• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Re: Overload question
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Overload question


  • Subject: Re: Overload question
  • From: Jeff Moore <email@hidden>
  • Date: Tue, 11 Oct 2005 11:13:31 -0700

It would help if I could see all of the telemetry, especially the cycles before and after. A screen shot would be fine. It would also help to know what the IO cycle duration in samples is.

At any rate, the start time for the IO cycle is not necessarily the same value as the host time displayed for the cycle event. They are taken from different fields in the event data sometimes.

Feel free to talk to me off-list if you have large attachments to send.

On Oct 10, 2005, at 6:24 PM, Ted Barram wrote:
Hi,

I’m looking for the cause of some sporadic overload notifications with one of our usb devices. When I analyze the overload in the IO Cycle Telemetry window, I see that there is a large diff between the main ‘Start’ timestamp (next to the UI arrow) and the ‘->Cycle” timestamp (about 1ms). This diff appears to be the cause of the overload, because all the other timestamp diffs look reasonable (and are the same as in other non-overload cycles.)


Here’s a snippet of the first few entries on the first 2 lines (the V represents the UI arrow):



V 93327.517 ( +2.663 +0.030)

   ->Cycle:     34993 93328.604 (  +0.000)


The diff between 93328.604 and 93327.517 is what appears top be causing the overload. In all other non-overload cycles this diff is much smaller, typically .003 ms.



So it looks as if something is pre-empting the IOProc, but it always happens at the same place, ie right at the very beginning, so I wonder of something else is going on.



One other difference that I see in the overload cycle compared with the other cycles is that the Wake:samp time is later, it’s 181 samples on the overload cycle as compared to 128 samples on the succesful cycles.



Thanks in advance, if any more data is needed let me know.


-Ted





--

Jeff Moore
Core Audio
Apple


_______________________________________________ Do not post admin requests to the list. They will be ignored. Coreaudio-api mailing list (email@hidden) Help/Unsubscribe/Update your Subscription: This email sent to email@hidden
References: 
 >Overload question (From: "Ted Barram" <email@hidden>)

  • Prev by Date: Re: ClassInfo & PresentPreset confusion
  • Next by Date: Re: Audio Units - newbie questions
  • Previous by thread: Overload question
  • Next by thread: MTCoreAudio, CAAudioFileWriter and conversion
  • Index(es):
    • Date
    • Thread