search for: bolts

Hi, I was interested in trying out LLVM BOLT and generated profile data using Linux perf using the following: perf record -e cycles:u -o perf.data <command> This is without the use of LBR so I understand the performance improvements may not be much but this was more for becoming familiar with BOLT's commands. I then run: perf2bolt -nl -p perf.data -o perf.fdata <binary> and I

As planned the new release is available at: ftp://libvirt.org/libvirt/ It is a rather large release with near 400 commits included. From an user point of view the main improvement is likely to be when using migration as various work has been done to extend the protocol and to avoid having the migration command stop other concurrent operations (like virsh list). See below for a number of

As planned and after most of the clang detected problems got fixed (thanks Eric !) the new release is available at: ftp://libvirt.org/libvirt/ It's a mixed release, it includes a number of improvements as well as many bug fixes and a few new features: Features: - support various persistent domain updates (KAMEZAWA Hiroyuki) - improvements on memory APIs (Taku Izumi) - Add

Hi all, I’ve been working on a Java-esque object-oriented language Bolt that targets LLVM IR. I’m agonisingly close to getting a virtual table working, and was hoping one of you could point out the gap in my understanding. I’ve linked the C++ code snippets relevant to the vtable below. Example IR generated (partially displayed below):

Hello Maksim, On Fri, Oct 18, 2019 at 10:57 AM Maksim Panchenko <maks at fb.com> wrote: > Cool. The new numbers look good. If you run BOLT with jemalloc library > > preloaded, you will likely get a runtime closer to 1 minute. We’ve noticed > that > > compared to the default malloc, it improves the multithreaded > > performance and brings down memory usage

Hello Maksim, On Wed, Oct 16, 2019 at 3:52 PM Maksim Panchenko <maks at fb.com> wrote: > Hi Sri, > > > > I want to clarify one thing before sending a detailed reply: did you > evaluate > > BOLT on Clang built with basic block sections? > In the makefile you reference, > > there are two versions: a “vanilla” and a default built with function > sections.

We are going to be at the llvm-dev meeting the next two days. We will get back to you after that. Sri On Mon, Oct 21, 2019 at 10:07 PM Maksim Panchenko <maks at fb.com> wrote: > Hi Sri, > > > > Thank you for replying to our feedback. 7 out 12 high-level concerns have > been > > answered; 2 of them are fully addressed. The rest are being tracked at the > >

Greetings, We, at Google, recently evaluated Facebook’s BOLT, a Post Link Optimizer framework, on large google benchmarks and noticed that it improves key performance metrics of these benchmarks by 2% to 6%, which is pretty impressive as this is over and above a baseline binaryalready heavily optimized with ThinLTO + PGO. Furthermore, BOLT is also able to improve the performance of binaries

Hello, I wanted to consolidate all the discussions and our final thoughts on the concerns raised. I have attached a document consolidating it. BOLT’s performance gains inspired this work and we believe BOLT is a great piece of engineering. However, there are build environments where scalability is critical and memory limits per process are tight : * Debug Fission,

Following Dan advice, I decided to not wait for more patches and push the current git head as the release. Let's plan to have another release by this month end with the QEmu debugging and hacking APIs. The release is available as usual at ftp://libvirt.org/libvirt Quite a lot of bug fixes during the two months since 0.8.1, and a few new feature. It also tagged more commits as being

Is there large value from deferring the block ordering to link time? That is, does the block layout algorithm need to consider global layout issues when deciding which blocks to put together and which to relegate to the far-away part of the code? Or, could the propellor-optimized compile step instead split each function into only 2 pieces -- one containing an "optimally-ordered" set of

Thanks a ton for the reply Matthias. I tried your suggestion but now I got one more message saying: *virsh # connect esx://172.16.73.14/?no_verify=1* *Enter username for 172.16.73.14 [root]: root* *Enter root's password for 172.16.73.14: * *error: Failed to connect to the hypervisor* *error: internal error: curl_easy_perform() returned an error: Unsupported protocol (1) : Protocol https not

Some more information about the relaxation pass whose effectiveness and convergence guarantees were listed as a concern: TLDR; Our relaxation pass is similar to what LLVM’s MCAssembler does but with a caveat for efficiency. Our experimental results show it is efficient and convergence is guaranteed. Our relaxation pass is very similar to what MCAssembler does as it needs to solve the same

Hi, &nbsp;I was finally able to connect to the xenapi hypervisor.&nbsp;&nbsp;However, once I connect to virsh and give a 'quit' command I get&nbsp; "* glibc detected *** virsh: double free or corruption (fasttop):". Please find below the actual memory dump. &nbsp; Note: I have disabled the SSL certificate verification in xenapi_driver.c by&nbsp;setting

Having newly compiled a Libvirt on a Windows Vista machine (many thanks to Matthias Bolte) I have attempted to connect to a remote Ubuntu machine using virsh.exe. However, the connection fails as follows: $ virsh -c qemu+tcp:///system error: unable to connect to libvirtd at 'localhost': errno=10061 error: failed to connect to the hypervisor 1. I have modified

On Wed, Sep 25, 2019 at 5:02 PM Eli Friedman via llvm-dev < llvm-dev at lists.llvm.org> wrote: > My biggest question about this architecture is about when propeller runs > basic block reordering within a function. It seems like a lot of the > complexity comes from using the proposed -fbasicblock-sections to generated > mangled ELF, and then re-parsing the mangled ELF as a

please see my pull request on github (patch attached as well to this mail): > https://github.com/openssh/openssh-portable/pull/21 thank you! -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-sshd-make-c-option-usable.patch Type: text/x-patch Size: 569 bytes Desc: not available URL:

Thanks a ton for the reply Matthias. I am little confused here. How to add volume on my esx server using virsh? I did SCP and copied my vmdk image to [datastore1] testNode/cluster.vmdk and also modified my xml file as you suggested. Now when I tried starting the node it gave me an error message saying: "error: Failed to start testNode error: internal error: Could not start domain:

Hi, I was trying to connect to ESX server using libvirt's virsh console but I am getting the following error: *virsh# connect esx://172.16.73.14* *Enter username for 172.16.73.14 [root]: root* *Enter root's password for 172.16.73.14:* *error: Failed to connect to the hypervisor* *error: internal error curl_easy_perform() returned an error: Peer certificate cannot be authenticated with

Just curious but how long has ESX support been in libvirt? Also I noticed this error when accessing one of my windows vms: error: internal error Expecting VMX entry 'scsi0.virtualDev' to be 'buslogic' or 'lsilogic' but found 'lsisas1068' Not sure whether this is an enhancement, defect, or not supported at this time. Thanks. -Jon