search for: flops

Hi all, Since proc.time return three different kind of times (user, system and elapsed) , I am wondering which one is right for calculating flops. In New S Language (Becker et. al. ) , it seems to be the user because " the user time measures the processor time used in S and the system time measures the operating system in response to S's request". But in R Help , system time sounds better as " The first two entries(refer...

After looking through the archives I haven''t been able to assess the reliability of a backup procedure which employs zfs send and recv. Currently I''m attempting to create a script that will allow me to write a zfs stream to a tape via tar like below. # zfs send -R pool at something | tar -c > /dev/tape I''m primarily concerned with in the possibility

Jean-Marc, I'm an embedded systems designer and am in the process of selecting a hardware platform to run a Linux-Speex VoIP application. Being concerned with size and power consumption I must choose a processor with a minimal amount of computational power. To that end I've narrowed my search to number of different products that reference the popular "SiSoft Sandra(TM)"

On Tue, Feb 14, 2017 at 11:21:34AM -0500, James Hartig wrote: > Those 2 boxes are in the same subnet and have addresses of 10.240.0.4 and > 10.240.0.5, respectively, on their eth0 interface. Port 655 on tcp and udp > is open to the world. The tinc_test_2 box has a ConnectTo of tinc_test_1. > When tinc_test_2 is started, it prints out: > UDP address of tinc_test_1 set to

...r people the opportunity to perform their own performance measurements. — Performance Gains — SingleSource/Benchmarks/Misc/matmul_f64_4x4 -53.68% MultiSource/Benchmarks/Olden/power/power -18.55% MultiSource/Benchmarks/TSVC/LoopRerolling-flt/LoopRerolling-flt -14.71% SingleSource/Benchmarks/Misc/flops-6 -11.02% SingleSource/Benchmarks/Misc/flops-5 -10.03% MultiSource/Benchmarks/TSVC/LinearDependence-flt/LinearDependence-flt -8.37% External/Nurbs/nurbs -7.98% SingleSource/Benchmarks/Misc/pi -7.29% External/SPEC/CINT2000/252_eon/252_eon -5.78% External/SPEC/CFP2006/444_namd/444_namd -4.52% Ext...

...st as bad as > the predictable number generator issue. > Well, heratbleed and shellshock were pretty much global: all systems (not only Linuxes, not to say particular Linux distributions - my FreeBSD boxes were affected too) using openssl or bash were affected... Same bad, yet these were not flops of particular distribution, so whichever system you decided to stick with , you had these. Not certain about you, but this kind of makes difference for me. When I say I'm happy about [me choosing way back] RedHat heartbleed, no heartbleed, no difference. Valeri +++++++++++++++++++++++++++++++...

On Tue, Feb 14, 2017 at 1:22 PM, Etienne Dechamps <etienne at edechamps.fr> wrote: > > Can you specify which version of tinc you're using? There are vast differences in the way LocalDiscovery works between 1.0 and 1.1. The former uses broadcast, the latter unicast to explicitly advertised local addresses. I'm using tinc 1.1pre14. I noticed there's an option,

...me overhead: > >I believe so to, but please verify with -debug-pass=Structure I have verified. It indeed does not involve Polly Dependence analysis. "Polly Dependence Pass" for flop is still high for some benchmarks as we have discussed before. >> SingleSource/Benchmarks/Misc/flops 28.57% >> MultiSource/Benchmarks/MiBench/security-sha/security-sha 22.22% >> MultiSource/Benchmarks/VersaBench/ecbdes/ecbdes 21.05% >> When I look into the compile-time for the flop benchmark using "-ftime-report", I find the extra compile-time overhead mainly comes from...

On Sun, January 11, 2015 5:16 pm, Keith Keller wrote: > On 2015-01-11, Valeri Galtsev <galtsev at kicp.uchicago.edu> wrote: >> >> Indeed. Or another system altogether (sihg). I'm just extending your >> thought half a step farther ;-) > > Or going even farther, if you like CentOS but not systemd, do the work > to get CentOS working without it. Unhappy Debian

Can you specify which version of tinc you're using? There are vast differences in the way LocalDiscovery works between 1.0 and 1.1. The former uses broadcast, the latter unicast to explicitly advertised local addresses. You say that tinc_test_1's eth0 interface is configured with 10.240.0.4, and tinc_test_2's eth0 interface is configured with 10.240.0.5. How are the public addresses

...44. As shown in the results, 13 benchmarks have >5% compile-time performance improvements by simply removing the first "createInstructionCombiningPass". The top 5 benchmarks are listed as follows: SingleSource/Regression/C++/2003-09-29-NonPODsByValue-38.46% SingleSource/Benchmarks/Misc/flops-19.30% SingleSource/Benchmarks/Misc/himenobmtxpa-12.94% MultiSource/Benchmarks/VersaBench/ecbdes/ecbdes-12.68% MultiSource/Benchmarks/ASCI_Purple/SMG2000/smg2000-10.68% Unfortunately, there are also two serious execution-time performance regressions: SingleSource/Benchmarks/Adobe-C++/simple_types_c...

We are testing tinc inside Google Compute within a single region and an external region. Two boxes are created as follows: /etc/tinc/test/tinc_test_1 Subnet = 10.240.0.0/16 Subnet = 10.240.0.4/32 Address = 104.154.59.151 /etc/tinc/test/tinc_test_2 Subnet = 10.240.0.0/16 Subnet = 10.240.0.5/32 Address = 104.197.132.141 /etc/tinc/test/tinc.conf Name = $HOST AddressFamily = ipv4 Interface = tun0

On 14 February 2017 at 18:59, James Hartig <james at levenlabs.com> wrote: > When you say "and to the local network" what IP does it try to send to > on the local network? The subnet address? No. The Subnet option deals with routing *inside* the VPN, not the underlying "real" network. In tinc 1.1, the address that local discovery probes are sent to is the local

Is there a way to boot PXELINUX 100% over HTTP instead of TFTP? I can pull over pxelinux.0 via HTTP (DHCP config) but all the config files are pulled over via TFTP. DHCPD config file example: filename "pxelinux.0" ; # boots via TFTP filename "http://10.1.1.10/pxelinux.0" # boots via HTTP My goal is to alternate booting between Linux and Memtest for some hardware stress

OK, seem to have answered my own question... I think. We'll see how it works out. Basically I should've been asking for some help in the dosemu forums because that's where I needed the help. I think I get it now.... mkdiskimage whatever size you need Set up dosemu to boot the precise DOS flavor you want from mounted folder also mount the image as secondary HDD fdisk, format, sys, etc

On Tue, Feb 14, 2017 at 1:46 PM, Guus Sliepen <guus at tinc-vpn.org> wrote: > On Tue, Feb 14, 2017 at 11:21:34AM -0500, James Hartig wrote: > >> Those 2 boxes are in the same subnet and have addresses of 10.240.0.4 and >> 10.240.0.5, respectively, on their eth0 interface. Port 655 on tcp and udp >> is open to the world. The tinc_test_2 box has a ConnectTo of

....8390 0.0009 Performance Improvements - Execution Time Δ Previous Current σ SingleSource/Benchmarks/Misc/matmul_f64_4x4 -53.67% 1.4064 0.6516 0.0007 External/Nurbs/nurbs -19.47% 2.5389 2.0445 0.0029 MultiSource/Benchmarks/Olden/power/power -18.49% 1.2572 1.0248 0.0004 SingleSource/Benchmarks/Misc/flops-4 -15.93% 0.7767 0.6530 0.0348 MultiSource/Benchmarks/TSVC/LoopRerolling-flt/LoopRerolling-flt -14.72% 2.3925 2.0404 0.0013 SingleSource/Benchmarks/Misc/flops-6 -11.05% 1.1427 1.0164 0.0009 SingleSource/Benchmarks/Misc/flops-5 -10.43% 1.2771 1.1439 0.0015 MultiSource/Benchmarks/TSVC/LinearDependenc...

...size and compile time impact is. -Chris > > — Performance Gains — > SingleSource/Benchmarks/Misc/matmul_f64_4x4 -53.68% > MultiSource/Benchmarks/Olden/power/power -18.55% > MultiSource/Benchmarks/TSVC/LoopRerolling-flt/LoopRerolling-flt -14.71% > SingleSource/Benchmarks/Misc/flops-6 -11.02% > SingleSource/Benchmarks/Misc/flops-5 -10.03% > MultiSource/Benchmarks/TSVC/LinearDependence-flt/LinearDependence-flt -8.37% > External/Nurbs/nurbs -7.98% > SingleSource/Benchmarks/Misc/pi -7.29% > External/SPEC/CINT2000/252_eon/252_eon -5.78% > External/SPEC/CFP200...

...some domains and we do want to vectorize them. I skimmed through the IR and the assembly and I did not see anything too bad. The next step would be to do a binary search on the places where the vectorizer fires to locate the bad pattern. On AVX we have another regression that I did not mention: Flops-7. When we vectorize we cause more spills because we do a poor job scheduling non-destructive source instructions (related to PR10928). Hopefully Andy’s scheduler will fix this regression once it is enabled. I did not measure code size, but I did measure compile time. There are 4-5 workloads (n...

...zer=none -mllvm -polly-code-generator=none -mllvm -polly-codegen-scev > Fortunately, with the option "-polly-codegen-scev", only three benchmark shows >20% extra compile-time overhead: I believe so to, but please verify with -debug-pass=Structure > SingleSource/Benchmarks/Misc/flops 28.57% > MultiSource/Benchmarks/MiBench/security-sha/security-sha 22.22% > MultiSource/Benchmarks/VersaBench/ecbdes/ecbdes 21.05% > When I look into the compile-time for the flop benchmark using "-ftime-report", I find the extra compile-time overhead mainly comes from the "C...