Until now, only 32-bit DMA addressing was allowed, following a report on
some old Intel machine that dropped 64-bit PCIe packets, even though
pci_set_dma_mask() was successful with DMA_BIT_MASK(64).

But then came TI's Keystone II chip (ARM Cortex A15 + DSPs), which refuses
32-bit DMA addressing (for good reasons). So 64-bit DMA is allowed as a
fallback option.
Signed-off-by: NEli Billauer <eli.billauer@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This driver has been functional and stable throughout the year it has spent
in the staging area. It has been patched for minor bugs, coding style issues
and improvements during this period.

This is the second submission of this move-out, after making several style
improvements, as suggested by Dan Carpenter.
Signed-off-by: NEli Billauer <eli.billauer@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Signed-off-by: NEli Billauer <eli.billauer@gmail.com>
Reviewed-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Suggested-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NEli Billauer <eli.billauer@gmail.com>
Reviewed-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

"rc" is often used to hold the return value of a function call. This patch
removes unnecessary assignments to this variable, and makes a few related
execution flow improvements.
Suggested-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NEli Billauer <eli.billauer@gmail.com>
Reviewed-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Managed device resource API replaces code that reinvents it for memory
allocation, page allocation and DMA mapping.

devm_add_action() is used for unwinding DMA mappings, since there is no
devm_* API for dma_map_single(). A recent patch that introduces such API
was rejected, mainly on the grounds that it may cause an unnecessary waste of
resources.
Suggested-by: NBaruch Siach <baruch@tkos.co.il>
Signed-off-by: NEli Billauer <eli.billauer@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Suggested-by: NBaruch Siach <baruch@tkos.co.il>
Signed-off-by: NEli Billauer <eli.billauer@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Fix to return negative error code -EIO from the error handling
case instead of 0.
Signed-off-by: NWei Yongjun <yongjun_wei@trendmicro.com.cn>
Reviewed-by: NJingoo Han <jg1.han@samsung.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Don't use DEFINE_PCI_DEVICE_TABLE macro, because this macro
is not preferred.
Signed-off-by: NJingoo Han <jg1.han@samsung.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Suggested-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NEli Billauer <eli.billauer@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This is necessary so that xillybus_core uses the correct device pointer
for PCIe devices in diagnostic message calls (dev_err, dev_warn and dev_info)
Signed-off-by: NEli Billauer <eli.billauer@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Changes:

* xillybus.h: __iomem added to struct xilly_endpoint -> registers to suppress
  "different address spaces" errors.

* xillybus_core.c: __user added as required for the same reason.

* The two member names of struct xilly_endpoint_hardware of the form
  sync_single_for_{cpu,device} were changed to something that won't look
  like the well-known functions.

* All *.c files: Variables and functions made static as required.
Reported-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NEli Billauer <eli.billauer@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NEli Billauer <eli.billauer@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This is the driver for Xillybus, which is a general-purpose interface for
data communication with FPGAs (programmable logic). Please refer to the
README included in this patch for a detailed explanation.

It was previously submitted for misc-devices, but it appears like noone's
willing to review the code (which I can understand, given its magnitude).
Hence submitted as a staging driver.
Signed-off-by: NEli Billauer <eli.billauer@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>