This is the mail archive of the
ecos-discuss@sources.redhat.com
mailing list for the eCos project.
RE: Interrupt latency Jitter in A&M's AdderII (PPC MPC852) board
- From: "Paul D. DeRocco" <pderocco at ix dot netcom dot com>
- To: "eCos Discuss" <ecos-discuss at ecos dot sourceware dot org>
- Date: Wed, 2 Jun 2004 23:52:33 -0700
- Subject: RE: [ECOS] Interrupt latency Jitter in A&M's AdderII (PPC MPC852) board
> From: Tzahi [mailto:tzahi@terasync.net]
>
> * I'm sorry for not being too clear
> * I am using an on-board (The CPM's GPTIMER) accurate HW Timer
> that generate
> an Interrupt via the MPC852T and the eCos interrupt handling
> * I do not mind about the absolute size of the interrupt latency, but I
> would like it to be constant.
> * The actual code executed for the same event should run the same
> (the same
> branching is expected)
> * So what in the eCos interrupt handling may have caused this?
You were completely clear. The problem isn't in your interrupt handler, it's
that the main code consists of instructions that take different numbers of
clocks to execute. The latency from when the timer generates an interrupt to
when the first instruction of your interrupt handler executes depends upon
how many clocks it takes to complete whatever instruction is executing.
There's nothing you can do about that. If you need a hardware signal with no
jitter, then you'll have to generate it directly with hardware. I have no
idea what kind of hardware resources are available in the system you're
using, but a pulse generated by an interrupt handler will always have some
jitter in it.
--
Ciao, Paul D. DeRocco
Paul mailto:pderocco@ix.netcom.com
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss