similar to: A problem of is.list function

Under platform x86_64-pc-linux-gnu arch x86_64 os linux-gnu system x86_64, linux-gnu status major 2 minor 13.1 year 2011 month 07

Hi folks, I am working with repeated measures data and I ran into issues where the paired t-test results did not match those obtained by employing glht() contrasts on a lme model. While the lme model itself appears to be fine, there seems to be some discrepancy with using glht() on the lme model (unless I am missing something here). I was wondering if someone could help identify the issue. On

Hi. macosx 10.6.8 With R-2.13.1 and also revision 56948 I get the following repeatable segfault: wt118:~% R --vanilla --quiet > R.Version() $platform [1] "x86_64-apple-darwin9.8.0" $arch [1] "x86_64" $os [1] "darwin9.8.0" $system [1] "x86_64, darwin9.8.0" $status [1] "" $major [1] "2" $minor [1] "13.1" $year [1]

On Oct 17, 2013, at 10:51 PM, lian li <lianli at gmail.com> wrote: > Thanks, Chris. > > We are interested in contributing it to LLVM itself. Our manager > agrees to commit resources for maintenance needs if it is accepted by > the community. This is great. Please make sure Oracle legal sign off on explicitly granting LLVM the use of the patents associated with the work. On

Hi Evan, We did an experiment using the LLVM test suite: we compare the overhead of using our analysis to the LLVM default, both with -O2 option. The overall overhead of compiling the whole test suite using our analysis is 36.5%. The biggest overhead is observed in "SingleSource/Benchmarks/Misc/flops-5", where we are 5 times slower: 0.07s (with our analysis) compared to

On Fri, Oct 18, 2013 at 7:27 AM, lian li <lianli at gmail.com> wrote: > Hi Daniel, > > I want to clarify that our analysis is not based on CFL-reachability. > We apply CFL-reachability to matching context information where the > exist from a function to a call-site must match > the entry from the corresponding call-site. Yes, sorry, I pulled the wrong quote, it was late.

Hi Daniel, I want to clarify that our analysis is not based on CFL-reachability. We apply CFL-reachability to matching context information where the exist from a function to a call-site must match the entry from the corresponding call-site. The problem is a simple balanced parentheses problem in CFL-reachability, and it can be computed efficiently. The paper you mentioned is a very nice paper

On Tue, Oct 22, 2013 at 9:55 AM, Hal Finkel <hfinkel at anl.gov> wrote: > ----- Original Message ----- > > Hi Evan, > > > > We did an experiment using the LLVM test suite: we compare the > > overhead of using our analysis to the LLVM default, both with -O2 > > option. > > It might also be interesting to try with -O3; I don't know if we have any

----- Original Message ----- > Hi Evan, > > We did an experiment using the LLVM test suite: we compare the > overhead of using our analysis to the LLVM default, both with -O2 > option. It might also be interesting to try with -O3; I don't know if we have any significant vectorizable loops in the test suite with a large number of arrays, but if we do, this kind of analysis

Thanks, Chris. We are interested in contributing it to LLVM itself. Our manager agrees to commit resources for maintenance needs if it is accepted by the community. Regards, Lian On Fri, Oct 18, 2013 at 3:43 PM, Chris Lattner <clattner at apple.com> wrote: > > On Oct 17, 2013, at 5:20 PM, lian li <lianli at gmail.com> wrote: > > Hi All, > > This is Lian Li from

I notice you guys formulate your CFL reachability problem as a balanced parentheses problem. What algorithm do you use to solve it? Are you aware of recent work that comes up with linear time and n log n time algorithms to solve this class of problems: http://www.cse.cuhk.edu.hk/lyu/_media/paper/pldi2013.pdf In particular, the time bound from the paper: "However, if we need the precise

Hi Hal, Thanks for your interest. We tested with the following existing compiler optimizations in LLVM with SPECINT2006 benchmarks: -dse (dead store elimination), -gvn (global value numbering), -licm (loop invariant code motion), -bb-vectorize (basic block vectorization), -memcpyopt (memcpy optimization), -sink (code sinking), -loop-idom (recognize loop idioms), -argpromotion (argument

Good Day, I fit a multivariate linear regression model with 3 dependent variables and several predictors using the lm function. I would like to use stepwise variable selection to produce a set of candidate models. However, when I pass the fitted lm object to step() I get the following error: Error from R: Error in drop1.mlm(fit, scope$drop, scale = scale, trace = trace, k = k, : no

Hi Lian, I am certainly interested in seeing this; do you have performance numbers (compile time)? Also, can you share more information about the promising optimization results you mentioned? Thanks, Hal ----- Original Message ----- > Hi All, > > This is Lian Li from Oracle Labs in Brisbane Australia. > > We have developed a precise and highly efficient pointer analysis >

On Oct 17, 2013, at 5:20 PM, lian li <lianli at gmail.com> wrote: > Hi All, > > This is Lian Li from Oracle Labs in Brisbane Australia. > > We have developed a precise and highly efficient pointer analysis > framework on top of LLVM, The approach is flow, context, and field > sensitive, details are described in the two papers below: > > "Boosting the

Hi All, This is Lian Li from Oracle Labs in Brisbane Australia. We have developed a precise and highly efficient pointer analysis framework on top of LLVM, The approach is flow, context, and field sensitive, details are described in the two papers below: "Boosting the performance of flow-sensitive points-to analysis using value flow" (in ESEC-FSE 2011), and "Precise and

On Wed, Oct 23, 2013 at 3:07 AM, Chandler Carruth <chandlerc at google.com> wrote: > On Tue, Oct 22, 2013 at 9:55 AM, Hal Finkel <hfinkel at anl.gov> wrote: >> >> ----- Original Message ----- >> >> It might also be interesting to try with -O3; I don't know if we have any >> significant vectorizable loops in the test suite with a large number of

There seems to be a bug in "image" in R 13.1 (on windows 32 & 64 bits) and on R-devel that is not present in R 13.0 and before. The below creates a plot with many white lines, horizontal and vertical, more or less regularly spaced. The effect is particularly dramatic when the plotting window is made very small. image(t(volcano)[ncol(volcano):1,]) Robert Hijmans University of

Dear R-Users, I think I have encountered a potential bug (or at least unwanted behavior), but I'm not sure so I wanted to post here first. Lately I've been encountering an error when running a package I put together. I have my package set up to check for updates when it loads but this error occurs and stops the package from loading: Error : .onLoad failed in loadNamespace() for

On Oct 22, 2013, at 1:29 PM, llvmdev-request at cs.uiuc.edu wrote: > Send LLVMdev mailing list submissions to > llvmdev at cs.uiuc.edu > > To subscribe or unsubscribe via the World Wide Web, visit > http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev > or, via email, send a message with subject or body 'help' to > llvmdev-request at cs.uiuc.edu > > You can