iounmap

iounmap

[Ronald Wahl]

Hi,

I'm currently writing a device driver for an pci controller and have to
map very large amounts of pci memory and i/o ranges via ioremap. The
problem is that there is no iounmap and ioremap returns NULL after some
insmod/rmmod cycles. When gets this implemented?

ron

--
\\                 Ronald Wahl   |   Peppercon AG                 //
 \\\          rwa@peppercon.de  |||  http://www.peppercon.de/   ///
  \OO  -----------------------  OOO  ------------------------  OO/
   OO  GnuPG/PGP key available  OOO  ||||||||||||||||||||||||  OO


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: iounmap

[Geert Uytterhoeven]

On Thu, 3 Aug 2000, Ronald Wahl wrote:
> I'm currently writing a device driver for an pci controller and have to
> map very large amounts of pci memory and i/o ranges via ioremap. The
> problem is that there is no iounmap and ioremap returns NULL after some
> insmod/rmmod cycles. When gets this implemented?

When someone steals some spare time and writes the code.

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


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: iounmap

[Matt Porter]

On Thu, Aug 03, 2000 at 02:44:12PM +0200, Geert Uytterhoeven wrote:
>
> On Thu, 3 Aug 2000, Ronald Wahl wrote:
> > I'm currently writing a device driver for an pci controller and have to
> > map very large amounts of pci memory and i/o ranges via ioremap. The
> > problem is that there is no iounmap and ioremap returns NULL after some
> > insmod/rmmod cycles. When gets this implemented?
>
> When someone steals some spare time and writes the code.

Done.  The hot swap PCI kernel patches from MCG have iounmap implemented.
Very useful when ripping a PCI peripheral out of its slot. :)

See them at http://www.mcg.mot.com/linux under HA Linux/Downloads.

--
Matt Porter
MontaVista Software, Inc.
mporter@mvista.com

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: iounmap

[Ronald Wahl]

On Thu, 3 Aug 2000, Matt Porter wrote:

> On Thu, Aug 03, 2000 at 02:44:12PM +0200, Geert Uytterhoeven wrote:
> >
> > On Thu, 3 Aug 2000, Ronald Wahl wrote:
> > > I'm currently writing a device driver for an pci controller and have to
> > > map very large amounts of pci memory and i/o ranges via ioremap. The
> > > problem is that there is no iounmap and ioremap returns NULL after some
> > > insmod/rmmod cycles. When gets this implemented?
> >
> > When someone steals some spare time and writes the code.
>
> Done.  The hot swap PCI kernel patches from MCG have iounmap implemented.
> Very useful when ripping a PCI peripheral out of its slot. :)
>
> See them at http://www.mcg.mot.com/linux under HA Linux/Downloads.

Call me blind, but I havn't seen an implementation of iounmap in these
patches.

ron


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: iounmap

[Matt Porter]

On Fri, Aug 04, 2000 at 10:06:26AM +0200, Ronald Wahl wrote:
> On Thu, 3 Aug 2000, Matt Porter wrote:
>
> > On Thu, Aug 03, 2000 at 02:44:12PM +0200, Geert Uytterhoeven wrote:
> > >
> > > On Thu, 3 Aug 2000, Ronald Wahl wrote:
> > > > I'm currently writing a device driver for an pci controller and have to
> > > > map very large amounts of pci memory and i/o ranges via ioremap. The
> > > > problem is that there is no iounmap and ioremap returns NULL after some
> > > > insmod/rmmod cycles. When gets this implemented?
> > >
> > > When someone steals some spare time and writes the code.
> >
> > Done.  The hot swap PCI kernel patches from MCG have iounmap implemented.
> > Very useful when ripping a PCI peripheral out of its slot. :)
> >
> > See them at http://www.mcg.mot.com/linux under HA Linux/Downloads.
>
> Call me blind, but I havn't seen an implementation of iounmap in these
> patches.

Whoops, sorry.  When I was at MCG I know another engineer did an iounmap
implementation for HA support.  I thought it was in those patches... :-/

--
Matt Porter
MontaVista Software, Inc.
mporter@mvista.com

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: iounmap

[Paul Mackerras]

Ronald Wahl writes:

> I'm currently writing a device driver for an pci controller and have to
> map very large amounts of pci memory and i/o ranges via ioremap. The
> problem is that there is no iounmap and ioremap returns NULL after some
> insmod/rmmod cycles. When gets this implemented?

This should do the trick:

void iounmap(void *addr)
{
	if (addr > high_memory && (unsigned long) addr < ioremap_bot)
		vfree((void *) (PAGE_MASK & (unsigned long) addr));
}

Could you replace the null iounmap in arch/ppc/mm/init.c with that,
test it, and let me know if it works?

Paul.

--
Paul Mackerras, Senior Open Source Researcher, Linuxcare, Inc.
+61 2 6262 8990 tel, +61 2 6262 8991 fax
paulus@linuxcare.com.au, http://www.linuxcare.com.au/
Linuxcare.  Support for the revolution.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: iounmap

[Ronald Wahl]

On Fri, 4 Aug 2000, Paul Mackerras wrote:

> Ronald Wahl writes:
>
> > I'm currently writing a device driver for an pci controller and have to
> > map very large amounts of pci memory and i/o ranges via ioremap. The
> > problem is that there is no iounmap and ioremap returns NULL after some
> > insmod/rmmod cycles. When gets this implemented?
>
> This should do the trick:
>
> void iounmap(void *addr)
> {
> 	if (addr > high_memory && (unsigned long) addr < ioremap_bot)
> 		vfree((void *) (PAGE_MASK & (unsigned long) addr));
> }
>
> Could you replace the null iounmap in arch/ppc/mm/init.c with that,
> test it, and let me know if it works?

Yes, it seems to work. Thanx!

ron


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/