similar to: [LLVMdev] some of Basic blocks are deleted after running embedded ExecutionEngine

Dear Dibyendu and Mobi, Thanks for your help! :-) I finally figure it out. The solution is really simple. I just need to generate a new bitcode file with the following command: ----- opt -mem2reg -indvars test1.bc -o test2.bc ----- Then the load/store for induction variables will be removed and replaced by PHI instructions and all remaining load/store instructions are those I am interested in. I

Dear Dibyendu, Thanks for your response. :-) > If you are looking for only dependences which are inter-iteration (dependence distance != 0 ) you can do a post-pass on the ld/st addresses collected Yes, I am more interested in inter-iteration dependence. Could you provide more information or some links on post-pass approach? I have no idea on your method. :-) > eliminate such

Hi LLVM-ers, I try to develop my custom dynamic data dependence tool (focusing on nested loops), currently I can successfully get the trace including load/store address, loop information, etc. However, when I try to analyze dynamic data dependence based on the pairwise method described in [1], the load/store for iteration variables may interfere my analysis (I only care about the load/store for

Hi, all; I got an unusual error while using mclust library 2.1-7 on R 2.0.1. When I tried to run Mclust(rnorm(100)), I got segmentation fault error. Does anyone know what causes this problem? I came across the same problem even when I tried to run the example shown in Mclust help using iris data. Thanks in advance, Tae-Hoon Chung -------------------------------------------------- Tae-Hoon Chung

Basically, the `Detail' section of ?tune says it all: Details: As performance measure, the classification error is used for classification, and the mean squared error for regression. ... Andy > From: Tae-Hoon Chung > > Hi, all; > > Basically, the subject contains the all information I need to know. > In e1071 library, there are functions to tune parameters.

Hi, I am looking for Chinese Japanese and Korean tokenizer that could can be use to tokenize terms for CJK languages. I am not very familiar with these languages however I think that these languages contains one or more words in one symbol which it make more difficult to tokenize into searchable terms. Lucene has CJK Tokenizer ... and I am looking around if there is some open source that we

Hi, All; I have a question. In R, what is the best way to make R idle for a while and try something again later? For example, suppose there is an R job which accesses a file that may be shared with other active jobs. So when the file is being accessed by other job, your job will not be able to access the file and your job will crash because of that. To avoid this, you want your job to try to

Hi, I have a question about how to get the residuals and estimations of the parameters in my program. For example, x<-c(1,2,3,4,5,6) y<-c(2.9, 1.24, 1.71, 2.989358, 1.455979, 1.4) nls(y ~ a+sin(b*x),start=list(a=2.2,b=1.8),trace=TRUE) I can get the estimation of a, b and residuals from the output but I need them to be vectors in order to do futhur calculations in a program. Is there any

Hi all, A few of us have been discussing whether we should have a Xapian social gathering of some kind. The current idea is meeting up in a pub in London some time in autumn for drinks and food. However all of this really depends on who might be able to come! It would be a chance to meet other Xapian enthusiasts in an informal social setting and talk about all things search-related (and

Dear LLVMdev, We have a question regarding the semantics of the undefined value in LLVM IR. As far as we understand, in order to give a semantics to the undefined value, the semantics of LLVM IR is generalized to a rather unusual set-based semantics. More specifically, the notion of value in LLVM IR is generalized to a set of values, rather than a single value. Then the undefined value is simply

Dear all, I have setup a samba PDC on a debian lenny machine and everything works fine. Recently, i want to integrate Samba with LDAP. When I use smbldap-populate, the corresponding structures are created in the LDAP tree, however, the unix groups are not created at all! smbldap-populate did not complaint anything. I have tested by using smbldap-useradd. The same problem appears. What are the

Hi, all; Our group is planning to develop a web-based analysis package with R. I have some questions. (1) Can we use R as a daemon-like way such that, after invoking R and making it run in a kind of background mode, just put some R script into R using some pipe-like mechanism and retrieve the result out of it? (2) If this is possible, then how can we do it? If this is not possible, then what

Hi, all. Something strange happened to my ESS. I use Mac OS X. When R is used in stand alone mode, it recognizes external libraries like "affy", "e1071" etc. However, when R is invoked through ESS in Emacs, it produces errors like: Error in library(affy) : There is no package called 'affy' Interestingly, I remember I successfully used these libraries through ESS in

Hi, I heard that Theora will be used on OLPC. Is anyone working on optimization for OLPC? Regards, Chih-Chung Chang

Dear R users, Suppose I have 2 R script files: 'test1.R' and 'test2.R' and one R file 'main.R' which sources each of them. I wonder if there is a way to skip rest of code in 'test1.R' once a condition is met but still continue to run rest of script from the 'main'R' file. A simple example is shown below. I hope only 'message1' and

hello, I have finished reading the papers, and i think it is time to design my project. First step will be determine the input characters are Chinese. i see the past post that cjk-tokenizer is just dealing with UTF-8 and unicode, but i see some other code system such as gbk and big5. i am wondering that should i just deal with UTF-8 and unicode?

Hi, Attached is a patch against 1.0alpha7 to speed up Theora decoding. It is about 15~20% faster in my test. It consists of following things: * MMX loop filter based on Rudolf Marek's patch in http://lists.xiph.org/pipermail/theora-dev/2005-August/002838.html * MMX IDCT based on Rudolf Marek's patch in http://lists.xiph.org/pipermail/theora-dev/2005-July/002816.html and the code in

Hello. By way of background, I am running out of memory when attempting to normalize the data from 160 affymetrix microarrays using justRMA (from the affy package). This is despite making 6 gigabytes of swap space available on our sgi irix machine (which has 2 gigabytes of ram). I have seen in various discussions statements such as "you will need at least 6 gigabytes of memory to normalize

Hi, All; When I generate a "special" paper postscript image larger than "a4" or "letter" using R, I can only see one-page portion of all image, of course. What will be the simple solution for this? Is there any way I can set the bounding box information on the image? Or any other suggestions? Thanks in advance; Tae-Hoon Chung

I''ve just started to recently look at CBQ and HTB, but I''ve been wondering whether or not something like the following is possible: I''ve got two external links: +--+ | |-----ISP1 |FW| | |-----ISP2 +--+ One link has a bandwidth cap, and the other doesn''t. What I want to do is, punt all outbound traffic out via ISP1