similar to: [LLVMdev] Interested in working on Predictive Commoning

Yin Ma wrote: > Hello, > > It seems there is the no redundancy elimination over loop iterations, such > > as predictive communing or scalar replacement in LLVM. They are quite > usefully > > for computation code. I am wondering if any party is working or has plan to > > implement those optimizations? I don't know of any, but I was wondering if you could point me

There is a non-internal techpub for this somewhere, though it escapes me. My recollection is that TPO (the middle end of XLC) had moved on to 2nd order predictive quite a while ago. At least, the implementation I saw 6+ years ago in TPO was much closer to what is described in the 2nd order paper. On Sat, Dec 1, 2012 at 8:36 PM, Krzysztof Parzyszek <kparzysz at codeaurora.org> wrote: >

Thanks Arpith for this pointer but it did not contain what I was looking for. However, I solved my issue by finding that the BitCastInst used when outlining OpenMP parallel region offers a method for retrieving the original value within the bitcast via stripPointerCasts(). So what I did to retrieve omp_outlined..45 is: //bitcast is the third argument of __kmpc_fork_call Value *vcall =

On 12/1/2012 7:10 PM, Nick Lewycky wrote: > > I don't know of any, but I was wondering if you could point me to the > paper that describes predictive commoning? I could only find the > second-order predictive commoning paper, which if I understand correctly > is a much newer and different algorithm. I think the original paper was some internal IBM publication. The idea is

I'm curious if any of the core R developers have considered the possibility of hosting R (v2?) on the parrot common runtime environment. Perl6 will generate byte-code for parrot, as will some future version of python. I can imagine both drawbacks and advantages. Some advantages would be fast byte-code execution and freely mixing perl, python and R modules. Anyone looked into this? Tim

On 12/1/2012 10:38 PM, Daniel Berlin wrote: > There is a non-internal techpub for this somewhere, though it escapes me. > My recollection is that TPO (the middle end of XLC) had moved on to > 2nd order predictive quite a while ago. > > At least, the implementation I saw 6+ years ago in TPO was much closer > to what is described in the 2nd order paper. TPO never had any other

On Sun, Dec 2, 2012 at 7:48 AM, Krzysztof Parzyszek <kparzysz at codeaurora.org> wrote: > On 12/1/2012 10:38 PM, Daniel Berlin wrote: >> >> There is a non-internal techpub for this somewhere, though it escapes me. >> My recollection is that TPO (the middle end of XLC) had moved on to >> 2nd order predictive quite a while ago. >> >> At least, the

I see some code in R to attach finalizers to external pointer references (Register[C]Finalizer). Anyone have an example of how to code the finalizer? R_RegisterCFinalizer accepts a C function, but I can't see how to get it to operate on the pointer, since the pointer is not passed to the function when its called. RegisterFinalizer takes an R closure, but how is it called from R (tried

Hi all, It seems like there is a possible race on delalloc_bytes in btrfs in kernel 2.6.33.Please examine this issue and let me know if this is a problem or not.The race could occur between stat system call and writeback kernel thread. The lines are fs/btrfs/inode.c 5422 fs/btrfs/inode.c 1378 The stack traces are For file line fs/btrfs/inode.c 5422 btrfs_getattr+0x141/0x15e

Thanks for your reply. Yes, I was about to recurse over the use list of the argument in the called function. I did not want to pursue that because with this solution, I am going to implement the interprocedural part myself and was wondering if that was not already done. I was not also 100% sure that this will work for any type of arguments. If, based on your response, this is my only solution

Hello, I have been using the USE class to access the use-def chains of different values. However, what I have noticed is that the set of users of a particular value is limited for the appearance of that variable in the current function. How can I get the interprocedural use of a particular value? For example, if a variable *a* is used as an argument in a function call *foo*, the USE analysis

Hi, A question. When I run gelman.diag and gelman.plot with mcmc lists obtained from MCMCregress, the results are following. > post.R <- MCMCregress(Size~Age+Status, data = data, burnin = 5000, mcmc = 100000, + thin = 10, verbose = FALSE, beta.start = NA, sigma2.start = NA, + b0 = 0, B0 = 0, nu = 0.001, delta = 0.001) > post1.R <- MCMCregress(Size~Age+Status, data

Hi Everyone, I posted a similar question about a week ago, but haven't gotten any replies -- I'm afraid that's because my previous question was too vague. Let me try again with a more specific question, and I hope someone can help. NOTE, I know I should be using the newer lme4 package, I just haven't had a chance to update my version of R yet, so the question below relates

Hi, is there R function that does factor analysis? Thanks in advance. Kenny Ye Assistant Professor Department of Applied Math and Statistics SUNY at Stony Brook Stony Brook, New York 11794-3600 (516)632 9344 -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info",

The latest version of `MCMCvis` is now available on CRAN. `MCMCvis` is an R package used to visualize, manipulate, and summarize MCMC output. MCMC output may be derived from Bayesian model output fit with JAGS, Stan, or other MCMC samplers. Improvements since the last CRAN release (0.8.1) include: * ability to calculate and plot prior posterior overlap * ability to return number of effective

Hello, The third argument of the OpenMP runtime call __kmpc_fork_call is the outlined function resulting from omp parallel directive; this is of type kmpc_micro. I can get this argument as an LLVM value by using (callInst->getArgOperand...), but once I get this value, i am stuck in knowing its class type. An example of this value can be: void (i32*, i32*, ...)* bitcast (void (i32*, i32*,

When I use the function cancor of mva package, I found that it doesn't work when one of the matrix has only one column, or both have only one column. The function in Splus 5 with the same name works under those situations. The version of R I am using is 0.63.2 (released on Jan., 1999) on Solaris. Kenny Ye Assistant Professor Department of Applied Math and Statistics SUNY at Stony Brook

I seem to remember something about an "internal" keyword for R doc files. How do we use it? Tim -- Timothy H. Keitt Department of Ecology and Evolution State University of New York at Stony Brook Stony Brook, New York 11794 USA Phone: 631-632-1101, FAX: 631-632-7626 http://life.bio.sunysb.edu/ee/keitt/ -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-

Its not clear from the documentation whether one should do \keyword[key1, key2} or \keyword{key1} \keyword{key2} in .Rd files. I believe the latter is correct. (At least 'R CMD check' complains when given the first version.) It would help to make this explicit in the R extension writing docs. Tim -- Timothy H. Keitt Department of Ecology and Evolution State University of New

I have an application where I need to preserve object attributes across calls to 'lapply'. The current definition is: lapply <- function (X, FUN, ...) { FUN <- match.fun(FUN) if (!is.list(X)) X <- as.list(X) rval <- .Internal(lapply(X, FUN)) names(rval) <- names(X) return(rval) } Would it make sense to replace names(rval) <- names(X)