This patch series converts the PNP resource option structures
to a unified linked list. This preserves resource order, which
is important for some devices. The bulk of the series is in
the last patch, and it has more a more detailed changelog.
Rene Herman did some testing of ISA devices and reviewed
v1 of this series. I think I addressed all his comments.
Any comments would be welcome.
This depends on some patches that are in -mm, but not yet
upstream. In mmotm, these would probably go after
pnp-dont-sort-by-type-in-sys-resources.patch
Note that I incorporated Rene's ioport.h comment fix
(pnpacpi-fix-irq-flag-decoding-comment-fix.patch) into
this series, so you should drop that if you pick up this
series.
Change between v1 and v2:
- rewrote 8250_pnp modem identification (first patch in series)
- used bitmap_copy() rather than explicit loop to copy IRQ bitmaps
- added "(optional)" note in /sys/.../options for IORESOURCE_IRQ_OPTIONAL
- split out ISAPNP ENDDEP bug fix into a separate patch
- restored "clear auto resources" when auto-assignment fails
- removed 0x100 from option priority
- allocated 12 bits for dependent option priority, 16 bits for set number
- restored check for invalid dependent function priority
- fixed AD1815 quirk (set OPTIONAL flag in the correct "flags" field)
- fixed ioport.h ISA PnP comment (replaces Rene's
pnpacpi-fix-irq-flag-decoding-comment-fix.patch)
- rename "pnp_dependent_option()" to "pnp_new_dependent_set()"
Bjorn
--
On 05-06-08 00:09, Bjorn Helgaas wrote:
> This patch series converts the PNP resource option structures
> to a unified linked list. This preserves resource order, which
> is important for some devices. The bulk of the series is in
> the last patch, and it has more a more detailed changelog.
>
> Rene Herman did some testing of ISA devices and reviewed
> v1 of this series. I think I addressed all his comments.
> Any comments would be welcome.
>
> This depends on some patches that are in -mm, but not yet
> upstream. In mmotm, these would probably go after
> pnp-dont-sort-by-type-in-sys-resources.patch
Perhaps it would've been different if I'd been a full-blown git wizzard
but this is getting rather painful to test against mainline. Would it be
possible for you to repost (privately or publicly) that which this
depends on against current mainline?
I had to blow away my local trees here and while rebuilding the needed
manual adjustments were just just enough to make me worry that I'd end
up with something not entirely correct.
Rene.
On Monday 09 June 2008 09:00:13 pm Rene Herman wrote:
> On 05-06-08 00:09, Bjorn Helgaas wrote:
>
> > This patch series converts the PNP resource option structures
> > to a unified linked list. This preserves resource order, which
> > is important for some devices. The bulk of the series is in
> > the last patch, and it has more a more detailed changelog.
> >
> > Rene Herman did some testing of ISA devices and reviewed
> > v1 of this series. I think I addressed all his comments.
> > Any comments would be welcome.
> >
> > This depends on some patches that are in -mm, but not yet
> > upstream. In mmotm, these would probably go after
> > pnp-dont-sort-by-type-in-sys-resources.patch
>
> Perhaps it would've been different if I'd been a full-blown git wizzard
> but this is getting rather painful to test against mainline. Would it be
> possible for you to repost (privately or publicly) that which this
> depends on against current mainline?
Sure. Here's my current quilt "patches" directory:
http://www.kernel.org/pub/linux/kernel/people/helgaas/2.6.26-rc5-pnp1.tar.bz2