search for: nbench

As per [1] I am going to add [2] to the llvm test-suite. I hope nobody is currently working on that area (if anybody is working please do let me know) Rajika [1] - http://llvm.org/OpenProjects.html#llvmtest [2] - http://www.tux.org/~mayer/linux/bmark.html -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hi All, I have samba 3.6 and samba 4.2 running on 2 different freebsd machines. Both are virtual machines and have 1GB Ethernet and exact same configurations. Samba is joined to a 2K8 server(security = ADS). Both use a minimal similar smb.conf. Am using smbtorture bench.nbench from another freebsd machine. Samba 3.6 Throughput is at: 12.6165 MB/sec Samba 4.2 Throughput is at: 1.5359 MB/sec 4.2 appears almost 10X slower. What could be going wrong for me ? Is there an official bench-marking of Samba 3 and Samba 4, which I could refer to ? btw.. netbench seems to be go...

Hi, I got problem while using smbtorture with the NBENCH test. The error happens on the "Mkdir" operation. I'm using the file "client.txt" from dbench-3.04. There are "Mkdir" operations, but in the file "torture.c", this operation is not handled. Few other operations are not handled. I did compare operations f...

Why does GCC produce faster code using "-march=pentium2 -mtune=pentium4" on a Pentium 4 chip versus plain -march=pentium4? Try it... CPUTYPE=pentium2 CFLAGS+= -mtune=pentium4 COPTFLAGS+= -mtune=pentium4 -- BSD Podcasts @ http://bsdtalk.blogspot.com/

...nchmarks/Ptrdist/yacr2/yacr2 5.62% Performance Improvements - execution_time Change SingleSource/Benchmarks/Misc-C++/Large/sphereflake -4.43% External/SPEC/CINT2006/456.hmmer/456.hmmer -2.50% External/SPEC/CINT2006/464.h264ref/464.h264ref -1.60% MultiSource/Benchmarks/nbench/nbench -1.19% SingleSource/Benchmarks/Adobe-C++/functionobjects -1.07% I had a brief look at the regressions and they all look to be caused by getting bad luck with branch mispredictions: I looked into the Shootout-ary3 and yacr2 cases and in both the hot code path was the same with...

Hi, I've just tagged 3.5.1-rc2, so testing can begin. Remember to use the 3.5 branch of the test-suite. If you have any questions, let me know. -Tom

...re-list for this builder? It allows faster catch and fix other failures (if we get them). I also wandering, may be some of them just incompatible with freebsd platform, or with CPU they were tested at? Below some details: MultiSource/Benchmarks/SciMark2-C - fp difference. MultiSource/Benchmarks/nbench - timeout. SingleSource/UnitTests, Output/initp1.out-llc - empty stdout. SingleSource/Benchmarks/Misc/mandel - fp difference. After short analysis, I found out, that first 3 tests were disabled on clang-X86_64-freebsd machine. "mandel" test looks is new one here, that's why I creat...

...Now per my reading's to run the smbtorture test tool required in the dbench config.txt file. Is this still ture? I am using distribution samba-3.0.27a. Results from initial run: host=10.60.100.131 share=moab2 user=engsmoke/admin1 myname=rh-scale-144 Did not find a test named 1 Running NBENCH 1 clients started ERROR: You are using a dbench 1 load file Throughput 0 MB/sec NBENCH took 0.043605 secs Are these smb operation code still valid? Deltree "\clients\client1" NT_STATUS_OK Mkdir "\clients" NT_STATUS_OK NTCreateX "\clients\clien...

....62% > > Performance Improvements - execution_time          Change > SingleSource/Benchmarks/Misc-C++/Large/sphereflake -4.43% > External/SPEC/CINT2006/456.hmmer/456.hmmer  -2.50% > External/SPEC/CINT2006/464.h264ref/464.h264ref  -1.60% > MultiSource/Benchmarks/nbench/nbench  -1.19% > SingleSource/Benchmarks/Adobe-C++/functionobjects -1.07% > > I had a brief look at the regressions and they all look to be > caused by > getting bad luck with branch mispredictions: I looked into the > Shootout-ary3 and > yacr2 cases and...

...ance Improvements - execution_time Change >>> SingleSource/Benchmarks/Misc-C++/Large/sphereflake -4.43% >>> External/SPEC/CINT2006/456.hmmer/456.hmmer -2.50% >>> External/SPEC/CINT2006/464.h264ref/464.h264ref -1.60% >>> MultiSource/Benchmarks/nbench/nbench -1.19% >>> SingleSource/Benchmarks/Adobe-C++/functionobjects -1.07% >>> >>> I had a brief look at the regressions and they all look to be caused by >>> getting bad luck with branch mispredictions: I looked into the Shootout-ary3 and >&gt...

...orted by qemu (x86, x86-64, ARM, SPARC, PowerPC, MIPS, m68k) can be translated to LLVM IR this way (adding support for one of these architectures only requires minor changes to llvm-qemu). The end result was that llvm-qemu was running about half the speed of regular qemu on the synthetic benchmark nbench (using a hotspot-like approach: interpretation of blocks with few executions and JITing of blocks with high execution counts). However, there is still potential for improvement, one being an efficient implementation of direct block chaining (in certain cases a block can directly jump to its success...

...167.00 475.9% test-suite...T2006/445.gobmk/445.gobmk.test 19.00 88.00 363.2% test-suite.../Applications/SPASS/SPASS.test 49.00 155.00 216.3% test-suite...lications/ClamAV/clamscan.test 72.00 227.00 215.3% test-suite.../Benchmarks/nbench/nbench.test 30.00 92.00 206.7% test-suite...T2000/256.bzip2/256.bzip2.test 31.00 85.00 174.2% test-suite...chmarks/MallocBench/gs/gs.test 57.00 133.00 133.3% test-suite...000/255.vortex/255.vortex.test 95.00 200.00...

...SEC backend 'gssapi_krb5' registered GENSEC backend 'gssapi_krb5_sasl' registered GENSEC backend 'ntlmssp' registered NTPTR backend 'simple_ldb' NTVFS backend 'simple' for type 1 registered NTVFS backend 'cifs' for type 1 registered NTVFS backend 'nbench' for type 1 registered NTVFS backend 'unixuid' for type 1 registered NTVFS backend 'unixuid' for type 3 registered NTVFS backend 'unixuid' for type 2 registered NTVFS backend 'cifsposix' for type 1 registered NTVFS backend 'smb2' for type 1 registered NTV...

On Tue, Jan 29, 2013 at 3:59 PM, Murali, Sriram <sriram.murali at intel.com> wrote: > Our benchmark results show that the compilation time performance improved by > ~0.5%. That's fairly small; what was the standard deviation, confidence interval, etc? -- Sean Silva

...h264ref 100.00 102.10 470.lbm 100.00 100.07 471.omnetpp 100.00 100.36 bitmnp01 100.00 100.44 cjpegv2data6 100.00 100.76 idctrn01 100.00 99.83 libquake2 100.00 100.08 libquake_portable 100.00 100.02 libxcsoar 100.00 100.08 linpack 100.00 101.07 matrix01 100.00 100.28 nbench 100.00 100.76 tblook01 100.00 101.52 ttsprk01 100.00 100.82 Geomean 100.00 100.51 -----Original Message----- From: Sean Silva [mailto:silvas at purdue.edu] Sent: Tuesday, January 29, 2013 6:41 PM To: Murali, Sriram Cc: llvmdev at cs.uiuc.edu Subject: Re: [LLVMdev] [Patch][Review Req...

...s are: FDPASS LOCK1 LOCK2 LOCK3 UNLINK BROWSE ATTR TRANS2 MAXFID TORTURE RANDOMIPC NBW95 NBWNT default test is ALL But the smbtorture compiled with Gentoo's 3.0.2a gives me tests are: FDPASS LOCK1 LOCK2 LOCK3 LOCK4 LOCK5 LOCK6 LOCK7 UNLINK BROWSE ATTR TRANS2 MAXFID TORTURE RANDOMIPC NEGNOWAIT NBENCH OPLOCK1 OPLOCK2 OPLOCK3 DIR DIR1 DENY1 DENY2 TCON TCONDEV RW1 RW2 RW3 OPEN OPENATTR XCOPY RENAME DELETE PROPERTIES MANGLE W2K TRANS2SCAN NTTRANSSCAN UTABLE CASETABLE ERRMAPEXTRACT PIPE_NUMBER TCON2 IOCTL CHKPATH FDSESS default test is ALL What is the new equivalent to NBW95 and NBWNT, and what do...

...shared filesystem needed for ctdb is on a ocfs2 share stored on a ISCSI target. The cluster is running fine and ip takeover etc is working fine as well. To find out how the cluster would performe in real life with many clients accessing samba shares I compiled smbtorture (from samba4) to run the nbenchmark test using the loadfile client.txt from the dbench4.0 distribution. What I found out is really strange: I first tried to simulate 50 clients on one of the cluster nodes: $ bin/smbtorture //host1/smbtest1 -UUNIKO/smbtest1%password bench.nbench --loadfile=dbench-4.0/client.txt --num-progs=100...

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> </head> <body bgcolor="#ffffff" text="#000000"> Hello,<br> <br> We are a research project in joint french-chinese laboratory. We are considering using<br> LLVM in our project but we'd like to have some additional info before we dive in.

Hello, This patch aims to improve compile time performance by increasing the SCEV vector size in LoopStrengthReduce. It is observed that the BaseRegs vector size is 4 in most cases, and elements are frequently copied when it is initialized as SmallVector<const SCEV *, 2> BaseRegs. Our benchmark results show that the compilation time performance improved by ~0.5%. Patch by Wan Xiaofei.

Dear list, with linux compiling samba 4 works pretty fine but with opensolaris 'snv_111b' 64bit I have problems with the prototype declaration of getpwent_r(): (Using /opt/SunStudioExpress/bin/cc) Compiling torture/../../lib/nss_wrapper/testsuite.c "torture/../../lib/nss_wrapper/testsuite.c", line 336: prototype mismatch: 4 args passed, 3 expected