similar to: [LLVMdev] Question: Motivation of the semantics of the undefined value?

I think calling the attribute "used" is confusing. I'd suggest the following: "not_poison": If an argument is marked not_poison, and the argument is poison at runtime, the call is instant UB. Whether an argument is poison is checked after the rules for other attributes like "nonnull" and "align" are applied. This makes it clear that the IR semantics

Hi Johannes, >> Not sure the semantics of "used" you propose is sufficient. AFAIU the >> proposal, "used" could only be used in cases where the function will >> always trigger UB if poison is passed as argument. The semantics of >> attributes is usually the other way around, since function calls need >> to have UB as strong as the worst

>> A use of freeze is to enable speculative execution. For example, loop >> switching does the following transformation: >> while (C) { >> if (C2) { >> A >> } else { >> B >> } >> } >> => >> if (C2) { >> while (C) >> A >> } else { >> while (C) >> B >> } >>

>> 1) a bit in the output is poison if flipping any set of poison bits >> in the >> input may yield different values for the output bit. > > This is the definition I'd expect. > >> For example (bitwise notation): >> ppp * 000 == 000 (since flipping any of the poison bits cannot yield >> a >> result other than zero) >> 00p + 00p == 0pp

Hi, Thanks everybody that showed up in our talk at the LLVM dev meeting and to those that provided feedback so far. The slides are already online: http://llvm.org/devmtg/2016-11/Slides/Lopes-LongLivePoison.pdf The main question that some people raised was whether we could have bitwise poison instead of value-wise poison, since that semantics seems to be more natural as values continue to be just

Hi, Over the past few years we've been trying to kill poison somehow. There have been a few proposals, but they've all failed to pass the bar and/or to gather significant support (my own proposals included). We (David, Gil, John, Juneyoung, Sanjoy, Youngju, Yoonseung, and myself) have a new proposal to kill undef instead and replace it with poison + a new 'freeze' instruction. We

I''ve been wrestling with this all night, I''m hoping someone can help. I followed the exact steps in: http://wiki.rubyonrails.org/rails/pages/HowtoUseRailsWithSubversion ..but when I open a new ''Checkout project from SVN'' in RadRails, it opens up the second level dirs as the project dirs (ie. app, log, script, etc), leaving me with a mess of projects. I redid

Hi Rhelpers, I would like to have some help with a plot which is beyond my capabilities. This plot that I am seeking involves an overlay of two different barcharts and one xyplot. The code that I have used is the following : #save(df1,file="M:\\KBR\\df1.RData") load(file="M:\\KBR\\df1.RData") # df1$Year.ord created to obtain the right order i.e. 2015M < 2015K

This small patch extends configure_socket_options to support FreeBSD''s accf_http(9), which defers accept() until there''s a full HTTP request to read. Seems to work fine on 6.1-STABLE. DragonflyBSD should work too provided the /freebsd/ line is modified to match it. accf_http(9): http://www.freebsd.org/cgi/man.cgi?query=accf_http&sektion=9 -- Thomas

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

Hello Railists and Rubyists, I know that the topic of handling credit card numbers has been expounded before on this list (I''ve searched the archives) but I want to ask the question in a slightly different way, and get your feedback: Previous discussions on the list center around what to do with the card number *after* you have processed the credit card with a payment gateway

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 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