Re: [Jack-Devel] softmode for dummy driver

PrevNext  Index
DateFri, 29 Aug 2014 01:10:28 +1000
From Patrick Shirkey <[hidden] at boosthardware dot com>
ToJACK <[hidden] at lists dot jackaudio dot org>
In-Reply-ToPaul Davis Re: [Jack-Devel] softmode for dummy driver
Follow-UpJoakim Hernberg Re: [Jack-Devel] softmode for dummy driver
On Fri, August 29, 2014 12:55 am, Paul Davis wrote:
> On Thu, Aug 28, 2014 at 10:43 AM, Patrick Shirkey <
> [hidden]> wrote:
>
>>
>> On Fri, August 29, 2014 12:35 am, Paul Davis wrote:
>> > On Thu, Aug 28, 2014 at 9:52 AM, Patrick Shirkey
>> > <[hidden]
>> >> wrote:
>> >
>> >> Hi,
>> >>
>> >> On the version of jack1 that I am running jackd 0.124.1 it seems that
>> >> the
>> >> alsa dummy driver does not support softmode. Am I missing something
>> or
>> >> if
>> >> not is there a specific reason it is not already available other than
>> no
>> >> one has bothered to implement it?
>> >>
>> >
>> > my understanding of softmode is that it is a property of the server
>> not
>> > the
>> > backend. but perhaps you mean something else?
>> >
>>
>> I meant to say the dummy driver not the alsa dummy driver. In case it is
>> of interest I am trying to track down the cause of the following error
>> so
>> I thought softmode might be useful in this case.
>>
>
> ah, *that* softmode.
>
> the answer is that softmode relates to xrun handling, and the dummy driver
> nevers gets xruns.
>
> xruns are not the same as conditions where a client takes too long,
> measured by some metric, which is what is happening in your case.
> obviously, clients taking too long can *lead* to xruns, but softmode in
> the
> ALSA driver just affects how it responds to reports of xruns from ALSA
> itself. since the dummy backend has no such concept ... no softmode.
>

Thanks for that explanation.

The app I am using loads a midi file and triggers the timeout error while
loading the file or just after starting playback (upto 200ms). However the
issue is happening randomly. Sometimes the same file loads and plays just
fine and sometimes the app triggers the error.

As the dummy driver is not affected by xruns is there an easy/obvious way
to identify the actual cause of the delay that is triggering the subgraph
timeout error?

I have tried extending the periods upto -p 4096 but that is not helping.
With varying period rates I I can run the app for a few hours and other
times it triggers the timeout error after a few minutes.

Could it be related to the other apps in the processing graph or is it
most likely an internal issue to the application that is causing the
error.


--
Patrick Shirkey
Boost Hardware Ltd
PrevNext  Index

1409238637.21079_0.ltw:2, <61845.86.107.254.57.1409238628.squirrel at boosthardware dot com>