search for: 17.30

Dear R users, I have a txt file of the form 10092007 24.62 24.31 24.90 11092007 19.20 23.17 22.10 13092007 24.71 27.33 23.10 14092007 27.33 27.90 24.10 15092007 28.22 28.55 24.30 16092007 28.53 29.24 27.40 17092007 24.19 30.64 26.80 18092007 22.60 20.62 28.40 19092007 8.89 1.70 14.70 20092007 21.27 2.92 17.30 21092007 22.38 24.72 8.80 22092007 23.94 24.73 20.40 23092007 25.33 25.12 22.60

Thank you for information. I’ll build clang without the hack and re-run the benchmark tomorrow. -Evgeny From: Sanjay Patel [mailto:spatel at rotateright.com] Sent: Sunday, January 22, 2017 8:00 PM To: Evgeny Astigeevich Cc: llvm-dev; nd Subject: Re: [InstCombine] rL292492 affected LoopVectorizer and caused 17.30%/11.37% perf regressions on Cortex-A53/Cortex-A15 LNT machines > Do you mean to

Confirm there is no change in IR if the hack is disabled in the sources. David wrote that these instructions are created by SCEV. Are other targets affected by the changes, e.g. X86? Kind regards, Evgeny Astigeevich Senior Compiler Engineer Compilation Tools ARM From: Sanjay Patel [mailto:spatel at rotateright.com] Sent: Sunday, January 22, 2017 10:45 PM To: Evgeny Astigeevich Cc: llvm-dev; nd

Hi Sanjay, The benchmark source file: http://www.llvm.org/viewvc/llvm-project/test-suite/trunk/SingleSource/Benchmarks/Shootout/sieve.c?view=markup Clang options used to produce the initial IR: clang -DNDEBUG -O3 -DNDEBUG -mcpu=cortex-a53 -fomit-frame-pointer -O3 -DNDEBUG -w -Werror=date-time -c sieve.c -S -emit-llvm -mllvm -disable-llvm-optzns --target=aarch64-arm-linux Opt options: opt -O3

Hi folks, I've been playing around with an Asterisk server in my office for a few weeks now, and I've got it pretty much nailed down the way I want it, which is nice. One of the features I'm using is the ability to switch different contexts in & out of the dialplan on a schedule. So, for example, I've got the "official" tel number ringing my desk phone between

On Tue, Jan 24, 2017 at 1:20 PM, Sanjay Patel via llvm-dev < llvm-dev at lists.llvm.org> wrote: > > I started looking at the log files that you attached, and I'm confused. > The code that is supposedly causing the perf regression is created by the > loop vectorizer, right? Except the bad code is not in the "vector.body", so > is there something peculiar about

Hi Sanjay, Thank you for your analysis. It’s interesting why the x86 machine is not affected. Maybe the x86 backend is smarter than the AArch64 backend, or it might be micro-architectural differences. I don’t mind to keep the changes on trunk. What I’d like to see is who will/should be involved in solving the issue. What kind of help/support is needed? Should we (ARM Compilation Tools) start

> On Jan 23, 2017, at 3:48 PM, Sanjay Patel via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > All targets are likely affected in some way by the icmp+shl fold introduced with r292492. It's a basic pattern that occurs in lots of code. Did you see any perf wins on your targets with this commit? > > Sadly, it is also likely that many (all?) targets are negatively

Hi, We found that today's 17.30%/11.37% performance regressions in LNT SingleSource/Benchmarks/Shootout/sieve on LNT-AArch64-A53-O3__clang_DEV__aarch64 and LNT-Thumb2v7-A15-O3__clang_DEV__thumbv7 (http://llvm.org/perf/db_default/v4/nts/daily_report/2017/1/20?filter-machine-regex=aarch64%7Carm%7Cthumb%7Cgreen) are caused by changes [rL292492] in InstCombine: https://reviews.llvm.org/D28406

> On Jan 24, 2017, at 7:18 AM, Sanjay Patel <spatel at rotateright.com> wrote: > > > > On Mon, Jan 23, 2017 at 10:53 PM, Mehdi Amini <mehdi.amini at apple.com <mailto:mehdi.amini at apple.com>> wrote: > >> On Jan 23, 2017, at 3:48 PM, Sanjay Patel via llvm-dev <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote: >>

Dear R-users, I have to deal with (txt) files of the form 10092007 24.62 24.31 24.90 11092007 19.20 23.17 22.10 13092007 24.71 27.33 23.10 14092007 27.33 27.90 24.10 15092007 28.22 28.55 24.30 16092007 28.53 29.24 27.40 17092007 24.19 30.64 26.80 18092007 22.60 20.62 28.40 19092007 18.89 21.70 14.70 20092007 21.27 22.92 17.30 21092007 22.38 24.72 18.80 22092007 23.94 24.73 20.40 23092007 25.33

Hi there, This seems like it should be simple. I have a data frame of climate data sampled every 10 min. I want to average the entire data frame into 30 min values (i.e., one value for each half hour). Functions like running.mean give me a moving average but I want to reduce the size of the entire frame.. Any ideas how? Cheers! Example of my data timestamp Voltage LwTempDownelling

Dear Sirs, I am Professor at Indira Gandhi Krishi Vishwavidyalaya, Raipur, Chhattisgarh, India. While taking classes, I found the *by() *function producing following error when I use FUN=mean or median and some other functions, however, FUN=summary works. Given below is the output of the example I used on a built-in dataset "mtcars", along with error message reproduced herewith: >

Hi, I'm a total beginner in R and this question is probably very simple but I've spent hours reading about it and can't find the answer. I'm trying to reshape a data table from long to wide format. I've tried reshape() and cast() but I get error messages every time and I can't figure why. In my data, I have the length of two fish from each family. My data table (called

Hi All, I have the following text file (mytextfile.txt) 738307 527178 714456 557955 #N/A 17.42 6.22 4.73 #N/A 17.3 6.23 4.75 #N/A 17.29 6.17 4.7 #N/A 17.07 6.12 4.6 #N/A 17.27 6.19 4.7 #N/A 17.72 6.4 4.78 #N/A 17.12 6.19 4.75 #N/A 17.07 6.15 4.65 #N/A 17.03 6.07 4.64 #N/A 17.38 6.13 4.7 #N/A 17.38 6.13 4.7 #N/A 17.38 6.13 4.7 #N/A 17.38 6.13 4.7 #N/A

I think you are not using the best function for what your intentions are. Try: > by(data=mtcars, INDICES=list(as.factor(mtcars$am)), FUN=colMeans) : 0 mpg cyl disp hp drat wt qsec vs 17.1473684 6.9473684 290.3789474 160.2631579 3.2863158 3.7688947 18.1831579 0.3684211 am gear carb 0.0000000

[originally reported for ZFS on FreeBSD but Pawel Jakub Dawid says this problem also exists on Solaris hence this email.] Summary: on ZFS, overhead for reading a hole seems far worse than actual reading from a disk. Small buffers are used to make this overhead more visible. I ran the following script on both ZFS and UF2 filesystems. [Note that on FreeBSD cat uses a 4k buffer and md5 uses a 1k

Dear All, Thanks for your help. However, I would like to draw your attention to the following: Actually, I was replicating the Example 2.3, using the dataset "brainsize.txt" given in Section 2.3.3 ("Summarize by group") at page 55, of a famous book "R by Example" written by "Jim Albert and Maria Rizzo" published in Springers (2012) in a Use R! Series. The

Hello, Tatu > Is that correct? I think it's just something to be aware of. Currently we're aggressively merging globals by default. Do you think it will be better to provide special flag to control this behavior? -- WBR, Anton Korobeynikov

Good morning, I'm trying to write a function to assign the mean separation grouping letters to the factors of an experiment. A client wants this so I thought I'd write my own function for future use. But I'm having trouble doing it because I don't understand the logic of the assignment. As a general example, consider the following data and the results from SAS and its weak