

Hi, I'm trying to turn on a flag (-inline-generic-switch-cost) introduced in r301649 which improved the inline cost heuristic for switch. In my experiment for LLVM test suite and spec2000/2006 in AArch64, +17.82% performance and -8% code size reduce was observed in spec2000/vertex with O3 LTO. No significant code size / performance regression was found in O3/O2/Os. Before we turn this

> On 2016-May-27, at 10:47, Mehdi Amini <mehdi.amini at apple.com> wrote: > >> >> On May 27, 2016, at 9:57 AM, Duncan P. N. Exon Smith via llvm-dev <llvm-dev at lists.llvm.org> wrote: >> >>> >>> On 2016-May-27, at 05:55, Teresa Johnson via llvm-dev <llvm-dev at lists.llvm.org> wrote: >>> >>> >>>

On 2009-12-23 08:22, Li Shengmei wrote: > > Hi, all > > I use llvm-gcc –O4 to compile SPEC 2000, some benchmarks > can’t be built successfully, such as 164.gzip, 175.vpr etc. > > The error messages are as follows. > > > > …… > > zh.o unlzw.o unpack.o unzip.o util.o zip.o -o gzip > > bits.o: file not recognized: File format not recognized

Hi, all I use llvm-gcc -O4 to compile SPEC 2000, some benchmarks can't be built successfully, such as 164.gzip, 175.vpr etc. The error messages are as follows. .. zh.o unlzw.o unpack.o unzip.o util.o zip.o -o gzip bits.o: file not recognized: File format not recognized collect2: ld returned 1 exit status specmake: *** [gzip] Error 1 specmake options 2> options.err |

Looking at the existing flow of passes for LTO, it appears that most all passes are run on a per file basis, before the call to the gold linker. I'm looking to get people's feedback on whether there would be an advantage to waiting to run a number of these passes until the linking stage. For example, I believe I saw a post a little while back about postponing vectorization until the

Hello, I want to run SPEC2000 benchmark in LLVM to check correctness of a llvm module which I wrote. How to run SPEC2000 benchmark in LLVM? Unnikrishnan C -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.

Hi David, By "dead" do you mean unreachable? My understanding was that the removal of dead code is simply another optimization, which should be disabled after adding "optnone" (and adding the function to llvm.used so the function doesn't later get deleted entirely). I am instrumenting certain basic blocks in an LLVM pass, and would like to compile a binary which

Hi All, After the initial learning curve, we're excited to have put together a completely gcc/binutils-free toolchain based on LLVM. Now that we have things working, we desperately need to optimize the resulting binaries. Our bin files are up to 10x their fully optimized gcc equivalent (1.5k vs 16k). This is for a bare metal ARM based system so this is significant. We're using lld for

Hi all, I'd like to make the following proposal to implement an optimization called virtual constant propagation. ==Introduction== After enabling control flow integrity protection in Chromium, we have observed an unacceptable performance regression in certain critical layout microbenchmarks. Profiling [0] revealed that the cause of the regression was a large number of virtual calls, each

> I managed to solve this problem myself.... The problem was not correctly specifiying the SPEC CPU2000 source directory. configure says the dir was found, but that doesn't mean the sources will be found. For future reference: When you get the error: make[2]: *** No rule to make target `Output/accum.rbc', needed by `Output/177.mesa.linked.rbc'. Stop. You've probably

Dear guys, I've got SPEC2000, and now I am trying to run the tests with it. What I am doing to test it is to configure the test with --enable-spec2000=<directory>, and then I call "make TEST=nightly report.html". I would like to test only the files on SPEC, so I would like to know if any of you guys has an example makefile that tests only the SPEC benchmarks and could