similar to: Compare test-suite benchmarks performance complied without TBAA, with default TBAA and with new TBAA struct path

Hello all, The performance comparison between Polly's Cloog and ISL code generator is posted on http://188.40.87.11:8000/db_default/v4/nts/59?compare_to=58&baseline=58 It seems their execution-time performance are comparable: Performance Regressions - Execution Time  (ISL over Cloog) MultiSource/Benchmarks/TSVC/ControlFlow-flt/ControlFlow-flt 8.49%

Now, we come to more evaluations on http://188.40.87.11:8000/db_default/v4/nts/recent_activity I mainly care about the compile-time and execution time impact for the following cases: pBasic (run 45): clang -O3 -load LLVMPolly.so pNoGenSCEV (run 44): clang -O3 -load LLVMPolly.so -polly-codegen-scev -polly -polly-optimizer=none -polly-code-generator=none pNoGenSCEV_nocan (run 47): same option

Hello all, I have done some basic experiments about Polly canonicalization passes and I found the SCEV canonicalization has significant impact on both compile-time and execution-time performance. Detailed results for SCEV and default canonicalization can be viewed on: http://188.40.87.11:8000/db_default/v4/nts/32 (or 33, 34) *pNoGen with SCEV canonicalization (run 32): -O3 -Xclang -load

Hi, Sean: I'm sorry I lie. I didn't mean to lie. I did try to avoid making a *BIG* change to the IPO pass-ordering for now. However, when I make a minor change to populateLTOPassManager() by separating module-pass and non-module-passes, I saw quite a few performance difference, most of them are degradations. Attacking these degradations one by one in a piecemeal manner is wasting

Hi Chris and others! I totally support any work in this direction. In the current state LNT’s regression detection system is too noisy, which makes it almost impossible to use in some cases. If after each run a developer gets a dozen of ‘regressions’, none of which happens to be real, he/she won’t care about such reports after a while. We clearly need to filter out as much noise as we can - and

Hi Folks, Looks like our LNT ARM buildbot with the vectorizer is running and producing good results. There are only 11 failures: FAIL: MultiSource/Applications/Burg/burg.execution_time (1 of 1104) FAIL: MultiSource/Applications/ClamAV/clamscan.execution_time (2 of 1104) FAIL: MultiSource/Applications/lemon/lemon.execution_time (3 of 1104) FAIL:

Summary ======= I'm planning on adjusting SimplifyCFG so that it doesn't turn two-entry phi nodes into selects until later in the pass pipeline, to give passes which can understand phis but not selects more opportunity to optimize. The thing I'm trying to do which made me think of doing this is described below, but from the benchmarking I've done it looks like this is overall a

Hi Renato, I noticed the bot yesterday. Thanks for working on this! On Feb 19, 2013, at 7:45 AM, Renato Golin <renato.golin at linaro.org> wrote: > Hi Folks, > > Looks like our LNT ARM buildbot with the vectorizer is running and producing good results. Do you have a base run with vectorization turned off? So we could see where we are degrading things? When you say good

Target: FreeBSD 6.2-STABLE on i386 autoconf says: configure:2122: checking build system type configure:2140: result: i386-unknown-freebsd6.2 [...] configure:2721: gcc -v >&5 Using built-in specs. Configured with: FreeBSD/i386 system compiler Thread model: posix gcc version 3.4.6 [FreeBSD] 20060305 [...] objdir != srcdir, for both llvm and gcc. Release build. llvm-gcc 4.2 from source.

Hi, LLVM 2.1-pre1 test results: Linux (SUSE) on x86 (P4) Release mode, but with assertions enabled LLVM srcdir == objdir # of expected passes 2250 # of expected failures 5 I ran the llvm-test suite on my desktop while I was also working on that PC, so don't put too much trust in the timing info. Especially during the "spiff" test the machine was swapping

On Fri, Sep 14, 2007 at 11:42:18PM -0700, Tanya Lattner wrote: > The 2.1 pre-release (version 1) is available for testing: > http://llvm.org/prereleases/2.1/version1/ > > [...] > > 2) Download llvm-2.1, llvm-test-2.1, and the llvm-gcc4.0 source. > Compile everything. Run "make check" and the full llvm-test suite > (make TEST=nightly report). > > Send

Target: FreeBSD 7.0-RC1 on amd64. autoconf says: configure:2122: checking build system type configure:2140: result: x86_64-unknown-freebsd7.0 [...] configure:2721: gcc -v >&5 Using built-in specs. Target: amd64-undermydesk-freebsd Configured with: FreeBSD/amd64 system compiler Thread model: posix gcc version 4.2.1 20070719 [FreeBSD] [...] objdir != srcdir, for both llvm and gcc. Release

Hello; I was able to build and install llvm(3.0) under Ubuntu 11.10 (using the ./configure script found under llvm source, and then make and make install). While configuring, I gave --prefix as a directory where I would like llvm to be installed. I did not give --with-llvmgccdir and the --enable-optimized argument to configure. Because 3.0 doesn't come with llvmgcc source/binaries and I

On Oct 20, 2009, at 6:02 AM, Duncan Sands wrote: > Hi Tanya, > >> 1) Compile llvm from source and untar the llvm-test in the projects >> directory (name it llvm-test or test-suite). Choose to use a pre- >> compiled llvm-gcc or re-compile it yourself. > > I compiled llvm and llvm-gcc with separate objects directories. > Platform is x86_64-linux-gnu. > Ok.

I'm concerned that we're focusing on one side of this.  Let me point out a few concerns w/changing the canonical form here: 1. LICM does not know how to hoist or sink regions.  It does know how to hoist and sink selects. 2. InstCombine has limited support for triangles/diamonds, but fairly extensive support for selects. 3. EarlyCSE and GVN do not know how to eliminate fully

Hi, Below you can see the updated benchmark results for the new SLP-vectorizer. As you can see, there is a small number of compile time regressions, a single major runtime *regression, and many performance gains. There is a tiny increase in code size: 30k for the whole test-suite. Based on the numbers below I would like to enable the SLP-vectorizer by default for -O3. Please let me know if you

Sorry for not posting sooner. I forgot to send an update with the results. I also have some benchmark data. It confirms much of what you posted -- binary size increase is essentially 0, performance increases across the board. It looks really good to me. However, there was one crash that I'd like to check if it still fires. Will update later today (feel free to ping me if you don't hear

Hi, Sorry for the delay in response. I measured the code size change and noticed small changes in both directions for individual programs. I found a 30k binary size growth for the entire testsuite + SPEC. I attached an updated performance report that includes both compile time and performance measurements. Thanks, Nadav On Jul 14, 2013, at 10:55 PM, Nadav Rotem <nrotem at apple.com>

Cool! What changes have you seen to generated code size? I'll take it for a spin on our benchmarks. On Sat, Jul 13, 2013 at 11:30 PM, Nadav Rotem <nrotem at apple.com> wrote: > Hi, > > LLVM’s SLP-vectorizer is a new pass that combines similar independent > instructions in a straight-line code. It is currently not enabled by > default, and people who want to

G'Day Tanya, Is it too late to bring in the following patches to fix some major brokenness in the AuroraUX tool chain for 2.6? http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Driver/Tools.cpp?r1=84468&r2=84469&view=diff&pathrev=84469 http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Driver/Tools.cpp?r1=84265&r2=84266&view=diff&pathrev=84266