similar to: [RFC] Remarks-based code size analysis tool

Hi, I would like to propose a new !annotation metadata kind that can be attached to arbitrary instructions to drive generating remarks that provide additional insight into transformations applied to a program. To motivate this, consider these specific questions we would like to get answered: * How many stores added for automatic variable initialization remain after optimizations? Where are

Cool! I really like the idea. I left a comment about metadata preservation below. Once this is available we will certainly employ it to understand OpenMP programs better. We could also think about a user facing version of this while we are at it ;) ~ Johannes On 11/4/20 3:57 PM, Florian Hahn via llvm-dev wrote: > Hi, > > > I would like to propose a new !annotation metadata kind

Dear llvm-dev community, I am trying to analyze the optimization remarks generated through clang using -fsave-optimization-remark with -O3. --- !Analysis Pass: loop-vectorize Name: CFGNotUnderstood DebugLoc: { File: c-ray-mt.c, Line: 177, Column: 2 } Function: main Args: - String: 'loop not vectorized: ' - String: loop control flow is not understood by vectorizer I tried to look for

14. 10. 2020 v 11:51, Victor Sudakov <vas at sibptus.ru>: > >> I had to do it with 10-20 live users and it went fine, nobody noticed >> anything (I made the change in the night hours just to be sure >> though). I had to go down and fix some erratic folder names that >> resulted from the botched separator handling on some macOS Mail >> accounts. > >

Does anybody have any insight into what could be causing this strange problem? I have Samba 3.02 (sorry, can't upgrade right now). My Linux box is running Mandrake 10 Official. I have Samba configured as a PDC. I can successfully add machines to my Domain -- and log on as ANY user from ANY machine. However on a SINGLE MACHINE, when I log on as USER_1 and access that user's roaming

Dear all, I have a list of precipitation record and a list of time I would like to sum them up per hour, or per day. Does such a function exist ? example: time<-c("2000-10-03 14:00:00","2000-10-03 14:10:00","2000-10-03 14:20:00","2000-10-03 15:30:00","2000-10-03 16:40:00","2000-10-03 16:50:00","2000-10-03

Hi all: I have a question on how to go about creating a correlation matrix. I have a huge amount of data....21 variables for 3471 times. I want to see how each of the variables correlate to each other. Any help would be appreciated, including which package and which functions I should use to do this. Thanks, Jessica Higgs Masters Student Department of Meteorology Penn State University

Dear all, I am working with a csv file. Some data of the file are not valid and they are marked with a star '*'. For example : *789. I have attached with this email a example file (test.txt) that looks like the data I have to work with. I see 2 possibilities ..thast I cannot manage anyway in R: 1-first & easiest solution: Read the data with read.csv in R, and define as na strings

Hi, I'm new to llvm and am trying to run benchmarks from the test-suite using lnt to check loop-vectorization for various benchmarks. Test are compiling and executing fine, but I am not getting optimization remarks while using flags like -Rpass-missed=loop-vectorize and -Rpass-analysis=loop-vectorize I've tried running it like this: lnt runtest test-suite --sandbox SANDBOX --cc

Hi Fellow R Users, I recently upgraded to Mac OS X 10.5 (Snow Leopard) and had some issues downloading and running R 2.10.1. I fixed the tcl/tk problem I was originally having but it was replaced with another. I run R out of the shell (terminal) and when I ask it to download.packages() it gives me the following message: *Loading Tcl/Tk interface ... done* *Error in file.info(x) : argument

Adam, thanks for all the suggestions! One nice aspect of the `-Rpass` family of options is that I can filter based on what I want. If I only want to see which inlines I missed, I could use `clang -Rpass-missed="inline"`, for example. On the other hand, optimization remark YAML always include remarks from all passes (as far as I can tell), which increases the amount of time it takes

[This email is either empty or too large to be displayed at this time]

Hi Violeta, I compiled with clang -Oz and clang -Oz -mno-red-zone for comparisons against Oz and clang -O0 and clang -O0 -mno-red-zone for comparisons against a default clang. I unfortunately don’t have the clang commit I worked with on my home laptop, and don’t have access to the computer I was using at Apple, so I can’t help you there. Jessica > On Sep 28, 2016, at 11:22 AM, Violeta

Dear R user, After I imported data (csv format) in R, I called it out. But it is in non-numeric format. Then using "as.numeric" function. However, the output is really awful !!!!! > PE[1,90:99] V90 V91 V92 V93 V94 V95 V96 V97 V98 V99 1 16.8467742 17.5853166 19.7400328 21.7277241

Hi, There was a recent discussion on generating non-temporal stores automatically[1]. This is a hard problem to get right. What seems like a much easier but still useful problem to solve is to provide diagnostics to the user where NT stores *may* help. Then the user can add the corresponding builtins and see if they are beneficial. My hope is that with the work to add profile-driven

> On Wed, Jun 28, 2017 at 8:13 AM, Hal Finkel <hfinkel at anl.gov> wrote: > > I don't object to adding some kind of filtering option, but in general it won't help. An important goal here is to provide analysis (and other) tools to users that present this information at a higher level. The users won't, and shouldn't, know exactly what kinds of messages the tools use.

Hi Jessica, This time I'm replying all.... On Fri, May 4, 2018 at 12:45 PM, Jessica Paquette <jpaquette at apple.com> wrote: > I would personally like a feature like that in FileCheck because it would > make it a lot easier to write MachineOutliner tests, and would make the > tests significantly smaller and easier to understand. > How do MachineOutliner tests accomplish

Hi All I have a column/variable called time difference. It has a whole list of numbers from 0 through to the hundreds eg 236. I want to assign a corresponding "name" to each variable from a predefined list: Month or less, 1 -2 months, 2-3 months etc So the result would look something like: Time Difference Month 1 Month or

Good afternoon! After applying the latest bash RPM listed at http://lists.centos.org/pipermail/centos-announce/2014-September/020594.html : The fixed RPM (bash-3.2-33.el5_10.4.x86_64.rpm) DOES work just fine on CentOS 5.10. However, it DOES NOT work on CentOS 5.4. That is, bash runs fine, but IS STILL VULNERABLE TO SHELLSHOCK! Scary screenie at: http://i.imgur.com/yR7sBjV.png It looks like

Hello! (+cc Adam Nemet, since he presented on optimization remarks at LLVM Dev Mtg 2016) I have a large C++ program, which I am compiling using a sampling profile generated via perf and AutoFDO. I'd like to use this profile in order to show the hotness of each code path that is displayed in the new optimization remarks viewer tool ( https://www.youtube.com/watch?v=qq0q1hfzidg). It seems,