[Scons-users] Segmentation faults in scons-2.2.0

[Evan Driscoll]

On 08/20/2013 11:39 AM, Chris Irvine wrote:

> I have used Cygwin in the past but I thought this Interix method would

> work, especially for very large where 64 bit is needed. datasets.  I’m

> finding that every program is tough to get going.


Do you actually need to run SCons from Interix, or could you just use
either native Python/SCons or Cygwin Python/SCons to run the Interix
compiler and build an Interix executable?

This may be a little obnoxious to arrange, but it might not be too bad.
My wild guess is the easiest might be to use native Python/SCons, use
the mingw tools, and fiddle with the PATH setting so it picks up the
Interix ones. (Well, assuming you're doing like a relatively simple
C/C++ build.)

Evan



> 

>  

> 

> The my problem is a configuration file that is missing from the Interix

> python setup.  It’s a file called pyconfig.h and its in the

> /usr/local/include/python2.6   directory.  Scons must need this.  I’ll

> have to get it created properly.

> 

>  

> 

>  

> 

>  

> 

> Cheers

> 

> --Chris

> 

>  

> 

> *From:*scons-users-bounces at scons.org

> [mailto:scons-users-bounces at scons.org] *On Behalf Of *Gary Oberbrunner

> *Sent:* Sunday, August 18, 2013 7:57 PM

> *To:* SCons users mailing list

> *Subject:* Re: [Scons-users] Segmentation faults in scons-2.2.0

> 

>  

> 

> In theory, no program written in python should be able to crash the

> python interpreter itself, which is what is happening to you.  What this

> means is there's a definitely a bug in python on Interix.  The crash may

> or may not be due to a higher-level bug (e.g. something in SCons), but

> at this point what you're seeing is too far from the SCons code to be

> able to tell.  In order to make further progress, you'd probably have to

> build python from source on your Interix box -- not at all impossible,

> if you have a C/C++ development environment, but might be more than

> you're willing to get into.

> 

>  

> 

> I know it's not really relevant, but might cygwin be a possible

> replacement for Interix for you?  It's much more widely used, I think,

> and probably better supported.

> 

>  

> 

> On Fri, Aug 16, 2013 at 7:04 PM, Chris Irvine <cirvine1 at telus.net

> <mailto:cirvine1 at telus.net>> wrote:

> 

> $ python --version

> 

> Python 2.6.2

> 

>  

> 

> The core can be looked at with that “gbd python core” command.  Thanks. 

> It shows the three warnings below and they are repeated numerous times,

> always as warnings.

> 

>  

> 

> This GDB was configured as "i586-pc-interix3"...

> 

> Core was generated by `python /usr/local/bin/scons -H'.

> 

> Reading symbols from /usr/lib/x86/libdl.so.5.2...done.

> 

> Loaded symbols for /usr/lib/x86/libdl.so.5.2

> 

> Reading symbols from /usr/local/lib/x86/libreadline.so...

> 

> warning: "_environ": indirect symbol does not have real one (-1)

> 

>  

> 

>  

> 

> warning: "_RaiseException at 16": indirect symbol does not have real one (-1)

> 

>  

> 

>  

> 

> warning: "_beThreadSafe": indirect symbol does not have real one (-1)

> 

>  

> 

> done.

> 

>  

> 

>  

> 

> My system is Interix 6.1 and 64 bit system.  Microsoft regard Interix as

> third-party software now.  The Microsoft link about vfork you had is for

> Interix 3 and written back in 2005.  I hope it has been fixed??  In the

> python makefile, it has  

> 

>  

> 

> MACHDEP=   interix6

> 

>  

> 

> The configuration may be the culprit. 

> 

>  

> 

>  

> 

> Cheers

> 

> --Chris

> 

>  

> 

> *From:*scons-users-bounces at scons.org

> <mailto:scons-users-bounces at scons.org>

> [mailto:scons-users-bounces at scons.org

> <mailto:scons-users-bounces at scons.org>] *On Behalf Of *Bill Deegan

> *Sent:* Thursday, August 15, 2013 7:34 PM

> *To:* SCons users mailing list

> 

> 

> *Subject:* Re: [Scons-users] Segmentation faults in scons-2.2.0

> 

>  

> 

> Try :

> python --version

> 

> ?

> 

> -Bill

> 

>  

> 

> On Thu, Aug 15, 2013 at 6:27 PM, Gary Oberbrunner <garyo at oberbrunner.com

> <mailto:garyo at oberbrunner.com>> wrote:

> 

>  

> 

>  

> 

> On Thu, Aug 15, 2013 at 9:07 PM, Chris Irvine <cirvine1 at telus.net

> <mailto:cirvine1 at telus.net>> wrote:

> 

> ...

> 

> Memory fault (core dumped)

> 

> $ scons

> 

> Memory fault (core dumped)

> 

> $ scons --debug=pdb

> 

> Memory fault (core dumped)

> 

>  

> 

>  

> 

> I don't know anything about Interix -- does Microsoft even still support

> it?  If it's Unix-like enough, you can try to do

> 

>   gdb python core

> 

> and then type 'where' to see the stack trace. But that will be at too

> low a level to see what python was trying to do at the time of failure.

>  And your log above shows it might not even be writing the core files it

> says it is.

> 

> I suppose it could be this issue: http://support.microsoft.com/kb/328684

> Might be worth trying that patch anyway.

> 

>  

> 

> Does python work at all?

> 

>  

> 

> -- 

> Gary

> 

> 

> _______________________________________________

> Scons-users mailing list

> Scons-users at scons.org <mailto:Scons-users at scons.org>

> http://four.pairlist.net/mailman/listinfo/scons-users

> 

>  

> 

> 

> _______________________________________________

> Scons-users mailing list

> Scons-users at scons.org <mailto:Scons-users at scons.org>

> http://four.pairlist.net/mailman/listinfo/scons-users

> 

> 

> 

>  

> 

> -- 

> Gary

> 

> 

> 

> _______________________________________________

> Scons-users mailing list

> Scons-users at scons.org

> http://four.pairlist.net/mailman/listinfo/scons-users

> 



-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 555 bytes
Desc: OpenPGP digital signature
Url : http://four.pairlist.net/pipermail/scons-users/attachments/20130820/cf3a0b5c/attachment.pgp