similar to: [LLVMdev] program specialization vs LTO in LLVM

> LTO and partial evaluation are completely orthogonal > from each other. One is an optimization technique > and the other is a time that optimization can > occur. Please explain a bit more about what you are > trying to accomplish. A contrived example I had in mind is a program that computes a power function. In pseudocode: int powerFunc( int x, int n ) { if n is 0

Hi, I have a question regarding ExecutionEngine on multicore machines. Say I have n different Function objects that I want to execute simultaneously on a multicore machine. So, one would want to create n different instances of ExecutionEngine and run each function. But, if I understand correctly, there can only be one instance of ExecutionEngine, so there would have to be a locking mechanism

Hi, When I call the *switch* function first time, it works. but when I call it at the second time, it does nothing. The version I use is R version 2.9.0 Under development (unstable) (2009-02-21 r47969) here is the output: > organism="human" > species <- switch(organism, human <- "Hs", fly <- "Dm", mouse <-

Is there anything I can do about it? BTW: gcc is loosing the AVR backend, so I would assume, there will be a greater interest to this in llvm compared to the past. Thanks, Wilhelm Am 03.04.20 um 15:09 schrieb Wilhelm Meier via llvm-dev: > Should I create an issue in bugzilla for this? Just to be reminded ... > > Am 31.03.20 um 09:34 schrieb Wilhelm Meier via llvm-dev: >> Hi

Hello, my data is sorted by start.ens (see below). And now I would like to extract all rows (so called* defined row*s) with type==Expression - subset (df, type==Expression) - and the aforegoing type==DNase HS (which is not necessarly row n-1 - assumung that the defined row is n). I dont know how to add this to my subset command. Is that possible? Thanks Hermann > df start.ens fc.trans

Hi Dylan, looks ok now. One thing: the ISR is now: __vector_21: ; @__vector_21 __vector_21$local: sei push r0 push r1 in r0, 63 push r0 clr r0 push r24 lds r24, v1 sts v2, r24 pop r24 pop r0 out 63, r0 pop r1 pop r0 reti There are unneccessary push/pops of r1 and r0 too, since the clr is useless ... GCC had the same

hello I want to have subset of my dataset that is equal to 0. But, I got this error message. > hs2=hs[hs$hey60==0 ,] Error in hs[hs$hey60 == 0, ] : (subscript) logical subscript too long regards -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.

On Jul 12, 2013, at 3:49 PM, Shuxin Yang <shuxin.llvm at gmail.com> wrote: > 3.2 Compile partitions independently > -------------------------------------- > > There are two camps: one camp advocate compiling partitions via multi-process, > the other one favor multi-thread. > > Inside Apple compiler teams, I'm the only one belong to the 1st comp. I think >

Hey Wilhelm, That's a bug, the "interrupt" attribute is not being recognized by the backend. I have fixed it in https://github.com/llvm/llvm-project/commit/339b34266c1b54a9b5ff2f83cfb1da9cd8c9d90a Pull the latest LLVM and it should be fixed. On Tue, Mar 31, 2020 at 8:00 AM Wilhelm Meier <wilhelm.meier at hs-kl.de> wrote: > Hi Dylan, > > I used the following

Hello, I'm using dovecot 2.2.9 and a director/backend setup. On the director I've the LMTP in proxy mode, mapping the users to one of the backends. The backends to quota check and return the OverQuota message already at RCPT TO time. Here is what I typed, connected to the director Connection to director1 2525 port [tcp/*] succeeded! 220 director1.rz.hs-example.de Dovecot

Dear R-core devs, I hope this email finds you well. Please see the proposed patch to R-devel below: Scenario: When loading a package using `library`, a package may not be found if the cases are not matching: ``` > library(ORG.Hs.eg.db) Error in library(ORG.Hs.eg.db) : there is no package called 'ORG.Hs.eg.db' ``` Suggested Patch: Returns a message matching what

dear all, I have a very simple question but I could not figure out. I need to make plots with grid in the background. something like I old retrive like this a=runif(100)*10 b=runif(100)*10 plot(a,b, pch=20, xlim=c(0, round(max(a))), ylim=c(0, round(max(b)))) vs=seq(0, max(a), 0.5) for(i in 1:length(vs)){ abline(v=vs[i], col="lightgrey") } hs=seq(0, max(b), 0.5) for(i in

I have 4 dell 2970 server , three server harddisk is 146Gx6 ,one hard disk is 72Gx6: each server mount info is /dev/sda4 on /exp1 type xfs (rw) /dev/sdb1 on /exp2 type xfs (rw) /dev/sdc1 on /exp3 type xfs (rw) /dev/sdd1 on /exp4 type xfs (rw) /dev/sde1 on /exp5 type xfs (rw) /dev/sdf1 on /exp6 type xfs (rw) I create a gluster volume have 4 stripe gluster volume create test-volume3 stripe 4

On May 27, 2010, at 6:08 AM, Nicolas Capens wrote: > Hi all, > > I'm attempting to use LLVM for run-time code specialization, but I'm facing a performance hurdle. I'm currently performing the specialization during the AST to LLVM IR translation, but unfortunately this leads to relatively slow recompiles as LLVM has to perform all the heavy (optimization) passes over and

Hi Chris, Thanks for pointing me to that presentation! It helped me come up with a strategy that I believe might work: 1) Use CloneFunction() to make a copy of the original unspecialized (but optimized) function. 2) Specialize it using a custom function pass which identifies the specialization parameters and substitutes them with given run-time constants. 3) Run the function through a

I don't think there is any infrastructure for this kind of specialization. The closest thing I can think of is the profile-guided optimization stuff that Andreas Neufstifter has worked on. Because compilation and optimization with LLVM is expensive, our approach with Unladen Swallow has been to try to wait longer to generate specialized code, so we don't have to recompile. If memory

On Nov 11, 2011, at 4:14 PM, Don Quixote de la Mancha wrote: > cmp.w *looks* like a 16-bit comparison with an immediate constant, but > in reality the constant is twelve bits. The ARM and Thumb instruction > sets have quite severe restrictions on the allowed ranges of immediate > values because the richness of the ARM and Thumb instruction set makes > it hard to find enough bits in

> On Apr 1, 2016, at 3:19 AM, Renato Golin via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > On 1 April 2016 at 08:56, James Molloy via llvm-dev > <llvm-dev at lists.llvm.org> wrote: >> However, LLVM knows none of this. Only if I force-inline >> std::vector::__append and friends does LLVM actually see the operator >> new(256) call - without that LLVM

A third approach is to decouple the backend compilation and parallelism strategy from the partitioning. The partitioning can spits out partition BC files and some action records in some standard format. All of this can be fed into some driver tools that converts the compilation action file into make/build file of the underlying build system of your choice: 1) it can simply a compiler driver that

I'm a little confused. I'd like to build an RPM for Asterisk 1.4. Is it better to modify and use the spec file under redhat/asterisk.spec and run a 'make rpm', OR is it better to build a custom spec file from scratch and use 'rpmbuid -ba' <specfile>? How do people normally do it? The problem I see with a custom spec file is that since the source is all contained