Re: [sc-dev] Fwd: Ugen crashes server repeatedly to start with and then start's behaving itself

[Dan Stowell <danstowell+sc3@xxxxxxxxx>]

2013/11/4 Tristan Strange <tristan.strange@xxxxxxxxx>:
> Thanks for the advice Dan... I've not found anything memory management wise
> but think I've found the issue.
>
> I've just tested my ugen about 30 times and it now seems to be running
> without issue. Of course with these sort of problems it can be difficult to
> tell if its fixed or not!
>
> Is it possible that initialising m_fbufnum to -1e9f could have fixed the
> issue?

Yes - if m_fbufnum is uninitialised then it could be anything. it's
very easy for it to coincidentally match up exactly with the buffer
number you want (because it's easy for it to have a stale value from
last time you used exactly that buffer), so the UGen might well
proceed as if it has acquired the buffer info, when in fact it hasn't.

> Is it possible all of this was down to the buffer not being ready for
> reading something?

Nah

Best
Dan


> Many thanks,
> Tris
>
>
>
>
> On 3 November 2013 22:12, Dan Stowell <danstowell+sc3@xxxxxxxxx> wrote:
>>
>> Hi Tristan,
>>
>> Well, I had a look, but I'm afraid I saw no glaring errors.
>> Inconsistent crashing often indicates bad memory accesses, and in SC
>> plugins these can be tricky to detect using developer tools. I guess
>> at the least, I'd suggest temporarily adding some checks that your
>> indices into the buffer data are sane.
>>
>> Best
>> Dan
>>
>>
>>
>>
>> 2013/11/3 Tristan Strange <tristan.strange@xxxxxxxxx>:
>> > ok here's the source: https://github.com/triss/BeatWarperUgen
>> >
>> > It's a touch messy at the moment but hopefully legible!
>> >
>> > And sorry for all the noise.
>> >
>> >
>> > ---------- Forwarded message ----------
>> > From: Tristan Strange <tristan.strange@xxxxxxxxx>
>> > Date: 3 November 2013 19:02
>> > Subject: Fwd: Ugen crashes server repeatedly to start with and then
>> > start's
>> > behaving itself
>> > To: sc-dev@xxxxxxxxxxxxxxxx
>> >
>> >
>> > Tristan Strange wrote:
>> >> Unfortunately the thing won't reliably start and causes the server to
>> >> crash 0 > - 20 times before starting up and performing as I expect it
>> >> to.
>> >> Once I've its > started once it runs fine form then onwards.
>> >
>> > Oh pants.
>> >
>> > Turns out it doesn't always run fine after its run fine once... 1 out of
>> > 50
>> > times it fails to start and causes the server to crash with RETURN = 0
>> > again.
>> >
>> >
>> >
>>
>>
>>
>> --
>> http://www.mcld.co.uk
>>
>> _______________________________________________
>> sc-dev mailing list
>>
>> info (subscription, etc.):
>> http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
>> archive: https://listarc.bham.ac.uk/marchives/sc-dev/
>> search: https://listarc.bham.ac.uk/lists/sc-dev/search/
>
>



-- 
http://www.mcld.co.uk

_______________________________________________
sc-dev mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: https://listarc.bham.ac.uk/marchives/sc-dev/
search: https://listarc.bham.ac.uk/lists/sc-dev/search/