Re: garbage-collection/trampoline error with beams and accidentals

[David Kastrup]

"N. Andrew Walsh" <address@hidden> writes:

> Hi David,
>
> On Fri, Jan 11, 2019 at 5:58 PM David Kastrup <address@hidden> wrote:
>
>>
>> Just put the whole path in "double quote marks".
>>
>
> thanks. OK, now I must confess I don't really know what was going wrong
> exactly, but I resolved the error.

You cannot "resolve" this error since it is nothing a user should be
able to produce in the first place.  You may have been able to avoid
triggering it.

> My main file had a bunch of variables defining each instrument, and a
> separate file for score layout that placed them. A bunch of those
> variables weren't yet defined; now that they are, the error no longer
> appears. I have no idea how that relates to whether the viola part had
> accidentals on some notes, or whether a rest was beamed, or whatever,
> but now it works again.
>
> Coding discipline saves the day, I guess.

Ah, that's just superstitious.  Coding chaos could have gotten the same
result in an error like that.  Basically any change can cause a
different garbage collection sequence/behavior and thus mask the
problem.

> Thanks for the help. Sorry I couldn't get anything more productive,
> but I couldn't for the life of me reproduce the error, at any point,
> using gdb.  It only failed in frescobaldi, and (apparently) only
> because some instruments were still undefined.

I think "because" is the wrong word.  Defining an instrument variable
will take up significant memory space, so garbage collection will happen
at a significantly different place.  That's all.

-- 
David Kastrup