fs: Add destroy_inode callbacks to filesystems ...

Several filesystems rely on the default function which frees
the struct inode * rather than the filesystem specific inode
which the inode is embedded in. This works because the inode
is the first element in the filesystem specific inode. Let's
not depend on this behaviour and for clarity add the destroy_inode
callbacks to all filesystems.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

treewide: remove references to CREDITS ...

The CREDITS file was removed from barebox in 2015 by commit 6570288f2d97
("Remove the CREDITS file"). Remove references to it from several files.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

cdev: Add discard_range hook ...

To pass though discard_range() to the underlying drivers add a
discard_range hook to struct cdev_operations.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: Make use of cdev_flush() ...

Drop extra checks and explicit indirect call in devfs_flush() in
favour of using cdev_flush(), since it already does all of the above.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: Make use of cdev_erase() ...

Drop extra checks and explicit indirect call in devfs_erase() in
favour of using cdev_erase(), since it already does all of the above.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: devfs: forbid truncation when cdev has no truncate operation ...

When a cdev doesn't have a truncate callback then forbid truncation
and fail with -EPERM.
Before this we had always failed with -ENOSPC in this situation.
We checked for f->fsdev->dev.num_resources being nonzero, but this
check was absolutely meaningless. It goes back to ancient times when
the resources of a device were automatically added to devfs.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: let truncate take a loff_t argument ...

loff_t is the correct type for file sizes. Use it to allow to truncate
to sizes bigger than 32bit.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: devfs: Change .lseek callbacks to return 'int' ...

Returning requested offset from .lseek() callback doesn't really give
us any new information while bringing unnecessary
complications. Change all .lseek() types (both in struct struct
cdev_operations and in struct fs_driver_d) to return 'int' and adjust
the rest of the codebase accordingly.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

devfs: Drop dev_lseek_default() ...

Only the following cdevs do not declare an .lseek() operation:

  - Console devices in common/console.c
  - Firmware framework in common/firmware.c
  - JTAG driver in drivers/misc/jtag.c
  - UBI in drivers/mtd/ubi/barebox.c

Of those four, first two are marked DEVFS_IS_CHARACTER_DEV and
implement only .write() operation and the last two don't implement
anything but .ioctl(). While there's probably no meaningful way to use
lseek() against any of those devices, there doesn't seem to be any
harm in allowing it either.

Change devfs_lseek() to ignore absense of .lseek() callback and drop
dev_lseek_default() and all references to it in the codebase.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: Update FILE position in lseek() ...

Instead on relying on driver callbacks to update 'pos' in FILE, do it
as a part of lseek() code. This allows us to drop a bit of repeating
code as well as making lseek() implementation consistent with write()
and read().

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

devfs: Fix incorrect error check for cdev->ops->lseek() ...

Cdev->ops->lseek() will either return a negative error code on failure
or requested position on success. In order to properly check for
errors we need to test if the return value is negative, not just that
it's not -1.

Returning ret - cdev->offset, doesn't appear to be correct either,
even if ret is -1, since on failure this will lead us to return (-1 -
cdev->offset). Simplify that part by just returning 'pos', which is
what we'd end up returning on success in original code as well.

Third, make sure to return -ENOSYS, when no .lseek() callback is
provided.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: devfs: Make locally used function static ...

devfs_iterate() is only used locally, so make it static.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: devfs: implement d_revalidate hook ...

The files in devfs can change withouut the fs layer noticing, so we
have to revalidate dentries before using them.

A failure could be triggered with:

ls /dev/nand0.root.ubi; ubiattach /dev/nand0.root; ls /dev/nand0.root.ubi

The first 'ls' would create a dentry for nand0.root.ubi with no inode
associated since it does not yet exist. 'ubiattach' then creates that
file, but the second 'ls' does not show it since the dentry is not
revalidated and thus no inode is added to that dentry. This patch
fixes this and also the opposite case when a file is removed (for
example with ubidetach).

Reported-by: Ladislav Michl <ladis@linux-mips.org>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: devfs: fix r/w permissions ...

The real r/w flags were lost in the switch to dentry cache
implementation. Restore them.

Fixes: 9137c41915 ("fs: devfs: Switch to dentry cache implementation")

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: devfs: Create device files as character devices ...

Without this files in /dev/ appear as regular files. copy_file()
stumbles upon this as it tries a truncate() on that files which
then fails. Create the files as character devices so that copy_file()
works as expected.

Fixes: 9137c41915 ("fs: devfs: Switch to dentry cache implementation")

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Tested-by: Giorgio Dal Molin <giorgio.nicole@arcor.de>

fs: devfs: Switch to dentry cache implementation ...

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

devfs: Add optional truncate callback for device files ...

Not all device files have trivial fix device sizes like static
ubi volumes. Here the device file size equals the image size it contains.

Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: Fix erasing of devices >4GiB ...

When a device >4GiB is erased, not only the offset can be bigger
than 4GiB, but also the size. This happens with the simplest command
to erase a device: erase /dev/nand0. Make the size argument a 64bit
type to make this work.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

devfs: Add symlink support ...

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: devfs: Allow mounting only on /dev/ ...

We have places in the code where we assume that devfs is mounted on
/dev/, so enforce this path to avoid surprises.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: rename inode member of struct filep to priv ...

Because that's what it is. 'inode' will become confusing
once we support real inodes.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: store pointer to fsdev instead of dev in struct filep ...

The struct device_d * in struct filep is never of interest, instead
it is always converted to a struct fs_device_d *, so simplify the code
by storing the struct fs_device_d * directly.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Treewide: remove address of the Free Software Foundation ...

The FSF address has changed in the past. Instead of updating it
each time the address changes, just drop it completely treewide.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs: fix standard zero, full devices ...

The standard devices are currently broken since they have
the size ~0. As now files use loff_t as file size which is a signed
type the read implementation gets confused and now returns -1.

The current implementation also has the (somewhat theorical) problem
that we do not have real streaming devices, so /dev/zero went out
of zeroes after reading 4GB (or now LLONG_MAX).

This patch introduces a new cdev flag DEVFS_IS_CHARACTER_DEV and a new
file size flag FILE_SIZE_STREAM which makes it possible to create
real stream devices instead.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'for-next/resource-size' ...

Conflicts:
	drivers/base/resource.c
	fs/fs.c

Merge branch 'for-next/misc'

resource: store 'end' instead of 'size' in struct resource ...

Storing the size instead of the resource end in struct resource was
a mistake. 'size' ranges from 0 to UINT[32|64]_MAX + 1 which obviously
leads to problems. 'end' on the other hand will never exceed
UINT[32|64]_MAX. Also this way we can express a iomem region covering
the whole address space.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

make cdev 64bit capable ...

Next step to 64bit support: Make cdev size a 64bit type.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

use loff_t for file offsets ...

This is a first step for 64bit file support: Make the file sizes/offsets
64bit.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

fs/devfs.c: disallow write and erase if a partition is read-only ...

Signed-off-by: Jan Luebbe <jlu@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>