No need to maintain statistics twice.

On arrival of a new message, kernelcapi used to call
capi20_appl::signal(), which, from the application, would call back
to capi20_get_message(). So we rather just push the message down
directly, saving this detour.

Instead of passing an opaque handle, pass the actual struct capi20_appl,
which is now known to the applications. Applications can store
a pointer to the private data into struct capi20_appl::private.

Now that we have the struct capi_appl *, let's pass this around
instead of just an index which would mean another useless lookup.

Inside the kernel, we rather go the standard way than converting
to/from indices to the actual data structs all the time.

It's not used anymore, registering/unregistering just happens
on a per-application basis.

The callback to be notified of added/removed controllers is
logically per application.

This will be replaced by the standardized mechanism anyway, so the
temporary capi20_set_callback will fortunately vanish later.

Why pass the callbacks via a struct when we can just call them directly?

During module init, we're safe from getting unloaded, so there's no
need to fiddle with the use count.

into home.transmeta.com:/home/torvalds/v2.5/linux

into tp1.ruhr-uni-bochum.de:/home/kai/kernel/v2.5/linux-2.5.make-as

This patch updates the orinoco wireless driver to 0.11b.  This fixes
several kfree()-of-bad-address bugs.

into home.transmeta.com:/home/torvalds/v2.5/linux

into tp1.ruhr-uni-bochum.de:/home/kai/kernel/v2.5/linux-2.5.isdn

into home.transmeta.com:/home/torvalds/v2.5/linux

into flint.arm.linux.org.uk:/usr/src/linux-bk-2.5/linux-2.5-rmk

Fixes lockup on SMP boxes, and fixes buggy map scanning code.

we fix ARM720T support - this CPU has unified writethrough caches
only, so we can't use the Harvard cache operations when copying
pages.  Also, we don't have to evict cache entries during copypage.

Use the standard new-style pci_driver::remove to unregister
drivers at module unload time.

PCMCIA obviously has its own way of handling removing, the ISA
drivers unregister at module unload time as well.

Don't use a special CAPI solution to tell the drivers about 
ISA cards but use module parameters, just as other drivers do.

Internally use struct pci_dev to save that data - hopefully
one day the device tree will provide a nicer way to achieve this.

Use common function for setting the revision strings.

We now control rely on ->owner and the upper level to control the
module use count.

into tp1.ruhr-uni-bochum.de:/home/kai/kernel/v2.5/linux-2.5.isdn

 - Rewrite ioctl handling.

 - Apply fix for hpt366 "hang on boot" by Andre.

 - Remove stale XXX_tune_req. It was no longer used.

 - Propagate rq through ide_error(), ide_end_drive_cmd(), ide_dump_status(),
   ide_wait_stat().

 - Push the current drive down to ata_channel from hwgroup.

 - Push the timer down to the ata_channel structure. Most probably it will end
   at the drive.

Any potential BE arm port can fix this up later.

into home.transmeta.com:/home/torvalds/v2.5/linux

into home.transmeta.com:/home/torvalds/v2.5/linux

drivers.  Most of this is from 2.4.  Almost all of this work was
done by Ben Herrenschmidt.

Ok, here it is.  The following arch are not covered:
Mips, Mips64 in 32-bit mode, parisc in __LP64__ mode.

In addition, x86_64 mentions jiffies in the existing script.
This may be a problem.

This prevents a deadlock in JFS.

into penguin.transmeta.com:/home/penguin/torvalds/repositories/kernel/linux

into tp1.ruhr-uni-bochum.de:/home/kai/kernel/v2.5/linux-2.5.isdn.capi

into penguin.transmeta.com:/home/penguin/torvalds/repositories/kernel/linux

Noted by Keith Owens.

This removes the old alternates to export_operations for exporting a
filesystem.

It removes fh_to_dentry, dentry_to_fh, and s_nfsd_free_path_sem.  It
also removes a lot of code.

The fs/ntfs change is because it was setting fh_to_dentry and
dentry_to_fh (which no longer exist) to NULL.