similar to: LLVM mtriple for aarch64-win32-msvc ?

Thanks Martin, I'm generating the code using LLVM (writing llvm::Triple myself and llvm::TargetRegistry::lookupTarget is working), and that's how my bitcode is generated then using LLC to cross-compile that. So using armv7-win32-msvc is getting me a bit closer, but what CPU, raspberry pi 3 is running a Cortext-A53, but when I specify that in -mcpu argument I get this error: > llc.exe

Renato et al. In the meantime, is there an out of tree branch I mean, other than LLVM trunk so I can test how much I can SVE vectorize my code with it? Arm seemed to gave taken down the GitHub branch for sometime. On Wed, Sep 11, 2019 at 20:41 Renato Golin <rengolin at gmail.com> wrote: > On Wed, 11 Sep 2019 at 06:13, Itaru Kitayama via llvm-dev > <llvm-dev at lists.llvm.org>

Motivation At the moment, when targeting armv7a, clang defaults to generate code as if -mcpu=cortex-a8 was specified. When targeting armv8a, it defaults to generate code as if -mcpu=cortex-a53 was specified. This leads to surprising code generation, by the compiler optimizing for a specific micro-architecture, whereas the intent from the user was probably to generate code that is

Thanks for everyone giving their feedback! I saw pretty unanimous support for making -mcpu=generic the default and making -mcpu=generic schedule for an in-order CPU (Cortex-A8 in this case). I'll be making those changes shortly. I think the comments also make clear that it's less obvious whether we'd want -mcpu=native to become a default. It's probably good for some use cases, but

Hi all, Firstly thanks so much to Apple for open sourcing this and Tim for going through the effort of committing it! Along with Bradley I've been looking at this today from a perspective of working out how best to get this merge completed. The one sentence summary is "I think we should use ARM64 as a base". My view on the backends is that the ARM64 backend is more performant but

I'm running into some pretty bad performance in llc.exe when compiling some large neural networks into code that contains some very large llvm::ConstantDataArrays, some are { size=102,760,448 }. There's a small about of actual code for processing the network, but the assembly is mostly global data. I'm finding that llc.exe memory spikes up around 30 gigabytes and the job takes 20-30

> 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: >>

I was working with the rmetrics portfolioBacktesting function and dug into the code to try to find why my formula with 113 items, i.e. A1 thru A113, was being truncated and I only get 85 items, not 113. Is it due to a string length limitation in R or is it a bug in the strsplit or gsub functions, or in my string? I'd very much appreciate any suggestions ============Input script:

Hi Gurus, I've been playing with CentOS 7's AltArch i386 builds with some good results on one machine, but can't get it to boot properly on another with a newer Bay Trail CPU. Previously CentOS 6 i386 worked on both, and that set a legacy I'd like to recreate... I *think* the problem is that the 7 kernel is non-PAE, and that has some peculiar knock on effect that prevents some

> 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

Hello, We are succesfully using Samba with a remote WINS server which is located on another site. We have a second site which has successfully been using the same WINS server and communicating with our Samba server. Recently the PCs on the second site have been unable to connect to our Samba server. The other site uses a different DHCP server to our site but it still gives out the same

https://bugzilla.mindrot.org/show_bug.cgi?id=3704 Bug ID: 3704 Summary: Implement an interface to capture port number of random remote port forwarding -R 0:localhost:22 Product: Portable OpenSSH Version: 9.7p1 Hardware: All OS: All Status: NEW Severity: enhancement Priority:

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

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

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

tl;dr in low data situations we don’t look at past information, and that increases the false positive regression rate. We should look at the possibly incorrect recent past runs to fix that. Motivation: LNT’s current regression detection system has false positive rate that is too high to make it useful. With test suites as large as the llvm “test-suite” a single report will show hundreds of

Hi folks, As Tim pointed out, we recently had the opportunity to collect 64-bit benchmark performance data for GCC 4.9, AArch64 and ARM64 compilers on a real hardware. It is a cortex-a53 device. Due to proprietary reasons we cannot share the full hardware configuration. The preliminary results were shared at the hackers lab at EuroLLVM yesterday. For those who could not make it, below is

Thanks for the example, that is very useful in working out the overall scope of the problem, which is now wider than I thought it was. I've put some comments inline. On 15 March 2018 at 19:12, Friedman, Eli <efriedma at codeaurora.org> wrote: > On 3/15/2018 9:43 AM, Peter Smith via llvm-dev wrote: >> >> Hello everyone, this is most likely Arm specific, but could affect

On 07/12/2016 03:36 AM, Yamaban wrote: > On Tue, 12 Jul 2016 10:15, Walter H. wrote: >> >> Hello, >> >> can CentOS 6.8 be run on a mini-PC like this? >> https://www.zotac.com/product/mini_pcs/zbox-ci323-nano >> >> would like to configure this as a Firewall, and this should be instead of >> my router (integratet firewall, NAT) and wlan-ap >>