Re: PREP IDE support is broken

Re: PREP IDE support is broken

[Matt Porter]

On Thu, 8 Jul 1999, Eric Vallete wrote:

> Do anyone used PREP boards using IDE disk!!! I do.
> Months ago soemone from Moto posted a mail fixing IDE
> support. With the introduction of machine dependendent
> function pointers for ide byte handling, someone broke 
> it again.

Yes, we are using CF IDE disks on an MCP750.  Right now we just have some
hacks for supporting the edge/level changes and are working on some clean
code.

I was just looking yesterday at what is the best way to clean up
prep_pci.c so that we can have a machine type available elsewhere as you
mention.  I want to move the machine detection code so it's running
earlier and just have a function to call for the PReP machine type.
Ultimately, I'd like to see board-specific items isolated in separate
files for a more modular architecture.

--
Matt Porter                                       Motorola Computer Group
matt_porter@mcg.mot.com                           CIBU Linux Support


[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]

PowerPC assembly question

[Charles Lepple]

Hope this isn't too far off topic.

I've been poking around with some disassembled C code, and I'm wondering
if there is a source of documentation on register usage guidelines
(maybe an ABI? I'm just not used to this concept of passing arguments in
registers as opposed to the stack). Any help would be appreciated.

--Charles Lepple
clepple@mitre.org

[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]

Re: PowerPC assembly question

[Josh Huber]

Charles Lepple writes:
> 
> Hope this isn't too far off topic.
> 
> I've been poking around with some disassembled C code, and I'm wondering
> if there is a source of documentation on register usage guidelines
> (maybe an ABI? I'm just not used to this concept of passing arguments in
> registers as opposed to the stack). Any help would be appreciated.
What you're looking for is linked from here:
http://www.dartmouth.edu/cgi-bin/cgiwrap/jonh/lppc/faq.pl?file=155

Josh

[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]

Re: PowerPC assembly question

[David Edelsohn]

>>>>> Charles Lepple writes:

Charles> I've been poking around with some disassembled C code, and I'm wondering
Charles> if there is a source of documentation on register usage guidelines
Charles> (maybe an ABI? I'm just not used to this concept of passing arguments in
Charles> registers as opposed to the stack). Any help would be appreciated.

	All of this is defined in the appropriate ABI (AIX PowerOpen and
SVR4 PowerPC ABI).  The ABI specifies the uses of each register when
targeting code to the calling convention (argument passing, volatile and
non-volatile registers, etc.)

David


[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]

Re: PREP IDE support is broken

[VALETTE Eric]

>>>>> "Matt" == Matt Porter <mporter@phx.mcd.mot.com> writes:

Matt> On Thu, 8 Jul 1999, Eric Vallete wrote:

Matt> Yes, we are using CF IDE disks on an MCP750.  Right now we just have some
Matt> hacks for supporting the edge/level changes and are working on some clean
Matt> code.

I have code that works quite well. Just the way calling it is bad...

Matt> I was just looking yesterday at what is the best way to clean up
Matt> prep_pci.c so that we can have a machine type available elsewhere as you
Matt> mention.  I want to move the machine detection code so it's running
Matt> earlier and just have a function to call for the PReP machine type.
Matt> Ultimately, I'd like to see board-specific items isolated in separate
Matt> files for a more modular architecture.

Thanks a lot. Of course I can send a patch but I've seen so many wasted...

  ----------------------------------- prep pci ----------------------
int mesquite __initdata = 0;
struct pci_dev *via_82c586_pci_isa_bridge;

static unsigned char mcp750_openpic_initsenses[] = {
    1,	/* MCP750_INT_PCB(8259) */
    0,	/* MCP750_INT_FALCON_ECC_ERR */
    1,	/* MCP750_INT_PCI_ETHERNET */
    1,	/* MCP750_INT_PCI_PMC */
    1,	/* MCP750_INT_PCI_WATCHDOG_TIMER1 */
    1,	/* MCP750_INT_PCI_PRST_SIGNAL */
    1,	/* MCP750_INT_PCI_FALL_SIGNAL */
    1,	/* MCP750_INT_PCI_DEG_SIGNAL */
    1,	/* MCP750_INT_PCI_BUS1_INTA */
    1,	/* MCP750_INT_PCI_BUS1_INTB */
    1,	/* MCP750_INT_PCI_BUS1_INTC */
    1,	/* MCP750_INT_PCI_BUS1_INTD */
    1,	/* MCP750_INT_PCI_BUS2_INTA */
    1,	/* MCP750_INT_PCI_BUS2_INTB */
    1,	/* MCP750_INT_PCI_BUS2_INTC */
    1,	/* MCP750_INT_PCI_BUS2_INTD */
};

__initfunc(int raven_init(void))
{
	unsigned int	devid;
	unsigned int	pci_membase;
	unsigned char	base_mod;

	/* Check to see if the Raven chip exists. */
	if ( _prep_type != _PREP_Motorola) {
		OpenPIC = NULL;
		return 0;
	}

	/* Check to see if this board is a type that might have a Raven. */
	if ((inb(MOTOROLA_CPUTYPE_REG) & 0xF0) != 0xE0) {
		OpenPIC = NULL;
		return 0;
	}

	/* Check the first PCI device to see if it is a Raven. */
	pcibios_read_config_dword(0, 0, PCI_VENDOR_ID, &devid);

	switch (devid & 0xffff0000) {
	case MPIC_RAVEN_ID:
		MotMPIC = MOT_RAVEN_PRESENT;
		break;
	case MPIC_HAWK_ID:
		MotMPIC = MOT_HAWK_PRESENT;
		break;
	default:
		OpenPIC = NULL;
		return 0;
	}


	/* Read the memory base register. */
	pcibios_read_config_dword(0, 0, PCI_BASE_ADDRESS_1, &pci_membase);

	if (pci_membase == 0) {
		OpenPIC = NULL;
		return 0;
	}

	/* Map the Raven MPIC registers to virtual memory. */
	OpenPIC = (struct OpenPIC *)ioremap(pci_membase+0xC0000000, 0x22000);

	if (mesquite) {
	  OpenPIC_InitSenses = mcp750_openpic_initsenses;
	  OpenPIC_NumInitSenses = sizeof(mcp750_openpic_initsenses);
	}
	else {
	  OpenPIC_InitSenses = mvme2600_openpic_initsenses;
	  OpenPIC_NumInitSenses = sizeof(mvme2600_openpic_initsenses);
	}

	ppc_md.do_IRQ = chrp_do_IRQ;
	
	/* If raven is present on Motorola store the system config register
	 * for later use.
	 */
	ProcInfo = (unsigned long *)ioremap(0xfef80400, 4);

	/* This is a hack.  If this is a 2300 or 2400 mot board then there is
	 * no keyboard controller and we have to indicate that.
	 */
	base_mod = inb(MOTOROLA_BASETYPE_REG);
	if ((MotMPIC == MOT_HAWK_PRESENT) || (base_mod == 0xF9) ||
	    (base_mod == 0xFA) || (base_mod == 0xE1))
		prep_keybd_present = 0;

	if (mesquite) {
	  struct pci_dev *dev;
	  unsigned int temp;
	  unsigned char tmp;
	  
	  for(dev=pci_devices; dev; dev=dev->next) {
	    pcibios_read_config_dword(dev->bus->number, dev->devfn, PCI_VENDOR_ID, &temp);
	     if ( (temp == (((unsigned short) PCI_VENDOR_ID_VIA) | (PCI_DEVICE_ID_VIA_82C586_1 << 16)))
		  ||
		  (temp == (((unsigned short) PCI_VENDOR_ID_VIA) | (PCI_DEVICE_ID_VIA_82C586_0 << 16)))
		  ) {
	       via_82c586_pci_isa_bridge = dev;
	       break;
	     }
	  }
	  tmp = inb(0x810);
	  if  ( !(tmp & 0x2)) {
	    printk("This is a second generation MCP750 board\n");
	    printk("We must reprogram the PCI/ISA bridge...\n");
	    pcibios_read_config_byte(via_82c586_pci_isa_bridge->bus->number,
				     via_82c586_pci_isa_bridge->devfn,
				     0x47,  &tmp);
	    printk(" PCI ISA bridge control2 = %x\n", (unsigned) tmp);
	    /*
	     * Enable 4D0/4D1 ISA interrupt level/edge config registers
	     */
	    tmp |= 0x20;
	    pcibios_write_config_byte(via_82c586_pci_isa_bridge->bus->number,
				     via_82c586_pci_isa_bridge->devfn,
				     0x47,  tmp);
	    /*
	     * Now program the ISA interrupt edge/level
	     */
	    tmp = ELCRS_INT9_LVL | ELCRS_INT10_LVL | ELCRS_INT11_LVL;
	    outb(tmp, ISA8259_S_ELCR);
	    tmp = ELCRM_INT5_LVL;
	    outb(tmp, ISA8259_M_ELCR);;
	    /*
	     * Set the Interrupt inputs to non-inverting level interrupt
	     */
	    pcibios_read_config_byte(via_82c586_pci_isa_bridge->bus->number,
				     via_82c586_pci_isa_bridge->devfn,
				     0x54, &tmp);
	    printk(" PCI ISA bridge PCI/IRQ Edge/Level Select = %x\n", (unsigned) tmp);
	    tmp = 0;
	    pcibios_write_config_byte(via_82c586_pci_isa_bridge->bus->number,
				     via_82c586_pci_isa_bridge->devfn,
				     0x54, tmp);
	  }

	}
	return 1;
}
  -------------------------------------------------------------------
-- eric

[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]

Re: PowerPC assembly question

[Charles A. Jolley]

There are two main ABI's that I have consulted in the past regarding how 
Linux handles linking, etc. on the PowerPC.  The primary one is the
PowerOpen ABI, which describes many of the general register usage
conventions.  The other is the "System V ABI, PowerPC Processor Supplement",
published by SunSoft.  This primarily describes how PowerPC images are held
in ELF files.  I have both of these plus some additional helper docs in PDF
format which I use for reference here, but I know they are also available on
the web somewhere and there are probably some more up-to-date versions.
(Mine were released in 1994).

I can send them if needed...the copyrights don't prohibit redistribution...
however I seem to remember getting all of these from either IBM or
Motorola's PowerPC site so you may want to check there for newer versions
first.

Cheers,
-Charles

----------
>From: Charles Lepple <clepple@mitre.org>
>To: linuxppc-dev@lists.linuxppc.org
>Subject: PowerPC assembly question
>Date: Thu, Jul 8, 1999, 12:39 PM
>

>
> Hope this isn't too far off topic.
>
> I've been poking around with some disassembled C code, and I'm wondering
> if there is a source of documentation on register usage guidelines
> (maybe an ABI? I'm just not used to this concept of passing arguments in
> registers as opposed to the stack). Any help would be appreciated.
>
> --Charles Lepple
> clepple@mitre.org
> 

[[ This message was sent via the linuxppc-dev mailing list.  Replies are ]]
[[ not  forced  back  to the list, so be sure to Cc linuxppc-dev if your ]]
[[ reply is of general interest. Please check http://lists.linuxppc.org/ ]]
[[ and http://www.linuxppc.org/ for useful information before posting.   ]]