linux.git/drivers/uio, branch v2.6.29

UIO: Pass information about ioports to userspace (V2)

2009-01-06T18:44:44+00:00

Devices sometimes have memory where all or parts of it can not be mapped to
userspace. But it might still be possible to access this memory from
userspace by other means. An example are PCI cards that advertise not only
mappable memory but also ioport ranges. On x86 architectures, these can be
accessed with ioperm, iopl, inb, outb, and friends. Mike Frysinger (CCed)
reported a similar problem on Blackfin arch where it doesn't seem to be easy
to mmap non-cached memory but it can still be accessed from userspace.

This patch allows kernel drivers to pass information about such ports to
userspace. Similar to the existing mem[] array, it adds a port[] array to
struct uio_info. Each port range is described by start, size, and porttype.

If a driver fills in at least one such port range, the UIO core will simply
pass this information to userspace by creating a new directory "portio"
underneath /sys/class/uio/uioN/. Similar to the "mem" directory, it will
contain a subdirectory (portX) for each port range given.

Note that UIO simply passes this information to userspace, it performs no
action whatsoever with this data. It's userspace's responsibility to obtain
access to these ports and to solve arch dependent issues. The "porttype"
attribute tells userspace what kind of port it is dealing with.

This mechanism could also be used to give userspace information about GPIOs
related to a device. You frequently find such hardware in embedded devices,
so I added a UIO_PORT_GPIO definition. I'm not really sure if this is a good
idea since there are other solutions to this problem, but it won't hurt much
anyway.

Signed-off-by: Hans J. Koch 
Signed-off-by: Greg Kroah-Hartman

UIO: uio_pdrv_genirq: allow custom irq_flags

2009-01-06T18:44:43+00:00

I can't think of a reason why the driver prevents people from setting any
custom bits in their platform device, but I can think of some reasons for
allowing custom flags.  Like setting the IRQF_TRIGGER_... bits.

Signed-off-by: Mike Frysinger 
Signed-off-by: Hans J. Koch 
Signed-off-by: Greg Kroah-Hartman

UIO: use pci_ioremap_bar() in drivers/uio

2009-01-06T18:44:43+00:00

Use the newly introduced pci_ioremap_bar() function in drivers/uio.
pci_ioremap_bar() just takes a pci device and a bar number, with the goal
of making it really hard to get wrong, while also having a central place
to stick sanity checks.

Signed-off-by: Arjan van de Ven 
Signed-off-by: Hans J. Koch 
Signed-off-by: Greg Kroah-Hartman

saner FASYNC handling on file close

2008-11-01T16:49:46+00:00

As it is, all instances of ->release() for files that have ->fasync()
need to remember to evict file from fasync lists; forgetting that
creates a hole and we actually have a bunch that *does* forget.

So let's keep our lives simple - let __fput() check FASYNC in
file->f_flags and call ->fasync() there if it's been set.  And lose that
crap in ->release() instances - leaving it there is still valid, but we
don't have to bother anymore.

Signed-off-by: Al Viro 
Signed-off-by: Linus Torvalds

Merge branch 'bkl-removal' of git://git.lwn.net/linux-2.6

2008-10-20T20:42:14+00:00

* 'bkl-removal' of git://git.lwn.net/linux-2.6:
  UIO: BKL removal

UIO: BKL removal

2008-10-16T18:32:26+00:00

Fill in needed locking around idr accesses, then remove the big kernel lock
from the UIO driver.  Since there are no in-tree UIO drivers with open()
methods, no further BKL pushdown is required.

Acked-by: Hans J. Koch 
Acked-by: Greg Kroah-Hartman 
Signed-off-by: Jonathan Corbet

UIO: Fix mapping of logical and virtual memory

2008-10-16T16:24:54+00:00

mmap() doesn't work as expected for UIO_MEM_LOGICAL or UIO_MEM_VIRTUAL
mappings. The offset into the memory needs to be added, otherwise
uio_vma_fault always returns the first page only. Note that for UIO
userspace calls mmap() with offset = N * getpagesize() to access
mapping N. This must be compensated when calculating the offset. A
comment was added to explain this since it is not obvious.

Signed-off-by: Andrew G. Harvey 
Signed-off-by: Hans J. Koch 
Signed-off-by: Greg Kroah-Hartman

UIO: add automata sercos3 pci card support

2008-10-16T16:24:53+00:00

Here is a new version of the patch to support the Automata Sercos III
PCI card driver. I now check that the IRQ is enabled before accepting
the interrupt.

I still use a logical OR to store the enabled interrupts and I've
added a second use of a logical OR when restoring the enabled
interrupts. I added an explanation of why I do this in comments at the
top of the source file.

Since I use a logical OR, I also removed the extra checks if the
Interrupt Enable Register and ier0_cache are 0.

Signed-off-by: John Ogness 
Signed-off-by: Hans J. Koch 
Signed-off-by: Greg Kroah-Hartman

UIO: Change driver name of uio_pdrv

2008-10-16T16:24:53+00:00

The generic UIO platform device driver should be given a unique driver ID and
not just "uio". This is especially important since we now have a similar driver
named uio_pdrv_genirq. Currently, there's no user of this driver in the
mainline kernel.

Signed-off-by: Hans J. Koch 
Signed-off-by: Greg Kroah-Hartman

UIO: Add alignment warnings for uio-mem

2008-10-16T16:24:53+00:00

This patch adds an "offset" attribute for UIO mappings. It shows the
difference between the actual start address of the memory and the start
address of the page.

Signed-off-by: Hans J. Koch 
Signed-off-by: Greg Kroah-Hartman