[Pkg-phototools-devel] Bug#513410: Stitching script generates error after it deletes the intermediate image files

[Andreas Metzler]

On 2009-05-09 jedd <jedd at progsoc.org> wrote:
> On Sunday 26 April 2009, Andreas Metzler wrote:
>> This looks similar to #525494. Do you perhaps have set the
>> environment variable MAKEFLAGS=-jX? (Witch X > 1, check with
>> printenv).

>  Yes, Andreas, you're bang on - it looks like the same bug, and I do
>  indeed have -j set - mostly for the benefits it gives, even on this
>  rinky Atom 1.6GHz.

I will merge the bug reports, thanks.

>  I'm loathe to disable the -j setting, however, for that reason.

>  I see in #525494 that you suggest the hugin Makefiles may be
>  the causal factor - is this something you guys are taking upstream
>  to the hugin developers? 

We have already forwarded the bug report upstream
http://sourceforge.net/tracker/?func=detail&aid=2781528&group_id=77506&atid=550441

>  It seems that it's not a Debian problem,
>  of course, so I'd be surprised if only us two Debian users have
>  struck the problem.  Even more so given that multi-core CPUs are
>  now the norm rather than the exception.

Well, the only thing in a usual stitching process that is easy to run in
parallel (nona) can make use of multiple CPUs by using threading without 
MAKEFLAGS=-jX. (Use File -> Settings -> # of CPUs).

>  What do you recommend (other than either what I'm doing now -
>  temporarily renaming rm - or going back to a single make thread
>  via MAKEFILES) for the long term?

Renaming rm seems a little bit heavy handed. ;-) How about running
"env -u MAKEFLAGS hugin" instead of "hugin"?

For the long term I expect upstream to fix this.

cu andreas
-- 
`What a good friend you are to him, Dr. Maturin. His other friends are
so grateful to you.'
`I sew his ears on from time to time, sure'