Re: [boost] Regressions in your Boost libraries as of 2005-06-27
Douglas Gregor <dgregor@cs.indiana.edu> writes:
You are receiving this report because one or more of the libraries you maintain has regression test failures that are not accounted for. A full version of the report is sent to the Boost developer's mailing list.
Detailed report: http://engineering.meta-comm.com/boost-regression/CVS-HEAD/developer/issues....
There are failures in these libraries you maintain: python (4) type_traits (2)
|python| extract: gcc-3.3.6-linux list: gcc-3.3.6-linux operators: gcc-3.3.6-linux slice: gcc-3.3.6-linux
Hi Doug, Is there a machine I can login to at OSL where I can GDB this setup? I could build that compiler myself, but... I wonder a little why we're testing 3.3.6 when there's a 3.4.0 release, but I s'pose my wondering won't change anything ;-)
|type_traits| decay_test: cw-8_3
This one should be fixed now.
is_enum_test: vc-8_0
The chart disagrees. -- Dave Abrahams Boost Consulting www.boost-consulting.com
On Jun 27, 2005, at 1:24 PM, David Abrahams wrote:
Is there a machine I can login to at OSL where I can GDB this setup? I could build that compiler myself, but...
eddie.osl.iu.edu You should have access already, and the toolset looks like this: { GCC_ROOT_DIRECTORY = "/u/dgregor/Build/i686-pc-linux-gnu/gcc-3.3.6" ; extends-toolset gcc ; PYTHON_VERSION = 2.3 ; PYTHON_ROOT = "/u/dgregor/Build/i686-pc-linux-gnu/gcc-3.3.6" ; } You might want to wait a day. Some transient network failure kept eddie from submitting today, so the error you're seeing is still using the system-provided Python (that was also causing problems for gcc-3.4.4). Tomorrow's results will use the gcc-3.3.6-compiler Python.
I wonder a little why we're testing 3.3.6 when there's a 3.4.0 release, but I s'pose my wondering won't change anything ;-)
Because there are lots of Linux distributions out there using GCC 3.3.x. You let me drop MSVC 7.0 from the BGL and I'll let you drop GCC 3.3.x from the Python lib :) Doug
Douglas Gregor <doug.gregor@gmail.com> writes:
On Jun 27, 2005, at 1:24 PM, David Abrahams wrote:
Is there a machine I can login to at OSL where I can GDB this setup? I could build that compiler myself, but...
eddie.osl.iu.edu
You should have access already, and the toolset looks like this:
{ GCC_ROOT_DIRECTORY = "/u/dgregor/Build/i686-pc-linux-gnu/gcc-3.3.6" ; extends-toolset gcc ; PYTHON_VERSION = 2.3 ; PYTHON_ROOT = "/u/dgregor/Build/i686-pc-linux-gnu/gcc-3.3.6" ; }
I'm testing now, but you should know that this method of setting up the Python location is not going to work in general. You only get one setting of those variables for your entire test run (two if you're on Windows, so you can test Cygwin as well), and setting them in the toolset may be too late. In principle they have to be set on the command-line or in the environment, like their doc says. -- Dave Abrahams Boost Consulting www.boost-consulting.com
David Abrahams <dave@boost-consulting.com> writes:
Douglas Gregor <doug.gregor@gmail.com> writes:
On Jun 27, 2005, at 1:24 PM, David Abrahams wrote:
Is there a machine I can login to at OSL where I can GDB this setup? I could build that compiler myself, but...
eddie.osl.iu.edu
You should have access already, and the toolset looks like this:
{ GCC_ROOT_DIRECTORY = "/u/dgregor/Build/i686-pc-linux-gnu/gcc-3.3.6" ; extends-toolset gcc ; PYTHON_VERSION = 2.3 ; PYTHON_ROOT = "/u/dgregor/Build/i686-pc-linux-gnu/gcc-3.3.6" ; }
I'm testing now, but you should know that this method of setting up the Python location is not going to work in general. You only get one setting of those variables for your entire test run (two if you're on Windows, so you can test Cygwin as well), and setting them in the toolset may be too late. In principle they have to be set on the command-line or in the environment, like their doc says.
And when I do things the right way, everything passes with flying colors. -- Dave Abrahams Boost Consulting www.boost-consulting.com
participants (2)
-
David Abrahams -
Douglas Gregor