similar to: [LLVMdev] [llvm-commits] [PATCH] BasicBlock Autovectorization Pass

On Sat, 2011-10-29 at 15:16 -0500, Hal Finkel wrote: > On Sat, 2011-10-29 at 14:02 -0500, Hal Finkel wrote: > > On Sat, 2011-10-29 at 12:30 -0500, Hal Finkel wrote: > > > Ralf, et al., > > > > > > Attached is the latest version of my autovectorization patch. llvmdev > > > has been CC'd (as had been suggested to me); this e-mail contains > >

On Sat, 2011-10-29 at 14:02 -0500, Hal Finkel wrote: > On Sat, 2011-10-29 at 12:30 -0500, Hal Finkel wrote: > > Ralf, et al., > > > > Attached is the latest version of my autovectorization patch. llvmdev > > has been CC'd (as had been suggested to me); this e-mail contains > > additional benchmark results. > > > > First, these are preliminary

On Sat, 2011-10-29 at 12:30 -0500, Hal Finkel wrote: > Ralf, et al., > > Attached is the latest version of my autovectorization patch. llvmdev > has been CC'd (as had been suggested to me); this e-mail contains > additional benchmark results. > > First, these are preliminary results because I did not do the things > necessary to make them real (explicitly quiet the

On 2016-05-25 19:13, Kelly Lesperance wrote: > Hdparm didn?t get far: > > [root at r1k1 ~] # hdparm -tT /dev/sda > > /dev/sda: > Timing cached reads: Alarm clock > [root at r1k1 ~] # Hi Kelly, Try running 'iostat -xdmc 1'. Look for a single drive that has substantially greater await than ~10msec. If all the drives except one are taking 6-8msec, but one is very

Dear R-helpers, I want to make a series of boxplots on several numeric univariates with two group variables (species and population, population nested in species, and with population as the X-axis). In order to get a proper order of the individual populations in X-axis, I need to assign a wanted order to the factor (population). I used the levels() function to do this assignment, but it seemed

is there a way to use the actual index value for plotting zoo objects this is the way that the index is set up and a sample range of what I would like to plot 01/01/06 00:00:00 - 01/01/06 23:45:00 { library(zoo) # chron library(chron) fmt.chron <- function(x) { chron(sub(" .*", "", x), gsub(".* (.*)", "\\1:00", x)) }} x <- structure(c(15.57, 15.5,

Hello, Using the following lines of code, I created the following graph:

Hello, Using the following lines of code, I created the following graph:

The Milwaukee Ruby User''s Group is pleased to announce it''s first meeting. Date: June 7, 2006 - Wednesday Time: 5:30 - 7:30 p.m. Location: SpiderLogic, Conference Room 1 on first floor 10000 West Innovation Dr. Milwaukee, WI 53266 Map ==> http://tinyurl.com/olprl Look for Medical College of Wisconsin / Wipfli sign outside Very Tentative Agenda: 5.30 -

Dear all, I'm writing my manuscript to publish after analysis my final data with ANOVA, ANCOVA, MANCOVA. In a section of my result, I did correlation of my data (2 categirical factors with 2 levels: Quantity & Quality; 2 dependent var: Irid.area & Casa.PC1, and 1 co-var: SL). But as some traits (here Irid.area) are significantly influenced by the covariate (standard length, SL), I

Hi, I am a total beginner with this whole thing so please have patience! I am trying to run an S-plus program with a certain line: spline(1:nrow(y), y[,1],n=100); This crashes with: Error: NAs in foreign function call (arg 8) Apparently, this is caused by the last command of spline: u <- seq(xmin, xmax, length.out = n) .C("spline_eval", z$method, length(u), x = u, y =

Dear List I hope you can help me: I?ve got a dataframe (df) within which I am looking for Peak Over Threshold values as well as the length of the events. An event starts when walevel equals 5.8 and it should end when walevel equals the lower threshold value (5.35). I tried ?clusters (?)? from ?evd package?, and varied r (see example) but it did not work for all events (again

I have a situation where lagged values of a time-series are used to predict future values. I have packed together the time-series and the lagged values into a data frame: > str(D) 'data.frame': 191 obs. of 13 variables: $ y : num -0.21 -2.28 -2.71 2.26 -1.11 1.71 2.63 -0.45 -0.11 4.79 ... $ y.l1 : num NA -0.21 -2.28 -2.71 2.26 -1.11 1.71 2.63 -0.45 -0.11 ... $ y.l2 : num

Hi 2 all ! I have problems with concurrent filesystem actions on a ocfs2 filesystem which is mounted by 2 nodes. OS=RH5ES and OCFS2=1.2.6 F.e.: If I have a LV called testlv which is mounted on /mnt on both servers and I do a "dd if=/dev/zero of=/mnt/test.a bs=1024 count=1000000" on server 1 and do at the same time a du -hs /mnt/test.a it takes about 5 seconds for du -hs to execute: 270M

This is very interesting work. Thanks. Quick questions / comments. 1. What's the algorithm complexity? 2. Please rename -vectorize to -bb-vectorize since this is a specific kind of vectorization. 3. Have you tried non-x86 targets? Does it work? If not, what would it take to make it work? I am concerned that LLVM currently does not have a cost model for vectorization. That's necessary for

On Tue, 2011-11-08 at 20:24 +0100, Tobias Grosser wrote: > On 11/08/2011 03:36 PM, Hal Finkel wrote: > > On Tue, 2011-11-08 at 12:12 +0100, Tobias Grosser wrote: > >> On 11/08/2011 11:45 AM, Hal Finkel wrote: > >>> I've attached the latest version of my autovectorization patch. > >>> > >>> Working through the test suite has proved to be a

On Nov 27, 2011, at 7:04 PM, Hal Finkel wrote: > On Sun, 2011-11-27 at 18:34 -0800, Evan Cheng wrote: >> This is very interesting work. Thanks. Quick questions / comments. >> >> 1. What's the algorithm complexity? > > Roughly, it is O(N^2), where N is the number of instructions in a basic > block, but: > - The run time only *really* grows with the number of

Dear Eliza, Try this: Lines1<-readLines(textConnection("1911.01.01?????? 7.87 1911.01.02?????? 9.26 1911.01.03?????? 8.06 1911.01.04?????? 8.13 1911.01.05????? 12.90 1911.02.06?????? 5.45 1911.02.07?????? 3.26 1911.03.08?????? 5.70 1911.03.09?????? 9.24 1911.04.10?????? 7.60 1911.05.11????? 14.82 1911.05.12????? 14.10 1911.06.13?????? 7.87 1911.06.14?????? 9.26

Hi, I started to play with the LoopVectorizer of LLVM trunk on the work-item loops produced by pocl's OpenCL C kernel compiler, in hopes of implementing multi-work-item work group autovectorization in a modular manner. The vectorizer seems to refuse to vectorize the loop if it sees multiple writes to the same memory object within the same iteration. In case of parallel loops such as the

On Thu, 2012-01-26 at 14:34 -0600, Sebastian Pop wrote: > On Tue, Jan 24, 2012 at 6:41 PM, Hal Finkel <hfinkel at anl.gov> wrote: > >> enabling vectorization gets the performance down by 80% on ARM. > >> I will prepare a reduced testcase and try to find out the reason. > >> As a first shot, I would say that this comes from the vectorization of > >> code