- Boost-users - lists.boost.org

[BGL] Using BGL on Linux clusters
by Erik Arner 07 Feb '03

07 Feb '03

Hi, I'm still trying to figure out how to use the BGL on massive graphs (see previous thread "[BGL] Using BGL for huge graphs"). My next idea is to distribute the graph over nodes in a Linux (Beowolf) cluster, or some other parallel computer. I'd much appreciate input from people who have tried this, with graph data structures in general and the BGL in particular. Thanks, Erik

1 0

Intel 7.0 Compiler and date_time library
by klapshin <klapshin＠yahoo.com> 07 Feb '03

07 Feb '03

First of all I would like to thank everybody who responded on my previous post. The real showstopper for me at the moment is date_time library. I did a small test -- compiled and run days_till_new_year.cpp against boost 1.29 (patched to not complain about __ICL 700) and current snapshot of boost CVS. Here is what I've got (cpp was renamed to test.cpp for short): ------------------------------------- C:\work\icl>icl Intel(R) C++ Compiler for 32-bit applications, Version 7.0 Build 20021018Z Copyright (C) 1985-2002 Intel Corporation. All rights reserved. 30 DAY EVALUATION LICENSE icl: Command line error: no files specified; for help type "icl -help" C:\work\icl>icl /MD /O2 /DDATE_TIME_INLINE /I..\boost_1_29_0 /nologo test.cpp test.cpp C:\work\icl>test Days till new year: -731982 C:\work\icl>icl /MD /Od /DDATE_TIME_INLINE /I..\boost_1_29_0 /nologo test.cpp test.cpp C:\work\icl>test Days till new year: 329 C:\work\icl>icl /MD /O2 /DDATE_TIME_INLINE /I..\boost /nologo test.cpp test.cpp C:\work\icl>test Days till new year: 3031841 C:\work\icl>icl /MD /Od /DDATE_TIME_INLINE /I..\boost /nologo test.cpp test.cpp C:\work\icl>test Days till new year: 329 ---------------------------------------------- As you can see both versions are broken. What I don't understand is how come that boost regression tests are Ok? Am I using wrong version of Intel Compiler, or am I missing some magic flags to compiler? I will send this report to Intel tech support; however I would appreciate if someone will verify these results, and contact Intel on behalf of Boost community. Regards, Kirill Lapshin.

2 3

RE: Re: [Boost-Users] HP, aCC and the standards
by Tom Matelich 06 Feb '03

06 Feb '03

The most common one is giving a name to bool template params. template <bool> becomes template <bool b> I'll see if I can track down the email address of the contact I was working with at HP, see if he'd be interested in working with you. > -----Original Message----- > From: Toon Knapen [mailto:toon.knapen@si-lab.org] > Sent: Thursday, February 06, 2003 12:02 AM > To: Boost-Users(a)yahoogroups.com > Subject: Re: Re: [Boost-Users] HP, aCC and the standards > > > On Thursday 06 February 2003 00:45, Tom Matelich wrote: > > Many of those errors are easily worked around, > That's good news. I have little experience with the newer > aCC's (I just converted from HPUX10.20) > so I don't know all the tricks but I'll be happy to learn > them (but I don't have direct support from > HP as you had ;-( > > > its just hard to get the > > developers to change their code. Or at least it was when I > was trying to > > get things going with aCC. > In my experience, whenever I contacted a developer for > portability issues on Aix or HPUX, > patches were accepted within 24 hours. > > toon > > > ------------------------ Yahoo! Groups Sponsor > ---------------------~--> > Get 128 Bit SSL Encryption! > http://us.click.yahoo.com/LIgTpC/vN2EAA/xGHJAA/EbFolB/TM > -------------------------------------------------------------- > -------~-> > > Info: <http://www.boost.org> > Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> > Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> > > > Your use of Yahoo! Groups is subject to > http://docs.yahoo.com/info/terms/ > > ----------------------------------------------------------------------- DISCLAIMER: Information contained in this message and/or attachment(s) may contain confidential information of Zetec, Inc. If you have received this transmission in error, please notify the sender by return email. -----------------------------------------------------------------------

1 0

RE: [Boost-Users] Intel 7.0 Compiler problems
by Tom Matelich 06 Feb '03

06 Feb '03

It looks like Beman has upgraded to 7.0, if you look at http://boost.sourceforge.net/regression-logs/cs-win32.html > -----Original Message----- > From: klapshin <klapshin(a)yahoo.com> [mailto:klapshin@yahoo.com] > Sent: Wednesday, February 05, 2003 5:06 PM > To: Boost-Users(a)yahoogroups.com > Subject: [Boost-Users] Intel 7.0 Compiler problems > > > Hello, > > > > > Tired of MSVC deficiencies, and hoping to get certain performance > increase I got evaluation version of Intel C++ compiler. I've heard > lots of good things about this compiler, and was under impression > that > transition from MSVC to ICL should be very smooth. Huh, I had all > kinds of troubles. > > > > > First of all boost 1.29 gave up saying that it is an unknown > compiler. > Looking into config I figured that last known ICL was 6.0, so I just > hacked it badly by redefining ICL as 600 prior to including any boost > headers. I understand that it is not a good way, but hey, I am just > evaluating it. > > > > > Ok now everything compiled, but code using regex failed to link. No > wonder, I have not build regex library with new compiler. It turns > out > that there is no makefile for any intel compiler, so I tried to do it > with bjam, but seems that bjam did not set CC variable; it invoked > compilation with just line of flags. At this point I gave up and > decided to focus on parts of my project which do not depend on regex. > > > > > Next surprise was when I tried to run tests, in debug mode everything > was fine, but release mode failed miserably. It appears that > date_time > library does not work at all with optimization. Even setting /O1 did > not help, only /Od worked. > > > > > Failures were in very basic date operations, for instance > > > > > boost::gregorian::date(2002,12,27) == > boost::gregorian::date(2002,12,27) > > > > > is not true! > > > > > If I first construct a date, then ask for year, it throws, saying > year > is out of range. > > > > > I searched through Boost Users and Dev mailing lists, found few posts > discussing turning on strict ansi checks on intel, and also that in > this case it is better to use STLport, since STL coming with MSVC > could not be compiled in ansi mode. I tried to do that but can't > compile STLport in strict mode either. Using STLport in regular mode > did not have any effect on problems I am having. > > > > > Here is version the compiler reports: > > > Intel(R) C++ Compiler for 32-bit applications, Version 7.0 Build > 20021018Z > > > Copyright (C) 1985-2002 Intel Corporation. All rights reserved. > > > > > Now, my question is: are these problems well known? If they are, is > there any plans to fix them in next release? > > > And while we at it, when should we expect new release? I know there > were some discussions about mid-February, is there any update? > > > > > Regards, > > > Kirill. > > > > ------------------------ Yahoo! Groups Sponsor > ---------------------~--> > Get 128 Bit SSL Encryption! > http://us.click.yahoo.com/LIgTpC/vN2EAA/xGHJAA/EbFolB/TM > -------------------------------------------------------------- > -------~-> > > Info: <http://www.boost.org> > Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> > Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> > > > Your use of Yahoo! Groups is subject to > http://docs.yahoo.com/info/terms/ > > ----------------------------------------------------------------------- DISCLAIMER: Information contained in this message and/or attachment(s) may contain confidential information of Zetec, Inc. If you have received this transmission in error, please notify the sender by return email. -----------------------------------------------------------------------

1 0

ublas: conversion looses qualifiers
by speedsnaii <speedsnaii＠yahoo.de> 06 Feb '03

06 Feb '03

A newbie question: I have: void foo(matrix<double>& Y, const matrix<double>& X) { ... Y[0] = X[0]; // the compiler chokes on this ... } What is wrong with this? Please tell me. Thank you. Roland

1 0

RE: [Boost-Users] Intel 7.0 Compiler problems
by Tom Matelich 06 Feb '03

06 Feb '03

Hmm, I'm not a very good evaluator I guess, I just went about my business using 7.0 without seeing if it worked correctly :) Rather than hacking ICL, you can modify boost/compiler/intel.hpp changing 600's to 700's. I have my own dsp's for building regex, so the Compiler Selection feature worked for me (Though I've renamed cl.exe to make sure its never called because it was getting called sometimes) This is not to say I haven't had any troubles, but none with boost until the subject of my post a few minutes ago. > -----Original Message----- > From: klapshin <klapshin(a)yahoo.com> [mailto:klapshin@yahoo.com] > Sent: Wednesday, February 05, 2003 5:06 PM > To: Boost-Users(a)yahoogroups.com > Subject: [Boost-Users] Intel 7.0 Compiler problems > > > Hello, > > > > > Tired of MSVC deficiencies, and hoping to get certain performance > increase I got evaluation version of Intel C++ compiler. I've heard > lots of good things about this compiler, and was under impression > that > transition from MSVC to ICL should be very smooth. Huh, I had all > kinds of troubles. > > > > > First of all boost 1.29 gave up saying that it is an unknown > compiler. > Looking into config I figured that last known ICL was 6.0, so I just > hacked it badly by redefining ICL as 600 prior to including any boost > headers. I understand that it is not a good way, but hey, I am just > evaluating it. > > > > > Ok now everything compiled, but code using regex failed to link. No > wonder, I have not build regex library with new compiler. It turns > out > that there is no makefile for any intel compiler, so I tried to do it > with bjam, but seems that bjam did not set CC variable; it invoked > compilation with just line of flags. At this point I gave up and > decided to focus on parts of my project which do not depend on regex. > > > > > Next surprise was when I tried to run tests, in debug mode everything > was fine, but release mode failed miserably. It appears that > date_time > library does not work at all with optimization. Even setting /O1 did > not help, only /Od worked. > > > > > Failures were in very basic date operations, for instance > > > > > boost::gregorian::date(2002,12,27) == > boost::gregorian::date(2002,12,27) > > > > > is not true! > > > > > If I first construct a date, then ask for year, it throws, saying > year > is out of range. > > > > > I searched through Boost Users and Dev mailing lists, found few posts > discussing turning on strict ansi checks on intel, and also that in > this case it is better to use STLport, since STL coming with MSVC > could not be compiled in ansi mode. I tried to do that but can't > compile STLport in strict mode either. Using STLport in regular mode > did not have any effect on problems I am having. > > > > > Here is version the compiler reports: > > > Intel(R) C++ Compiler for 32-bit applications, Version 7.0 Build > 20021018Z > > > Copyright (C) 1985-2002 Intel Corporation. All rights reserved. > > > > > Now, my question is: are these problems well known? If they are, is > there any plans to fix them in next release? > > > And while we at it, when should we expect new release? I know there > were some discussions about mid-February, is there any update? > > > > > Regards, > > > Kirill. > > > > ------------------------ Yahoo! Groups Sponsor > ---------------------~--> > Get 128 Bit SSL Encryption! > http://us.click.yahoo.com/LIgTpC/vN2EAA/xGHJAA/EbFolB/TM > -------------------------------------------------------------- > -------~-> > > Info: <http://www.boost.org> > Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> > Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> > > > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/ ----------------------------------------------------------------------- DISCLAIMER: Information contained in this message and/or attachment(s) may contain confidential information of Zetec, Inc. If you have received this transmission in error, please notify the sender by return email. -----------------------------------------------------------------------

1 0

BGL: BFSVisitor::cycle_edge() ?
by Tarjei Knapstad 06 Feb '03

06 Feb '03

This probably is an error in the documentation (and the book), but I thought I'd just check to make sure: The documentation for the EventVisitor concept states: <quote> For example, the BFSVisitor of breadth_first_search() has a cycle_edge() member function. The corresponding tag is on_cycle_edge. </quote> As far as I can see BFSVisitor has no cycle_edge() function, and there is no on_cycle_edge tag in the code. The same function is also mentioned in chapter 12.3.1 of the book in the documentation of gray_target(e,g) and black_target(e,g) where it says that "A call to this function is always preceded by a call to the cycle_edge() function.". Kind of strange as the cycle_edge() function is not documented in the same chapter :) Regards, Tarjei

1 0

Re: Intel 7.0 Compiler problems
by Bruno.Voigt＠ic3s.de 06 Feb '03

06 Feb '03

Under Linux Red Hat 8.1 Beta the lastest ICC/IA32 0.78 works without problems with boost regex 1.29. I just copiend over the libs/regex/build/ gcc.mak and gcc-shared.mak files to icc*.mak and changed the compiler name within to icc. Under RH8.0 I had some problems because the compiler chooked about the header containing the declarations for some tcp/ip data conversion functions htons() etc.. I've never used the bjam stuff up to now. Perhaps you can do the same with the regex vc7.mak file as a starting point. with best regards, Bruno -- bruno(a)ic3s.de [Non-text portions of this message have been removed]

1 0

Re: Re: [Boost-Users] HP, aCC and the standards
by Toon Knapen 06 Feb '03

06 Feb '03

On Thursday 06 February 2003 00:45, Tom Matelich wrote: > Many of those errors are easily worked around, That's good news. I have little experience with the newer aCC's (I just converted from HPUX10.20) so I don't know all the tricks but I'll be happy to learn them (but I don't have direct support from HP as you had ;-( > its just hard to get the > developers to change their code. Or at least it was when I was trying to > get things going with aCC. In my experience, whenever I contacted a developer for portability issues on Aix or HPUX, patches were accepted within 24 hours. toon

1 0

Intel 7.0 Compiler problems
by klapshin <klapshin＠yahoo.com> 06 Feb '03

06 Feb '03

Hello, Tired of MSVC deficiencies, and hoping to get certain performance increase I got evaluation version of Intel C++ compiler. I've heard lots of good things about this compiler, and was under impression that transition from MSVC to ICL should be very smooth. Huh, I had all kinds of troubles. First of all boost 1.29 gave up saying that it is an unknown compiler. Looking into config I figured that last known ICL was 6.0, so I just hacked it badly by redefining ICL as 600 prior to including any boost headers. I understand that it is not a good way, but hey, I am just evaluating it. Ok now everything compiled, but code using regex failed to link. No wonder, I have not build regex library with new compiler. It turns out that there is no makefile for any intel compiler, so I tried to do it with bjam, but seems that bjam did not set CC variable; it invoked compilation with just line of flags. At this point I gave up and decided to focus on parts of my project which do not depend on regex. Next surprise was when I tried to run tests, in debug mode everything was fine, but release mode failed miserably. It appears that date_time library does not work at all with optimization. Even setting /O1 did not help, only /Od worked. Failures were in very basic date operations, for instance boost::gregorian::date(2002,12,27) == boost::gregorian::date(2002,12,27) is not true! If I first construct a date, then ask for year, it throws, saying year is out of range. I searched through Boost Users and Dev mailing lists, found few posts discussing turning on strict ansi checks on intel, and also that in this case it is better to use STLport, since STL coming with MSVC could not be compiled in ansi mode. I tried to do that but can't compile STLport in strict mode either. Using STLport in regular mode did not have any effect on problems I am having. Here is version the compiler reports: Intel(R) C++ Compiler for 32-bit applications, Version 7.0 Build 20021018Z Copyright (C) 1985-2002 Intel Corporation. All rights reserved. Now, my question is: are these problems well known? If they are, is there any plans to fix them in next release? And while we at it, when should we expect new release? I know there were some discussions about mid-February, is there any update? Regards, Kirill.

2 1