Bug#729479: boost1.54: FTBFS on hppa -- BOOST_MATH_NO_LONG_DOUBLE_MATH_FUNCTIONS incorrectly defined

2013-12-01 Thread John David Anglin
On 1-Dec-13, at 1:20 AM, Steve M. Robbins wrote: Just to be absolutely clear: you were building the debian package sources and the debian/rules did specify --disable-long-double on the bjam command line, right? It looks like bjam has been removed from source: cd tools

Bug#729479: boost1.54: FTBFS on hppa -- BOOST_MATH_NO_LONG_DOUBLE_MATH_FUNCTIONS incorrectly defined

2013-12-01 Thread John David Anglin
Just to be absolutely clear: you were building the debian package sources and the debian/rules did specify --disable-long-double on the bjam command line, right? OK, I see it is created later: /home/dave/debian/boost1.54/boost1.54-1.54.0/bjam -d2 --disable-long- double --l ayout=system

Bug#729479: boost1.54: FTBFS on hppa -- BOOST_MATH_NO_LONG_DOUBLE_MATH_FUNCTIONS incorrectly defined

2013-12-01 Thread Steve M. Robbins
Hello, On December 1, 2013 12:49:12 PM John David Anglin wrote: Just to be absolutely clear: you were building the debian package sources and the debian/rules did specify --disable-long-double on the bjam command line, right? OK, I see it is created later:

Bug#729479: boost1.54: FTBFS on hppa -- BOOST_MATH_NO_LONG_DOUBLE_MATH_FUNCTIONS incorrectly defined

2013-12-01 Thread John David Anglin
On 1-Dec-13, at 3:25 PM, Steve M. Robbins wrote: I'm curious about another thing: the hppa buildd machines claim BD- Uninstallable for boost [1]. It claims that boost1.54 depends on debhelper that depends on perl which is missing. How did you get around that?

Bug#729479: boost1.54: FTBFS on hppa -- BOOST_MATH_NO_LONG_DOUBLE_MATH_FUNCTIONS incorrectly defined

2013-11-30 Thread Steve M. Robbins
On Wed, Nov 13, 2013 at 07:56:51AM -0500, John David Anglin wrote: Source: boost1.54 Version: 1.54.0-3 Severity: serious Tags: patch Thanks for the patch. Just to be absolutely clear: you were building the debian package sources and the debian/rules did specify --disable-long-double on the

Bug#729479: boost1.54: FTBFS on hppa -- BOOST_MATH_NO_LONG_DOUBLE_MATH_FUNCTIONS incorrectly defined

2013-11-13 Thread John David Anglin
Source: boost1.54 Version: 1.54.0-3 Severity: serious Tags: patch Justification: fails to build from source (but built successfully in the past) Build fails: cp bin.v2/libs/wave/build/gcc-4.6/release/debug-symbols-on/link-static/threading-multi/libboost_wave.a stage/lib/libboost_wave.a

Bug#729479: boost1.54: FTBFS on hppa -- BOOST_MATH_NO_LONG_DOUBLE_MATH_FUNCTIONS incorrectly defined

2013-11-13 Thread Cyril Brulebois
Control: severity -1 normal John David Anglin dave.ang...@bell.net (2013-11-13): Source: boost1.54 Version: 1.54.0-3 Severity: serious Tags: patch Justification: fails to build from source (but built successfully in the past) hppa isn't exactly a supported architecture anymore, so not a

Bug#729479: boost1.54: FTBFS on hppa -- BOOST_MATH_NO_LONG_DOUBLE_MATH_FUNCTIONS incorrectly defined

2013-11-13 Thread John David Anglin
On 11/13/2013 11:25 AM, Cyril Brulebois wrote: John David Anglindave.ang...@bell.net (2013-11-13): Then reportbug should be fixed to not set serious for arches in ports. The level is set automatically for any situation where fails to build is selected in the report. reportbug can't know which