User reports rarely include full information, so include this important
tidbit up front.  It's also good to know at a glance in general.
Signed-off-by: NRobin Getz <robin.getz@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Every Blackfin variant has the same DMA bit masks, so avoid duplicating
them over and over in each mach header.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The driver changed from "isp1760-hcd" to "isp1760", so update resources
to match.
Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

No need to set MSTR in .ctl_reg as the Blackfin SPI bus driver does this
already for all parts.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

There's no point in having mask defines when the entire MMR value is a
count or address.  i.e. applying a mask of -1 is pointless.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

There are no MXVR device drivers, and if someday there is, we can put
these in a dedicated header rather than polluting the global namespace.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

People should not be accessing OTP MMRs directly.  They should instead go
through the Blackfin ROM helper functions.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The watchdog code doesn't need these, and the other parts had these
punted, so keep the global namespace clean.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The RTC driver code doesn't need these, and the other parts had these
punted, so keep the global namespace clean.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

No point in duplicating entire lists when we can simply tail into other
parts for most of the MMRs.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

All the other BF54x parts had these defines renamed to avoid collision,
but it looks the BF542 was missed somehow.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The cache lock code was unused and punted, so punt the documentation too.
Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

We don't need the BKL and now people will stop looking at this.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The DMA channel status field was encoding redundant info wrt the DMA MMR
config register, and it was doing an incomplete job of checking all DMA
channels (some drivers write directly to the config register).  So drop
the tristate field in favor of a binary atomic field.  This simplifies
the code in general, removes the implicit need for sleeping, and forces
the suspend code to handle all channels properly.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The driver was moved during the merge process, so update the defines to
match the new location.
Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Avoid including unnecessary headers all the time as well as circular
includes with core requirements.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMike Frysinger <vapier@gentoo.org>

All the arches define a helper macro to make things easy for driver code.
Reported-by: NFrank Van Hooft <frank@frankvh.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NBarry Song <barry.song@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The GPIOs on ports C/D/E on the BF538/BF539 do not behave the same way as
the other ports on the part and the same way as all other Blackfin parts.
The MMRs are programmed slightly different and they cannot be used to
generate interrupts or wakeup a sleeping system.  Since these guys don't
fit into the existing code, create a simple gpiolib driver for them.
Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Have the C API trace funcs match the assembly API trace funcs.
Signed-off-by: NGraf Yang <graf.yang@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

PLL_LOCKCNT applies only to the PLL programming sequence which does not
apply to core and system clock dividers.  Writes to PLL_DIV to change the
CSEL/SSEL dividers take effect immediately.

There is still overhead in software in writing the new dividers, so just
use a value of 50us as this should be good enough.
Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Drop the CONFIG_USB_ISP1362_BFIN_GPIO_IRQ Kconfig as it never made it into
mainline and it was a bad interface into the board resources.  For boards
that actually used this, replace it with an actual IRQ define.  For boards
that didn't, simply drop the resources.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Since the link sizes never change at runtime, push the calculation out to
the linker script to save some useless calculation costs.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The access_ok() function did not accept ranges within the async banks
which made it impossible to do XIP in flash.  Fixing that also showed
that the current bfin_mem_access_type() code did not work with accesses
that spanned async banks (like a file system).  So split out and fix the
async bank checks so that all these scenarios work as expected.
Signed-off-by: NBernd Schmidt <bernds_cb1@t-online.de>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Rather than using our own data structures that basically boil down to a
bitmap, use the standard bitmap functions.
Reported-by: NPaul Mundt <lethal@linux-sh.org>
Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Now that the driver has been updated, convert the board resources to the
new i2c framework for managing slaves.

For boards that don't actually hook up to this hardware, simply drop the
resources altogether.
Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The requested peripheral is turned into an index into some state arrays,
so make sure the calculated index doesn't exceed the index.  This occurs
when using bogus pin values or the define headers are screwed up.  Now
we'll notice right away that something needs fixing instead of trying to
track down random memory corruption.
Signed-off-by: NBarry Song <barry.song@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

No point in redefining things that common code already does for us.  Also
use CYC2NS_SCALE_FACTOR to better reflect reality and for better precision.
Signed-off-by: NYi Li <yi.li@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Currently sched_clock() is only defined when using CYCLES as a clock
source.  Declare sched_clock() in common code and mark it with notrace to
prevent invoking sched_clock() recursively (because ftrace uses
sched_clock() to record time).
Signed-off-by: NYi Li <yi.li@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Some of the clocksource prototypes were updated, but the gptimer0 func was
missed in the process.  Not a big issue as the argument is ignored, but we
should fix the compile warning anyways.
Signed-off-by: NYi Li <yi.li@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NCliff Cai <cliff.cai@analog.com>
Signed-off-by: NYi Li <yi.li@analog.com>
Signed-off-by: NHarald Krapfenbauer <Harald.Krapfenbauer@bluetechnix.at>
Signed-off-by: NGraf Yang <graf.yang@analog.com>
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

The Blackfin port only implemented an optimized version of the
csum_tcpudp_nofold function, so convert everything else to the new
generic code.
Signed-off-by: NMike Frysinger <vapier@gentoo.org>

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-udf-2.6:
  udf: Avoid IO in udf_clear_inode
  udf: Try harder when looking for VAT inode
  udf: Fix compilation with UDFFS_DEBUG enabled

It is not very good to do IO in udf_clear_inode. First, VFS does not really
expect inode to become dirty there and thus we have to write it ourselves,
second, memory reclaim gets blocked waiting for IO when it does not really
expect it, third, the IO pattern (e.g. on umount) resulting from writes in
udf_clear_inode is bad and it slows down writing a lot.

The reason why UDF needed to do IO in udf_clear_inode is that UDF standard
mandates extent length to exactly match inode size. But when we allocate
extents to a file or directory, we don't really know what exactly the final
file size will be and thus temporarily set it to block boundary and later
truncate it to exact length in udf_clear_inode. Now, this is changed to
truncate to final file size in udf_release_file for regular files. For
directories and symlinks, we do the truncation at the moment when learn
what the final file size will be.
Signed-off-by: NJan Kara <jack@suse.cz>

Some disks do not contain VAT inode in the last recorded block as required
by the standard but a few blocks earlier (or the number of recorded blocks
is wrong). So look for the VAT inode a bit before the end of the media.
Signed-off-by: NJan Kara <jack@suse.cz>