On Thu, Jul 7, 2011 at 5:30 PM, Alan Braslau wrote:

On Thu, Jul 07, 2011 at 11:36:32AM +0200, Mojca Miklavec wrote:

...

You don't need to be sorry. It was fixed 20 minutes ago.

mtx-context     | fatal error: no return code, message: luatex: execution interrupted

Oops! luatex now dumps core. No minimal test file (yet). (must be triggered by some metapost code - looking into locating it.)

Alan Latest minimals with Linux 32 bit is ok with

\starttext \input knuth \stoptext -- luigi

On Thu, Jul 07, 2011 at 05:39:05PM +0200, luigi scarso wrote:

...

On Thu, Jul 7, 2011 at 5:30 PM, Alan Braslau wrote:

...

On Thu, Jul 07, 2011 at 11:36:32AM +0200, Mojca Miklavec wrote:

...

You don't need to be sorry. It was fixed 20 minutes ago.

mtx-context     | fatal error: no return code, message: luatex: execution interrupted

Oops! luatex now dumps core. No minimal test file (yet). (must be triggered by some metapost code - looking into locating it.)

Alan Latest minimals with Linux 32 bit is ok with

\starttext

\input knuth

\stoptext

-- luigi

The luatex core dump (both on freebsd-amd64 and linux-amd64) is due to a metapost bug (without any diagnostic messages). I do not believe that it is related to the recent updates but rather to a (recent) error in my coding.

Taco may want to chase this, but I will need to provide him with a reproducible minimal test file... Just as an indication, I found two big errors in my metapost code: 1) a typo resulting in assignment to an array element of an undeclared array; 2) reference to a loop variable outside of a loop (which, of course, does not work). I believe that the core dump was due to the first mistake, but I cannot reproduce it simply. Ah ok --- there was also an email about an error in metapost 1.5 on

On Fri, Jul 8, 2011 at 8:30 AM, Alan Braslau wrote: the metapost list, but maybe it's not related. -- luigi