[PATCH 2.4] ndelay typo?

[PATCH 2.4] ndelay typo?

[Brian Murphy]

Hi Ralf,
	I presume you meant this?

/Brian

Index: include/asm/delay.h
===================================================================
RCS file: /cvs/linux/include/asm-mips/delay.h,v
retrieving revision 1.10.2.4
diff -u -r1.10.2.4 delay.h
--- include/asm/delay.h	5 Jul 2003 03:23:46 -0000	1.10.2.4
+++ include/asm/delay.h	5 Jul 2003 13:18:24 -0000
@@ -73,6 +73,6 @@
 #endif
 
 #define udelay(usecs) __udelay((usecs),__udelay_val)
-#define ndelay(usecs) __udelay((usecs),__udelay_val)
+#define ndelay(usecs) __ndelay((usecs),__udelay_val)
 
 #endif /* _ASM_DELAY_H */

Re: [PATCH 2.4] ndelay typo?

[Ralf Baechle]

On Sat, Jul 05, 2003 at 03:22:22PM +0200, Brian Murphy wrote:

> 	I presume you meant this?

Yes, thanks, will fix.

> Index: include/asm/delay.h
> ===================================================================
> RCS file: /cvs/linux/include/asm-mips/delay.h,v
> retrieving revision 1.10.2.4
> diff -u -r1.10.2.4 delay.h
> --- include/asm/delay.h	5 Jul 2003 03:23:46 -0000	1.10.2.4

Btw, don't diff anything in include/asm/.  That's a symlink ...

Btw, sorry if I toasted the Lasat support with the Galileo cleanup.  There
was just so much code duplication, it was insane to maintain.

I'm wondering about the Nile4 support btw.   Vrc5074 == NILE4, right?

   Ralf

Re: [PATCH 2.4] ndelay typo?

[Geert Uytterhoeven]

On Sat, 5 Jul 2003, Ralf Baechle wrote:
> On Sat, Jul 05, 2003 at 03:22:22PM +0200, Brian Murphy wrote:
> > 	I presume you meant this?
> 
> Yes, thanks, will fix.

And don't you want to rename the `usecs' parameter of ndelay() to `nsecs'?

> I'm wondering about the Nile4 support btw.   Vrc5074 == NILE4, right?

Yep.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: [PATCH 2.4] ndelay typo?

[Ralf Baechle]

On Sat, Jul 05, 2003 at 11:16:49PM +0200, Geert Uytterhoeven wrote:

> And don't you want to rename the `usecs' parameter of ndelay() to `nsecs'?

You've not looked at what I actually checked in, I renamed the argument.

> > I'm wondering about the Nile4 support btw.   Vrc5074 == NILE4, right?
> 
> Yep.

Well, I was wondering because the code in arch/mips/pci/ops-nile4.c which
was extraced from the lasat code is completly different from
ddb5xxx/ddb5074/pci_ops.c, so it's hard to extract the commonc code into
a shared file.

  Ralf

Nile 4 (was: Re: [PATCH 2.4] ndelay typo?)

[Geert Uytterhoeven]

On Sun, 6 Jul 2003, Ralf Baechle wrote:
> On Sat, Jul 05, 2003 at 11:16:49PM +0200, Geert Uytterhoeven wrote:
> > > I'm wondering about the Nile4 support btw.   Vrc5074 == NILE4, right?
> > 
> > Yep.
> 
> Well, I was wondering because the code in arch/mips/pci/ops-nile4.c which
> was extraced from the lasat code is completly different from
> ddb5xxx/ddb5074/pci_ops.c, so it's hard to extract the commonc code into
> a shared file.

If you know the chip, they are actually quite similar :-)

The differences between the Lasat and the DDB code are these:
  - The Lasat code checks the PCI error registers to detect the presence of PCI
    devices, while the DDB code doesn't,
  - The Lasat code is limited to 8 PCI devices on bus 0, while the DDB code
    uses a different access scheme to access the extra devices,
  - The DDB code uses abstraction functions to access the Nile 4 registers,
    while the Lasat code accesses the registers directly.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: Nile 4

[Ralf Baechle]

On Wed, Jul 09, 2003 at 09:58:08AM +0200, Geert Uytterhoeven wrote:

> > Well, I was wondering because the code in arch/mips/pci/ops-nile4.c which
> > was extraced from the lasat code is completly different from
> > ddb5xxx/ddb5074/pci_ops.c, so it's hard to extract the commonc code into
> > a shared file.
> 
> If you know the chip, they are actually quite similar :-)
> 
> The differences between the Lasat and the DDB code are these:
>   - The Lasat code checks the PCI error registers to detect the presence of PCI
>     devices, while the DDB code doesn't,
>   - The Lasat code is limited to 8 PCI devices on bus 0, while the DDB code
>     uses a different access scheme to access the extra devices,
>   - The DDB code uses abstraction functions to access the Nile 4 registers,
>     while the Lasat code accesses the registers directly.

Time to cleanup that mess also then.

Does anybody still care about the DDB5074?  I was just told somebody tried
it and it didn't boot into userspace ...

  Ralf

Re: Nile 4

[Geert Uytterhoeven]

On Thu, 10 Jul 2003, Ralf Baechle wrote:
> Does anybody still care about the DDB5074?  I was just told somebody tried
> it and it didn't boot into userspace ...

A few weeks ago I gave it a try (2.4.21-pre4), but I didn't bother to set up a
Debian root on NFS to check userspace. At least the kernel booted until IP
autoconfig.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds