Not successfully coming through 'documentation' build

[John W. Eaton]

On  6-Feb-2008, Thomas Weber wrote:

| On 05/02/08 17:21 +0100, Thomas Treichl wrote:
| > Thomas Weber schrieb:
| >> Am Montag, den 04.02.2008, 20:57 +0100 schrieb Thomas Treichl:
| >>
| >>> <SNIP>
| >>>
| >>>    cp gplot.png HTML
| >>>    mkdir: HTML: File exists
| >>>    make[3]: *** [HTML/grid.png] Error 1
| >>>    make[3]: *** Waiting for unfinished jobs....
| >>>    make[2]: *** [interpreter] Error 2
| >>>    make[1]: *** [doc] Error 2
| >>>    make: *** [all] Error 2
| >>
| >> Do you build in parallel?
| >>
| >> 	Thomas
| >
| > Oh yes, I did a parallel build this time (normally I don't).
| > If you are asking me that way then I expect this wasn't a good idea?
| 
| Oh, it actually is a good idea; multi-core machines are increasingly
| available. But I saw this behaviour when doing parallel builds. There
| seems to be a race condition somewhere in the documentation generating
| Makefile.in.
| 
| I think (but currently can't check) I reported this to the bug list.
| 
| Does adding a 
| .NOTPARALLEL
| at the end of the corresponding Makefile.in help?

That would probably work, but it doesn't seem like the best solution
to turn off all parallel execution in the doc/interpreter directory,
which builds a lot of figures.

How about the following patch instead?

jwe


-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: diffs
Url: https://www.cae.wisc.edu/pipermail/bug-octave/attachments/20080206/8edd03ac/attachment.ksh