Handling interrupt in vxworks... - VxWorks
This is a discussion on Handling interrupt in vxworks... - VxWorks ; Hi,
When hardware interrupt occurs who is responsible, i mean which task,
or say who informs scheduler to stop the current task execution and
execute the ISR. How this complete flow goes?
Regards,
Suresh...
-
Handling interrupt in vxworks...
Hi,
When hardware interrupt occurs who is responsible, i mean which task,
or say who informs scheduler to stop the current task execution and
execute the ISR. How this complete flow goes?
Regards,
Suresh
-
Re: Handling interrupt in vxworks...
On 20 Feb., 07:48, "Suresh" wrote:
> Hi,
>
> When hardware interrupt occurs who is responsible, i mean which task,
> or say who informs scheduler to stop the current task execution and
> execute the ISR. How this complete flow goes?
>
> Regards,
> Suresh
Basicly the function "intConnect" is responsible to tell the sheduler
the name of a function call responsible for a specified IRQ number.
This is the so called interrupt service routine. In this routine you
may execute your code. Hint: keep it short, signal a flag to task that
handles your interrupt code.
HTH, Christian
-
Re: Handling interrupt in vxworks...
On Feb 20, 3:21 pm, Christian.Doppelba...@gmail.com wrote:
> On 20 Feb., 07:48, "Suresh" wrote:
>
> > Hi,
>
> > When hardware interrupt occurs who is responsible, i mean which task,
> > or say who informs scheduler to stop the current task execution and
> > execute the ISR. How this complete flow goes?
>
> > Regards,
> > Suresh
>
> Basicly the function "intConnect" is responsible to tell the sheduler
> the name of a function call responsible for a specified IRQ number.
> This is the so called interrupt service routine. In this routine you
> may execute your code. Hint: keep it short, signal a flag to task that
> handles your interrupt code.
> HTH, Christian
Thanks Cristian but yes that i am clear how to register and enable the
interrupt but my question is what OS would do when it sense an
hardware pin interrupting?
-
Re: Handling interrupt in vxworks...
Hi,
You need to understand the inside progress of intConnect.
When hardware interrupt is invoked, kernel acknowledges it and connect
this interrupt signal to the function name specified on intConnect.
That'all in vxworks application source level.
This acts just like ISR.
cheol
-
Re: Handling interrupt in vxworks...
On Feb 21, 2:43 am, "Suresh" wrote:
> On Feb 20, 3:21 pm, Christian.Doppelba...@gmail.com wrote:
>
>
>
> > On 20 Feb., 07:48, "Suresh" wrote:
>
> > > Hi,
>
> > > When hardware interrupt occurs who is responsible, i mean which task,
> > > or say who informs scheduler to stop the current task execution and
> > > execute the ISR. How this complete flow goes?
>
> > > Regards,
> > > Suresh
>
> > Basicly the function "intConnect" is responsible to tell the sheduler
> > the name of a function call responsible for a specified IRQ number.
> > This is the so called interrupt service routine. In this routine you
> > may execute your code. Hint: keep it short, signal a flag to task that
> > handles your interrupt code.
> > HTH, Christian
>
> Thanks Cristian but yes that i am clear how to register and enable the
> interrupt but my question is what OS would do when it sense an
> hardware pin interrupting?
The OS doesn't sense the interrupt pin.
In general, microprocessors respond to interrupts by (after completing
the currently executing instruction) vectoring to a fixed or variable
location in the physical address space.
Some minimum context is automatically saved and execution picks up at
the exception vector location.
The exact details here are architecture dependent and vary quite a bit
between processors.
Once execution resumes at the exception vector, the OS supplied code
will typically save some additional context, enough to call a C
function without losing or corrupting any of the interrupted context,
and then call the installed C interrupt handler.
It's only at this point that "what the OS would do" is a question for
which there is a meaningful answer.
Before that, it's all handled by the CPU.
HTH,
GV
-
Re: Handling interrupt in vxworks...
These intrupts covers only the vectored interpts listed in you
processor manul. But if you are looking for memory corruption, you
want to catch the at hardware level who is writing, then you need a
hardware supported hardware debugger such as vision ICE etc.
On Feb 21, 5:09 am, "gvarndell" wrote:
> On Feb 21, 2:43 am, "Suresh" wrote:
>
>
>
>
>
> > On Feb 20, 3:21 pm, Christian.Doppelba...@gmail.com wrote:
>
> > > On 20 Feb., 07:48, "Suresh" wrote:
>
> > > > Hi,
>
> > > > When hardware interrupt occurs who is responsible, i mean which task,
> > > > or say who informs scheduler to stop the current task execution and
> > > > execute the ISR. How this complete flow goes?
>
> > > > Regards,
> > > > Suresh
>
> > > Basicly the function "intConnect" is responsible to tell the sheduler
> > > the name of a function call responsible for a specified IRQ number.
> > > This is the so called interrupt service routine. In this routine you
> > > may execute your code. Hint: keep it short, signal a flag to task that
> > > handles your interrupt code.
> > > HTH, Christian
>
> > Thanks Cristian but yes that i am clear how to register and enable the
> > interrupt but my question is what OS would do when it sense an
> > hardware pin interrupting?
>
> The OS doesn't sense the interrupt pin.
> In general, microprocessors respond to interrupts by (after completing
> the currently executing instruction) vectoring to a fixed or variable
> location in the physical address space.
> Some minimum context is automatically saved and execution picks up at
> the exception vector location.
> The exact details here are architecture dependent and vary quite a bit
> between processors.
>
> Once execution resumes at the exception vector, the OS supplied code
> will typically save some additional context, enough to call a C
> function without losing or corrupting any of the interrupted context,
> and then call the installed C interrupt handler.
> It's only at this point that "what the OS would do" is a question for
> which there is a meaningful answer.
> Before that, it's all handled by the CPU.
>
> HTH,
> GV- Hide quoted text -
>
> - Show quoted text -
-
Re: Handling interrupt in vxworks...
On Feb 22, 8:30 am, clax...@hotmail.com wrote:
> These intrupts covers only the vectored interpts listed in you
> processor manul. But if you are looking for memory corruption, you
> want to catch the at hardware level who is writing, then you need a
> hardware supported hardware debugger such as vision ICE etc.
>
> On Feb 21, 5:09 am, "gvarndell" wrote:
>
>
>
> > On Feb 21, 2:43 am, "Suresh" wrote:
>
> > > On Feb 20, 3:21 pm, Christian.Doppelba...@gmail.com wrote:
>
> > > > On 20 Feb., 07:48, "Suresh" wrote:
>
> > > > > Hi,
>
> > > > > When hardware interrupt occurs who is responsible, i mean which task,
> > > > > or say who informs scheduler to stop the current task execution and
> > > > > execute the ISR. How this complete flow goes?
>
> > > > > Regards,
> > > > > Suresh
>
> > > > Basicly the function "intConnect" is responsible to tell the sheduler
> > > > the name of a function call responsible for a specified IRQ number.
> > > > This is the so called interrupt service routine. In this routine you
> > > > may execute your code. Hint: keep it short, signal a flag to task that
> > > > handles your interrupt code.
> > > > HTH, Christian
>
> > > Thanks Cristian but yes that i am clear how to register and enable the
> > > interrupt but my question is what OS would do when it sense an
> > > hardware pin interrupting?
>
> > The OS doesn't sense the interrupt pin.
> > In general, microprocessors respond to interrupts by (after completing
> > the currently executing instruction) vectoring to a fixed or variable
> > location in the physical address space.
> > Some minimum context is automatically saved and execution picks up at
> > the exception vector location.
> > The exact details here are architecture dependent and vary quite a bit
> > between processors.
>
> > Once execution resumes at the exception vector, the OS supplied code
> > will typically save some additional context, enough to call a C
> > function without losing or corrupting any of the interrupted context,
> > and then call the installed C interrupt handler.
> > It's only at this point that "what the OS would do" is a question for
> > which there is a meaningful answer.
> > Before that, it's all handled by the CPU.
>
> > HTH,
> > GV- Hide quoted text -
>
> > - Show quoted text -- Hide quoted text -
>
> - Show quoted text -
Guys,
Thanks a lot for valuable inputs for making me more clear.
Thanks again.
Regards,
Suresh