similar to: Changing Alignment of global variables in LLVM

If I know for sure I am accessing 32 byte chunks at a time, how can I go about changing the alignment of @u? Should I use DataLayout's reset method? I couldn't find a method to change alignment of one global variable. Thanks On Tue, Oct 3, 2017 at 6:34 PM, Matthias Braun <mbraun at apple.com> wrote: > The effective alignment is part of the load and store operations. Updating

I isolated the LLVM IR and the X86 instructions emitted for the function and are attached herewith and it is clearly emitting vector instructions. I am having a hard time figuring out where the vector instructions are formulated. For sure SLP and Loop vectorizer is not doing anything. On Mon, Aug 21, 2017 at 11:56 AM, Craig Topper <craig.topper at gmail.com> wrote: > The X86 backend

If count > MAX_UINT-4 your loop loops indefinitely with an increment of 4, I think. On Thu, Oct 13, 2016 at 4:42 PM, Charith Mendis via llvm-dev < llvm-dev at lists.llvm.org> wrote: > So, I tried unrolling the following simple loop. > > int unroll(unsigned * a, unsigned * b, unsigned *c, unsigned count){ > > for(unsigned i=0; i<count; i++){ > > a[i] =

Thanks for the explanation. But I am a little confused with the following fact. Can't LLVM keep vectorizable_elements as a symbolic value and convert the loop to say; for(unsigned i = 0; i < vectorizable_elements ; i += 2){ //main loop } for(unsigned i=0 ; i < vectorizable_elements % 2; i++){ //fix up } Why does it have to reason about the range of vectorizable_elements? Even

Hi all, Recently I compiled the attached .c file using Clang with "-mavx2 -mfma -m32 -O3" optimization flags. First I used -emit-llvm and inspected the LLVM IR and there are no vector instructions. Then I got the assembly output of the file in it I can clearly see vector instructions in it. Neither the SLPVectorizer or the LoopVectorizer is however doing any vectorization (also

How do you generate a SCEVAddRecExpr from a SCEV? It tried dyn_casting and it seems like that the SCEV returned by getSCEV is not a SCEVAddRecExpr. Thanks On Fri, Sep 30, 2016 at 4:16 PM, Friedman, Eli <efriedma at codeaurora.org> wrote: > On 9/30/2016 12:16 PM, Charith Mendis via llvm-dev wrote: > >> >> Hi all, >> >> What is the best way to get the symbolic

Hi all, What is the best way to get the symbolic expression for an address calculation in llvm specially when memory addresses are calculated within a loop. Use case: I want to know what loop induction variables are used for a particular address calculation and in what symbolic context. Thereby, I want to identify which stores and loads will be contiguous in memory if I unroll each of the

Hi all, Attached herewith is a fairly simple LLVM file (shuffle.ll) with lots of vector shuffles. When I use llc with -O3 -mcpu=core-avx2 the first shuffle sequence containing types of 128 wide gets reduced a single shuffle, where as the second shuffle sequence containing types of 256 wide doesn't get reduced to a single shuffle instruction in the resulting X86 code (Shuffle.s attached).

Hi all, Attached herewith is a simple vectorized function with loops performing a simple shuffle. I want all loops (inner and outer) to be unrolled by 2 and as such used -unroll-count=2 The inner loops(with k as the induction variable and having constant trip counts) unroll fully, but the outer loop with (j) fails to unroll. The llvm code is also attached with inner loops fully unrolled. To

I updated to the latest revision and now llvm does not build and quits cmake with CMake Error at cmake/modules/LLVMProcessSources.cmake:83 (message): Found unknown source file ../llvm-revec/lib/CodeGen/MachineFunctionAnalysis.cpp Please update ../llvm-revec/lib/CodeGen/CMakeLists.txt Thanks On Sat, Sep 3, 2016 at 2:09 AM, Craig Topper <craig.topper at gmail.com> wrote: >

Hi all, The attached LLVM assembly file fails to generate x86 code when compiled using llc. compilation command - ../llvm-build/bin/llc -filetype=asm -march=x86-64 -mcpu=core-avx2 ex4.ll The error message is, LLVM ERROR: Cannot select: t95: v8f32 = X86ISD::SUBV_BROADCAST t17 t17: v4f32,ch = load<LD16[%scevgep](tbaa=<0x4dbcd98>)> t0, t16, undef:i64 t16: i64 = add t2,

Good day to all! I'm doing git file replication task and need to sync source destination, using below sync command. rsync -atnvv --existing --exclude '.git' --progress source/ destination/; Here I, 1) need only sync directories that *exist on destination* ,no new folders should copy from source - [used --existing did the job but it not copying new files also.] 2) need to copy

Hi Rowland, I made the change you suggested to auto refresh kerberos. It didn't seem to fix the issue unfortunately, even after a machine restart. Following your line of reasoning that it is a Kerberos issue, I then tried to grab a new kerberos ticket on the server in question which appears to fail though. Perhaps this gives some further insight? pi at fs1:~ $ kinit administrator at

I joined a Windows 10 Pro system to my (still experimental) domain. The windows system actually hosts DC2 as a VM, and another Windows (Server 2008 R2) at another location hosts DC1 also as a VM. The two locations are connected via a VPN, both systems run only when needed. The windows system does not directly use DC2 for DNS but instead talks to a DNS resolver that delegates the samba Domain to

> -----Ursprüngliche Nachricht----- > Von: samba [mailto:samba-bounces at lists.samba.org] Im Auftrag von > Rowland penny > Gesendet: Sonntag, 5. Juni 2016 12:49 > An: samba at lists.samba.org > Betreff: Re: [Samba] inconsistent DNS information, windows domain > member issues.. > > On 05/06/16 10:05, Jo wrote: > > I joined a Windows 10 Pro system to my (still

Forgive me if I'm missing something obvious, but it seems that a number of core instructions—I'm specifically running in to `atomicrmw`, `fence`, and `cmpxchg` at the moment—cannot be constructed from the C bindings, and are therefore also inaccessible to the OCaml bindings. There are opcodes for each of these in the llvm-c/Core.h, but there seems to be no way to construct them. Is there

I am a new user of the R spatstat package and am having problems creating a polygonal observation window with owin(). Code follows: library("maps") library ("sp")` library("spatstat") mass.map <- map("state", "massachusetts:main", fill=T) # This returns a data frame includding x and y components that form a polygon of massachusetts mainland`

Hello, Let''s say I have a model for a "house". Each house has_a "city", which in turn has_a "state" which in turn has_a "country". The objective is to retrieve all houses in a given state, say "Massachusetts". Being new to Ruby on Rails what I like is that things that should be simple usually are simple, and since we have easy access to

Hi folks, there are two Samba4 DC server. The first one is the "PDC", and after I finished to set up that, I've joined the second one. There is a Linux client, where I configured the samba, and joined it to domain as member. Now I see these: # net ads status -U administrator objectClass: top objectClass: person objectClass: organizationalPerson objectClass: user objectClass:

On 11/06/2020 12:16, Bob Wooden via samba wrote: > When I "nslookup". I get: > > root@[dchost]:~# nslookup [dchost].[domain].work > Server:??? ??? 192.168.116.50 > Address:??? 192.168.116.50#53 > > Non-authoritative answer: > Name:??? [dchost].[domain].work > Address: xx.198.245.139 > > This local active directory is the first I have ever setup with a