[math] Warnings in version 1.46.1
I am trying to build the Boost 1.46.1 libraries on a Fedora 7 Linux system with the default g++ 4.1.2. The warnings from the math library produce zillions of lines of output. Each warning has a long description of the instantiation and then concludes with a couple of lines like this: ./boost/math/special_functions/binomial.hpp:65: instantiated from here ./boost/mpl/assert.hpp:79: warning: lowering visibility of 'int mpl_::assertion_failed(typename mpl_::assert<C>::type) [with bool C = false]' to match its type Is this problem known? I hope it doesn't continue in future releases. -- Dick Hadsell 203-992-6320 Fax: 203-992-6001 Reply-to: hadsell@blueskystudios.com Blue Sky Studios http://www.blueskystudios.com 1 American Lane, Greenwich, CT 06831-2560
I am trying to build the Boost 1.46.1 libraries on a Fedora 7 Linux system with the default g++ 4.1.2. The warnings from the math library produce zillions of lines of output. Each warning has a long description of the instantiation and then concludes with a couple of lines like this:
./boost/math/special_functions/binomial.hpp:65: instantiated from here ./boost/mpl/assert.hpp:79: warning: lowering visibility of 'int mpl_::assertion_failed(typename mpl_::assert<C>::type) [with bool C = false]' to match its type
Is this problem known? I hope it doesn't continue in future releases.
I can't immediately reproduce here, do you have a test case? Thanks, John.
John Maddock wrote:
I am trying to build the Boost 1.46.1 libraries on a Fedora 7 Linux system with the default g++ 4.1.2. The warnings from the math library produce zillions of lines of output. Each warning has a long description of the instantiation and then concludes with a couple of lines like this:
./boost/math/special_functions/binomial.hpp:65: instantiated from here ./boost/mpl/assert.hpp:79: warning: lowering visibility of 'int mpl_::assertion_failed(typename mpl_::assert<C>::type) [with bool C = false]' to match its type
Is this problem known? I hope it doesn't continue in future releases.
I can't immediately reproduce here, do you have a test case? I blew away bin.v2/libs/math, so that the math library would rebuild. Then I ran this command:
bin/bjam --user-config=user-config.LINUX_AMDF7 --debug-configuration --without-python threading=multi instruction-set=opteron address-model=64 variant=release link=static,shared --stagedir=stageF7 stage | & tee build.release.math.log I am attaching the first part of the log, showing the messages from the first gcc.compile.c++.pch and gcc.compile.c++ compilations. -- Dick Hadsell 203-992-6320 Fax: 203-992-6001 Reply-to: hadsell@blueskystudios.com Blue Sky Studios http://www.blueskystudios.com 1 American Lane, Greenwich, CT 06831-2560
I can't immediately reproduce here, do you have a test case? I blew away bin.v2/libs/math, so that the math library would rebuild. Then I ran this command:
bin/bjam --user-config=user-config.LINUX_AMDF7 --debug-configuration --without-python threading=multi instruction-set=opteron address-model=64 variant=release link=static,shared --stagedir=stageF7 stage | & tee build.release.math.log
I am attaching the first part of the log, showing the messages from the first gcc.compile.c++.pch and gcc.compile.c++ compilations.
Thanks, still can't reproduce here on Ubuntu, but I found a few messages about this warning from Googling around - it appears to be a Red-Hat specific extension to GCC that chucks out a warning for every single member of a class that has default visibility of hidden :-(( It may even be specific to Fedora-7 and your gcc-4.1 for all I can tell as it appears to have caused quite a few folks problems. Can you try again with the option cxxflags=-Wno-visibility added? Thanks, John.
I can't immediately reproduce here, do you have a test case? I blew away bin.v2/libs/math, so that the math library would rebuild. Then I ran this command:
bin/bjam --user-config=user-config.LINUX_AMDF7 --debug-configuration --without-python threading=multi instruction-set=opteron address-model=64 variant=release link=static,shared --stagedir=stageF7 stage | & tee build.release.math.log
I am attaching the first part of the log, showing the messages from the first gcc.compile.c++.pch and gcc.compile.c++ compilations.
Correction: I meant to say cxxflags=-Wno-attributes. John.
John Maddock wrote:
Correction: I meant to say cxxflags=-Wno-attributes.
Thank you. I added it to the <compileflags> and <linkflags> in my user-config file, and those hideous warnings are gone (suppressed and forgotten). -- Dick Hadsell 203-992-6320 Fax: 203-992-6001 Reply-to: hadsell@blueskystudios.com Blue Sky Studios http://www.blueskystudios.com 1 American Lane, Greenwich, CT 06831-2560
-----Original Message----- From: boost-bounces@lists.boost.org [mailto:boost-bounces@lists.boost.org] On Behalf Of Richard Hadsell Sent: Friday, March 25, 2011 5:47 PM To: boost@lists.boost.org Subject: Re: [boost] [math] Warnings in version 1.46.1
John Maddock wrote:
Correction: I meant to say cxxflags=-Wno-attributes.
Thank you. I added it to the <compileflags> and <linkflags> in my user-config file, and those hideous warnings are gone (suppressed and forgotten).
-- Dick Hadsell 203-992-6320 Fax: 203-992-6001 Reply-to: hadsell@blueskystudios.com Blue Sky Studios http://www.blueskystudios.com 1 American Lane, Greenwich, CT 06831-2560
Does anything about this need to be added to the https://svn.boost.org/trac/boost/wiki/Guidelines/WarningsGuidelines If only a note that this problem exists with this combination, and this is the 'workaround'? Paul --- Paul A. Bristow, Prizet Farmhouse, Kendal LA8 8AB UK +44 1539 561830 07714330204 pbristow@hetp.u-net.com
Does anything about this need to be added to the
https://svn.boost.org/trac/boost/wiki/Guidelines/WarningsGuidelines
If only a note that this problem exists with this combination, and this is the 'workaround'?
Please! Thanks, John.
-----Original Message----- From: boost-bounces@lists.boost.org [mailto:boost-bounces@lists.boost.org] On Behalf Of John Maddock Sent: Wednesday, March 30, 2011 6:11 PM To: boost@lists.boost.org Subject: Re: [boost] [math] Warnings in version 1.46.1
Does anything about this need to be added to the
https://svn.boost.org/trac/boost/wiki/Guidelines/WarningsGuidelines
If only a note that this problem exists with this combination, and this is the 'workaround'?
Please!
Done. Paul --- Paul A. Bristow, Prizet Farmhouse, Kendal LA8 8AB UK +44 1539 561830 07714330204 pbristow@hetp.u-net.com
Paul A. Bristow wrote:
Does anything about this need to be added to the
https://svn.boost.org/trac/boost/wiki/Guidelines/WarningsGuidelines
If only a note that this problem exists with this combination, and this is the 'workaround'?
This is up to you, specifically John M, who identified the warning suppression option my compiler needed. Also, I am not a user of the math library, so I don't know whether the warnings show up outside the build. I just build all of the libraries, so that they are available for our developers, if desired. BTW, there is a typo in the page you pointed to. There is a line with "-pendantic" instead of "-pedantic". (Would you call this comment pedantic?) -- Dick Hadsell 203-992-6320 Fax: 203-992-6001 Reply-to: hadsell@blueskystudios.com Blue Sky Studios http://www.blueskystudios.com 1 American Lane, Greenwich, CT 06831-2560
Does anything about this need to be added to the https://svn.boost.org/trac/boost/wiki/Guidelines/WarningsGuidelines
If only a note that this problem exists with this combination, and this is the 'workaround'?
This is up to you, specifically John M, who identified the warning suppression option my compiler needed. Also, I am not a user of the math library, so I don't know whether the warnings show up outside the build. I just build all of the libraries, so that they are available for our developers, if desired.
I don't *think* this will effect users... unless possibly they're building with -fvisibility=hidden. As soon as I get a chance I'll also install the latest fedora release and see if this is still an issue. John.
John Maddock wrote:
I don't *think* this will effect users... unless possibly they're building with -fvisibility=hidden.
As soon as I get a chance I'll also install the latest fedora release and see if this is still an issue. I did not see the warnings with g++ 4.5.1 on Fedora 14.
-- Dick Hadsell 203-992-6320 Fax: 203-992-6001 Reply-to: hadsell@blueskystudios.com Blue Sky Studios http://www.blueskystudios.com 1 American Lane, Greenwich, CT 06831-2560
I don't *think* this will effect users... unless possibly they're building with -fvisibility=hidden.
As soon as I get a chance I'll also install the latest fedora release and see if this is still an issue. I did not see the warnings with g++ 4.5.1 on Fedora 14.
OK thanks, looks like Red Hat fixed this then, whew! John.
-----Original Message----- From: boost-bounces@lists.boost.org [mailto:boost-bounces@lists.boost.org] On Behalf Of Richard Hadsell Sent: Wednesday, March 30, 2011 6:20 PM To: boost@lists.boost.org Subject: Re: [boost] [math] Warnings in version 1.46.1
Paul A. Bristow wrote:
Does anything about this need to be added to the
https://svn.boost.org/trac/boost/wiki/Guidelines/WarningsGuidelines
BTW, there is a typo in the page you pointed to. There is a line with "- pendantic" instead of "-pedantic". (Would you call this comment pedantic?)
Yes - but that is the most annoying things about pedants - they are always right! Thanks - and Fixed. Paul --- Paul A. Bristow, Prizet Farmhouse, Kendal LA8 8AB UK +44 1539 561830 07714330204 pbristow@hetp.u-net.com
participants (3)
-
John Maddock -
Paul A. Bristow -
Richard Hadsell