Compiling 2.1.1 on Mac OS X 10.1 - test failures

Steven D. Majewski sdm7g at Virginia.EDU
Fri Oct 5 10:44:21 EDT 2001


On Fri, 5 Oct 2001, Richard Jones wrote:

> Compiling went OK. Then I tried "make test". Oh dear.
> 
> test_largefile: took about 2.5 minutes. Top reported that it was mostly 
> "stuck" and occasionally "running" - using very little CPU and no disk that I 
> could detect (by the scientific aural detection process ;) When it was 
> finished, it printed the message "@test exists: 0". No idea if this normal - 
> my Linux machine doesn't have large file support.

I never have enough free space on my iBook for test_largefile to 
complete, so I don't know what SHOULD happen.

> test_re: got a segfault (with -O3, -O2 and no optimisation at all)

This one is a known problem. You have to set the stack limit in the 
shell with "limit stack 4096" (for tcsh -- 'ulimit' in bash or sh).
Nobody has yet figured out how to set this limit for the program -- 
I've tried some of the linker switches which would seem to control
this, but it still seems to need to be set in the shell. 

-- Steve Majewski






More information about the Python-list mailing list