Windows System Software -- Consulting, Training, Development -- Unique Expertise, Guaranteed Results
Before Posting...
Please check out the Community Guidelines in the Announcements and Administration Category.More Info on Driver Writing and Debugging
The free OSR Learning Library has more than 50 articles on a wide variety of topics about writing and debugging device drivers and Minifilters. From introductory level to advanced. All the articles have been recently reviewed and updated, and are written using the clear and definitive style you've come to expect from OSR over the years.
Check out The OSR Learning Library at: https://www.osr.com/osr-learning-library/
Re: A question about set 8259A interrupt controller.
OSR_Community_User
Member Posts: 110,217
in NTDEV
> I'm writing an ATM adapter driver. I want to ask two questions about set
>the 8259A Interrupt controller.
>
> 1.Does ddk supplies the function such as disable_irq to mask the 8259A
>interrupt controller? For example: in the linux, we can simply use the
>disable_irq(3) to disable interrupt 3.
On NT/Win2000 you don't directly deal with physical interrupt levels, you
work in terms of the logical IRQL level. Understanding IRQL's is critical
to understanding how to write NT/Win2000 drivers. To synchronize execution
with interrupt code, you use an appropriate function (like maybe
KeSynchronizeExecution) to raise your IRQL level, which may (or may not)
transparently adjust interrupt controller hardware . Also note that IRQL
levels are processor specific, so even if you raise the IRQL of one
processor, another processor may take an interrupt. This means you may need
a spinlock to protect shared data (KeSynchronizeExecution grabs the
interrupt object spinlock). Also, PCI devices sometimes share interrupt
levels, so physically disabling an interrupt levels might be bad for some
other device
As your device is a network device, and your probably writing an NDIS
driver, there is a special set of OS API's for doing all this (many of
these have names that start with Ndis...).
> 2.If ddk doesn't supply this kind of function, can I write the control
>word by using the WRITE_PORT_UCHAR( )?
Absolutely not, some systems don't use 8259A interrupt controllers (like
most multi processor capable systems ). The physical interrupt controller is owned by the OS, and is
only to be fooled with by it's owner.
- Jan
>the 8259A Interrupt controller.
>
> 1.Does ddk supplies the function such as disable_irq to mask the 8259A
>interrupt controller? For example: in the linux, we can simply use the
>disable_irq(3) to disable interrupt 3.
On NT/Win2000 you don't directly deal with physical interrupt levels, you
work in terms of the logical IRQL level. Understanding IRQL's is critical
to understanding how to write NT/Win2000 drivers. To synchronize execution
with interrupt code, you use an appropriate function (like maybe
KeSynchronizeExecution) to raise your IRQL level, which may (or may not)
transparently adjust interrupt controller hardware . Also note that IRQL
levels are processor specific, so even if you raise the IRQL of one
processor, another processor may take an interrupt. This means you may need
a spinlock to protect shared data (KeSynchronizeExecution grabs the
interrupt object spinlock). Also, PCI devices sometimes share interrupt
levels, so physically disabling an interrupt levels might be bad for some
other device
As your device is a network device, and your probably writing an NDIS
driver, there is a special set of OS API's for doing all this (many of
these have names that start with Ndis...).
> 2.If ddk doesn't supply this kind of function, can I write the control
>word by using the WRITE_PORT_UCHAR( )?
Absolutely not, some systems don't use 8259A interrupt controllers (like
most multi processor capable systems ). The physical interrupt controller is owned by the OS, and is
only to be fooled with by it's owner.
- Jan
Howdy, Stranger!
| Upcoming OSR Seminars | ||
|---|---|---|
| OSR has suspended in-person seminars due to the Covid-19 outbreak. But, don't miss your training! Attend via the internet instead! | ||
| Kernel Debugging | 13-17 May 2024 | Live, Online |
| Developing Minifilters | 1-5 Apr 2024 | Live, Online |
| Internals & Software Drivers | 11-15 Mar 2024 | Live, Online |
| Writing WDF Drivers | 26 Feb - 1 Mar 2024 | Live, Online |
Comments
Thank you for your timely answer.
I've now understood that I could't write control word to the PIC.
But the problem remains that our network adapter hasn't register which
can disable/enable itself producing interrupt . Instead, we have to control
the PIC to let PIC mask/unmask the interrupt.
So ,how can I disable/enable interrupt in my ISR function? Or does NT
automatically disable/enable interrupt when the drive enters the ISR
routine?
The same problem is with DMA.
So I wonder whether I should change our hardware design.
Thank you.
ÔÚ 00-2-23 12:54:00 ÄúдµÀ£º
>> I'm writing an ATM adapter driver. I want to ask two questions about set
>>the 8259A Interrupt controller.
>>
>> 1.Does ddk supplies the function such as disable_irq to mask the 8259A
>>interrupt controller? For example: in the linux, we can simply use the
>>disable_irq(3) to disable interrupt 3.
>
>On NT/Win2000 you don't directly deal with physical interrupt levels, you
>work in terms of the logical IRQL level. Understanding IRQL's is critical
>to understanding how to write NT/Win2000 drivers. To synchronize execution
>with interrupt code, you use an appropriate function (like maybe
>KeSynchronizeExecution) to raise your IRQL level, which may (or may not)
>transparently adjust interrupt controller hardware . Also note that IRQL
>levels are processor specific, so even if you raise the IRQL of one
>processor, another processor may take an interrupt. This means you may need
>a spinlock to protect shared data (KeSynchronizeExecution grabs the
>interrupt object spinlock). Also, PCI devices sometimes share interrupt
>levels, so physically disabling an interrupt levels might be bad for some
>other device
>
>As your device is a network device, and your probably writing an NDIS
>driver, there is a special set of OS API's for doing all this (many of
>these have names that start with Ndis...).
>
>- Jan
>
>
>
>---
>You are currently subscribed to ntdev as: [email protected]
>To unsubscribe send a blank email to $subst('Email.Unsub')
____________________________
Free Email/SMTP/POP, http://www.bn3.com, Hosting [email protected]
where you need to control a specific pc resource, you now need specific
support for each motherboard design that's slightly different. It shouldn't
be much of an issue to disable the interrupt at the adapter, and control
your DMA correctly.
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of encarta
> Sent: Thursday, February 24, 2000 9:24 AM
> To: NT Developers Interest List
> Subject: [ntdev] Re: A question about set 8259A interrupt controller.
>
>
> Hi,
>
> Thank you for your timely answer.
>
> I've now understood that I could't write control word to the PIC.
> But the problem remains that our network adapter hasn't register which
> can disable/enable itself producing interrupt . Instead, we have
> to control
> the PIC to let PIC mask/unmask the interrupt.
>
> So ,how can I disable/enable interrupt in my ISR function? Or does NT
> automatically disable/enable interrupt when the drive enters the ISR
> routine?
>
> The same problem is with DMA.
>
> So I wonder whether I should change our hardware design.
>
>
> Thank you.
>
>
>
>
> ÔÚ 00-2-23 12:54:00 ÄúдµÀ£º
> >> I'm writing an ATM adapter driver. I want to ask two
> questions about set
> >>the 8259A Interrupt controller.
> >>
> >> 1.Does ddk supplies the function such as disable_irq to
> mask the 8259A
> >>interrupt controller? For example: in the linux, we can
> simply use the
> >>disable_irq(3) to disable interrupt 3.
> >
> >On NT/Win2000 you don't directly deal with physical interrupt levels, you
> >work in terms of the logical IRQL level. Understanding IRQL's is critical
> >to understanding how to write NT/Win2000 drivers. To synchronize
> execution
> >with interrupt code, you use an appropriate function (like maybe
> >KeSynchronizeExecution) to raise your IRQL level, which may (or may not)
> >transparently adjust interrupt controller hardware . Also note that IRQL
> >levels are processor specific, so even if you raise the IRQL of one
> >processor, another processor may take an interrupt. This means
> you may need
> >a spinlock to protect shared data (KeSynchronizeExecution grabs the
> >interrupt object spinlock). Also, PCI devices sometimes share interrupt
> >levels, so physically disabling an interrupt levels might be bad for some
> >other device
> >
> >As your device is a network device, and your probably writing an NDIS
> >driver, there is a special set of OS API's for doing all this (many of
> >these have names that start with Ndis...).
> >
>
> >- Jan
> >
> >
> >
> >---
> >You are currently subscribed to ntdev as: [email protected]
> >To unsubscribe send a blank email to $subst('Email.Unsub')
>
>
>
> ____________________________
> Free Email/SMTP/POP, http://www.bn3.com, Hosting [email protected]
>
> ---
> You are currently subscribed to ntdev as: [email protected]
> To unsubscribe send a blank email to $subst('Email.Unsub')
>
> automatically disable/enable interrupt when the drive enters the ISR
> routine?
Yes. Outside the ISR, you can call KeSynchronizeExecution for this.
Max