misc questions about the device&driver arch

[John-Mark Gurney]

william wallace wrote this message on Wed, May 24, 2006 at 21:13 +0800:
> roger really good tips :)
> and so far as i know ,
> the PCI configure space is accessed by  pci_read_config,and so on
> the PCI IO           space is accessed by
bus_space_read_1 and friends...

> the PCI memory  space is accessed by   bus_space_read_1 and so on
> am i right sir?

Yep...

> On 5/24/06, John-Mark Gurney <gurney_j at resnet.uoregon.edu> wrote:
> >william wallace wrote this message on Wed, May 24, 2006 at 15:48 +0800:
> >> IN static device_method_t pci_methods[] = {
> >> what is the freeBSD's magic to connect pci_read_config_method    wtih
> >> PCI_READ_CONFIG? awk script?and so on?
> >> which
> >> DEVMETHOD(pci_read_config,    pci_read_config_method), in pci.c
> >> command = PCI_READ_CONFIG(dev, child, PCIR_COMMAND, 2); in
> >> pci_enable_io_method in pci.c
> >>
> >> and what if i create a new method in pci method interface ?
> >> such as   DEVMETHOD(pci_have_rest ,   pci_have_rest_method)
> >> what else should i do when i want to use PCI_HAVE_REST(dev ) point to
> >> pci_have_rest_method
> >
> >These are handled by the _if.m files...  look at sys/dev/pci/pci_if.m...
> >There is an awk script in sys/tools/makeobjops.awk that converts the
> >_if.m to the appropriate .h and .c files...
> >
> >Please read:
> >http://www.freebsd.org/doc/en_US.ISO8859-1/books/arch-handbook/kernel-objects-using.html
> >http://www.freebsd.org/doc/en_US.ISO8859-1/books/arch-handbook/newbus-overview.html
> >
> >The device methods are implemented using kobj...
> >
> >> On 5/23/06, Scott Long <scottl at samsco.org> wrote:
> >> >william wallace wrote:
> >> >
> >> >[...]
> >> >
> >> >> MSI:
> >> >> I've bantered around different suggestions for an API that will 
> >support
> >> >> this.  The basic thing that a driver needs from this is to know
> >> >> exactly how many message interrupt vectors are available to it.  It
> >> >> can't just register vectors and handlers blindly since the purpose of
> >> >> MSI is to assign special meanings to each vector and allow the driver 
> >to
> >> >> handle each one in specifically.
> >> >
> >> >[...]
> >> >
> >> >I just wanted to briefly say that an MSI implementation has been done
> >> >recently, and that it should start getting wider circulation and review
> >> >soon.  That's not to say that more work and design can't be done in this
> >> >area, but we should probably wait a bit and see what has been done
> >> >already.

-- 
  John-Mark Gurney				Voice: +1 415 225 5579

     "All that I will do, has been done, All that I have, has not."