search for: discrimination

Motivation: Many optimizations duplicate code. E.g. loop unroller duplicates the loop body, GVN duplicates computation, etc. The duplicated code will share the same debug info with the original code. For SamplePGO, the debug info is used to present the profile. Code duplication will affect profile accuracy. Taking loop unrolling for example: #1 foo(); #2 for (i = 0; i < N; i++) { #3 bar();

Hi all, Here I include an RFC for control flow sensitive AutoFDO (FS-AFDO). This is a joint work with David Li. Questions and feedback are welcome. Thanks, Rong ============= [RFC] Control Flow Sensitive AutoFDO (FS-AFDO) 1. Motivation AFDO profile is derived from PMU samples from running an earlier build binary. PMU samples are indexed by the IP addresses. An offline tool uses the debug

The impact to debug_line is actually not small. I only implemented the part 1 (encoding duplication factor) for loop unrolling and loop vectorization. The debug_line size overhead for "-O2 -g1" binary of speccpu C/C++ benchmarks: 433.milc 23.59% 444.namd 6.25% 447.dealII 8.43% 450.soplex 2.41% 453.povray 5.40% 470.lbm 0.00% 482.sphinx3 7.10% 400.perlbench 2.77% 401.bzip2 9.62% 403.gcc

Do you have an estimate of the debug_line size increase? I guess it will be small. David On Thu, Oct 27, 2016 at 11:39 AM, Dehao Chen <dehao at google.com> wrote: > Motivation: > Many optimizations duplicate code. E.g. loop unroller duplicates the loop > body, GVN duplicates computation, etc. The duplicated code will share the > same debug info with the original code. For

Is there prior art for this sort of thing (in GCC, for example) - I take it this isn't the first time this has come up as a problem for profile accuracy? (so it'd be helpful to know prior solutions to this (& if we're not doing whatever was done before, what it is about our situation that's different, etc), or why it hasn't been a problem, etc) On Thu, Oct 27, 2016 at

Hello everyone, I am totally new to the R program. I have had a look at some pdf documents that I downloaded and that explain how to do many things in R; however, I still cannot figure out how to do what I want to do, which is to perform Discriminant Correspondence Analysis on a rectangular matrix of data that I have in an Excel file. I know R users frown upon Excel and recommend converting Excel

The large percentages are from those tiny benchmarks. If you look at omnetpp (0.52%), and xalanc (1.46%), the increase is small. To get a better average increase, you can sum up total debug_line size before and after and compute percentage accordingly. David On Thu, Oct 27, 2016 at 1:11 PM, Dehao Chen <dehao at google.com> wrote: > The impact to debug_line is actually not small. I only

Hi Dehao, This is definitely an important problem, thanks for writing this up! There is a related problem that I think we can address at the same time: When we multiversion code, for example when we use runtime checks to enable the creation of a vectorized loop while retaining the scalar loop, and then we collect profiling data, we should be able to recover the relative running time of the

Hi Rong, This is a very interesting proposal. We've also observed profile quality degradation from CFG destructive pass like loop rotate, and I can see how this proposal would help improve quality of profile that drives later optimization passes in the pipeline. I have a few questions. * How does this affect today's AutoFDO? Specifically, can users upgrade compiler with FS-AutoFDO

It looks like the example doesn't use the encoding described in the text? Assume that the discriminator is uint32. The traditional discriminator is less than 256, let's take 8 bit for it. For duplication factor (type 1 duplication), we assume the maximum unroll_factor * vectorize_factor is less than 256, thus 8 bit for it. For unique number(type 2 duplication), we assume code is at most

----- Original Message ----- > From: "Dehao Chen" <dehao at google.com> > To: "Hal Finkel" <hfinkel at anl.gov> > Cc: "Xinliang David Li" <davidxl at google.com>, "llvm-dev" > <llvm-dev at lists.llvm.org> > Sent: Tuesday, November 1, 2016 11:43:41 AM > Subject: Re: [llvm-dev] (RFC) Encoding code duplication factor

----- Original Message ----- > From: "Dehao Chen" <dehao at google.com> > To: "Hal Finkel" <hfinkel at anl.gov> > Cc: "llvm-dev" <llvm-dev at lists.llvm.org>, "Xinliang David Li" > <davidxl at google.com> > Sent: Tuesday, November 1, 2016 8:24:30 PM > Subject: Re: [llvm-dev] (RFC) Encoding code duplication factor in

Dear All This is more of a statistics question than a question about help for R, so forgive me. I am using lda from the MASS package to perform linear discriminant function analysis. I have 14 cases belonging to two groups and have measured each of 37 variables. I want to find those variables that best discriminate between the two groups, and I want to visualise that and create a

Discussed with Hal, Adrain and Paul offline at the llvm dev meeting today. * trip count is not enough for vectorization, there is runtime check that might go false, which can be reflected in profile that we may want to preserve. * simply recording these context-profile may cause problems to iterative-sample-pgo. i.e. when you find a loop's vectorized version no executed (due to runtime

In many cases, the line-table fussing to improve autoFDO/sample-PGO would also likely help the debugging experience for optimized code, certainly in cases where line attribution is inherently ambiguous. In those cases, I have no problem with Just Doing It. Something likely to pad the line table to benefit profiling without similarly benefiting debugging… that's probably worth inventing a

----- Original Message ----- > From: "Dehao Chen" <dehao at google.com> > To: "Hal Finkel" <hfinkel at anl.gov> > Cc: "Xinliang David Li" <davidxl at google.com>, "llvm-dev" > <llvm-dev at lists.llvm.org> > Sent: Tuesday, November 1, 2016 1:24:01 PM > Subject: Re: [llvm-dev] (RFC) Encoding code duplication factor in

Hi everyone, I'm trying to solve a problem about how to get the Fisher's discriminant functions of a "lda" (linear discriminant analysis) object, I mean, the object obtained from doing "lda(formula, data)" function of the package MASS in R-project. This object gives me the canonical linear functions (n-1 coefficients matrix of n groups at least), and only with this

As illustrated in the above example, it is not like "vectorization has a distinct bit". All different optimizations make clones of code which will be labeled by UIDs represented by N (e.g. 8) bits. In this way, the space will be capped by the number of clones all optimizations have made, instead of # of optimizations that has applied. And it will be capped at 2^N-1. The cons of using uid

----- Original Message ----- > From: "Dehao Chen" <dehao at google.com> > To: "Hal Finkel" <hfinkel at anl.gov> > Cc: "llvm-dev" <llvm-dev at lists.llvm.org>, "Xinliang David Li" > <davidxl at google.com> > Sent: Tuesday, November 1, 2016 6:41:29 PM > Subject: Re: [llvm-dev] (RFC) Encoding code duplication factor in

On Jan 1, 2009, at 6:25 AM, Jon Harrop wrote: >> Exactly. I'm not especially interested in C-style unions, I'm >> interested >> in discriminated unions. But the actual discriminator field is easily >> represented in LLVM IR already, so there's no need to extend the IR >> to >> support them. That's why I am only asking for C-style union