From damarkowitz at gmail.com Mon Jul 7 20:04:57 2014 From: damarkowitz at gmail.com (Dale Markowitz) Date: Mon, 7 Jul 2014 16:04:57 -0400 Subject: Polygraph compilation error with gcc 4.8.2 Message-ID: Hi All, I'm trying to install Polygraph 4.3.2 on a Linux box running Fedora 19 with gcc version 4.8.2 20131212 (Red Hat 4.8.2-7) (GCC). The compilation fails with the error: g++ -DHAVE_CONFIG_H -I../.. -I../../src -I../.. -I../../src -fpermissive -I/usr/include -g -O1 -Wall -Wwrite-strings -Woverloaded-virtual -MT StatPhase.o -MD -MP -MF .deps/StatPhase.Tpo -c -o StatPhase.o StatPhase.cc StatPhase.cc: In member function 'virtual void StatPhase::noteXactEvent(BcastChannel*, const Xaction*)': StatPhase.cc:416:29: error: redeclaration of 'CompoundXactInfo& compound' static CompoundXactInfo &compound = *CompoundXactInfo::Create(); ^ StatPhase.cc:390:37: error: 'const CompoundXactInfo* const compound' previously declared here if (const CompoundXactInfo* const compound = x->partOf()) { ^ make[3]: *** [StatPhase.o] Error 1 This seems like it might have to do with my version of gcc, but I'm not sure. Any thoughts? Thanks! Dale -------------- next part -------------- An HTML attachment was scrubbed... URL: From pavel.kazlenka at measurement-factory.com Tue Jul 8 12:25:31 2014 From: pavel.kazlenka at measurement-factory.com (Pavel Kazlenka) Date: Tue, 08 Jul 2014 15:25:31 +0300 Subject: Polygraph compilation error with gcc 4.8.2 In-Reply-To: References: Message-ID: <53BBE33B.2080002@measurement-factory.com> Hi Dale, You're completely right. Old polygraph versions can not be built by modern gcc. IIRC, gcc of version 4.5 is definitely able to build polygraph-4.3.2 (and gcc 4.6 probably). This is a known issue. To work the one around, you can: a) find or set up machine/virtual machine with some old gcc version (e.g. ubuntu 12.04 LTS or Centos6), build polygraph there and move binaries to your main machine; b) wait until polygraph site will be updated and new polygraph version become available for download; Best wishes, Pavel On 07/07/2014 11:04 PM, Dale Markowitz wrote: > Hi All, > I'm trying to install Polygraph 4.3.2 on a Linux box running Fedora > 19 with gcc version 4.8.2 20131212 (Red Hat 4.8.2-7) (GCC). The > compilation fails with the error: > > g++ -DHAVE_CONFIG_H -I../.. -I../../src -I../.. -I../../src > -fpermissive -I/usr/include -g -O1 -Wall -Wwrite-strings > -Woverloaded-virtual -MT StatPhase.o -MD -MP -MF .deps/StatPhase.Tpo > -c -o StatPhase.o StatPhase.cc > StatPhase.cc: In member function 'virtual void > StatPhase::noteXactEvent(BcastChannel*, const Xaction*)': > StatPhase.cc:416:29: error: redeclaration of 'CompoundXactInfo& compound' > static CompoundXactInfo &compound = *CompoundXactInfo::Create(); > ^ > StatPhase.cc:390:37: error: 'const CompoundXactInfo* const compound' > previously declared here > if (const CompoundXactInfo* const compound = x->partOf()) { > ^ > make[3]: *** [StatPhase.o] Error 1 > > This seems like it might have to do with my version of gcc, but I'm > not sure. Any thoughts? > > Thanks! > Dale > > > _______________________________________________ > Users mailing list > Users at web-polygraph.org > http://www.web-polygraph.org/mailman/listinfo/users -------------- next part -------------- An HTML attachment was scrubbed... URL: From unjc.email at gmail.com Tue Jul 15 15:58:55 2014 From: unjc.email at gmail.com (unjc email) Date: Tue, 15 Jul 2014 11:58:55 -0400 Subject: SMP support Message-ID: Hello there, In the recent log change, I notice the latest private release has mentioned about "better SMP support". I would like to know more about the newly added feature. How capable does the latest version support SMP? Does it utilize multiple cores so that it is not necessary to start multiple Polyclt / Polysrv processes on a single box? Best Regards, Jacky -------------- next part -------------- An HTML attachment was scrubbed... URL: From rousskov at measurement-factory.com Tue Jul 15 17:33:01 2014 From: rousskov at measurement-factory.com (Alex Rousskov) Date: Tue, 15 Jul 2014 11:33:01 -0600 Subject: SMP support In-Reply-To: References: Message-ID: <53C565CD.6040706@measurement-factory.com> On 07/15/2014 09:58 AM, unjc email wrote: > In the recent log change, I notice the latest private release has > mentioned about "better SMP support". I would like to know more about > the newly added feature. How capable does the latest version support > SMP? Does it utilize multiple cores so that it is not necessary to > start multiple Polyclt / Polysrv processes on a single box? Not quite. Polygraph v4.8.0 makes it easier to use multiple processes on the same box. Here is a list of the related features: > * Added --worker command line option to specify SMP worker ID. > * Support %worker macro in command line option values. > * The local r.n.g. seed value now defaults to worker ID. This > change does not affect old test scripts running new Polygraph > versions because the default worker ID is one, same as the old > local r.n.g. default. > * Added worker() PGL function to return the current worker ID. > * "${worker}" macro support in ips2names, credentials, URL traces. > * pgl2ldif --worker ID command line option. The plan is for Polygraph v4.9.0 to start those multiple processes automatically so that you do not have to write a script to do that, but there is currently no sponsor for that feature. HTH, Alex.