This reverts commit 329cabce.

The commit that caused us to specify LE device endianness here,
29bb45f2 (regmap-mmio: Use native endianness for read/write,
2015-10-29), has been reverted in mainline so now when we specify
LE it actively breaks big endian kernels because the byte
swapping in regmap-mmio is incorrect. Let's revert this change
because it will 1) fix the big endian kernels and 2) be redundant
to specify LE because that will become the default soon.

Cc: Kevin Hilman <khilman@linaro.org>
Tested-by: NKevin Hilman <khilman@baylibre.com>
Cc: Mark Brown <broonie@kernel.org>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>

The Versatile syscon ICST driver OR:s the bits into place but
forgets to mask the previous value, making the code only work
if the register is zero or giving haphazard results. Mask the
19 bits used by the Versatile syscon interface register.

Regression caused and now fixed by yours truly.

Cc: Michael Turquette <mturquette@baylibre.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: linux-clk@vger.kernel.org
Fixes: 179c8fb3 ("clk: versatile-icst: convert to use regmap")
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>

Sparse reports the following warnings for structures and functions that
should be declared static:

drivers/clk/tegra/clk-tegra-super-gen4.c:70:35: warning: symbol
 'tegra_super_gen_info_gen4' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra-super-gen4.c:96:35: warning: symbol
 'tegra_super_gen_info_gen5' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra-super-gen4.c:174:13: warning: symbol
 'tegra_super_clk_init' was not declared. Should it be static?

Fix this by making the above static.
Signed-off-by: NJon Hunter <jonathanh@nvidia.com>
Acked-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

Sparse reports the following warnings for functions in clk-tegra210.c
that should be declared as static:

drivers/clk/tegra/clk-tegra210.c:460:6: warning: symbol
 'tegra210_pllcx_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:485:6: warning: symbol
 '_pllc_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:490:6: warning: symbol
 '_pllc2_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:495:6: warning: symbol
 '_pllc3_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:500:6: warning: symbol
 '_plla1_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:510:6: warning: symbol
 'tegra210_plla_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:562:6: warning: symbol
 'tegra210_plld_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:701:6: warning: symbol
 'tegra210_plld2_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:709:6: warning: symbol
 'tegra210_plldp_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:722:6: warning: symbol
 'tegra210_pllc4_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:731:6: warning: symbol
 'tegra210_pllre_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:844:6: warning: symbol
 'tegra210_pllx_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:904:6: warning: symbol
 'tegra210_pllmb_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:963:6: warning: symbol
 'tegra210_pllp_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:1025:6: warning: symbol
 'tegra210_pllu_set_defaults' was not declared. Should it be static?
drivers/clk/tegra/clk-tegra210.c:1215:15: warning: symbol
 'tegra210_clk_adjust_vco_min' was not declared. Should it be static?

Fix this by declaring the above as static.
Signed-off-by: NJon Hunter <jonathanh@nvidia.com>
Acked-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

Sparse generates the following warning for the pll_m params structure:

drivers/clk/tegra/clk-tegra210.c:1569:10: warning: Initializer entry
 defined twice
drivers/clk/tegra/clk-tegra210.c:1570:10:   also defined here

Fix this by correcting the index for the MISC1 register.

Fixes: b31eba5ff3f7 ("clk: tegra: Add support for Tegra210 clocks")
Signed-off-by: NJon Hunter <jonathanh@nvidia.com>
Acked-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

The definition, PLLU_BASE_OVERRIDE, for the pll_u OVERRIDE bit is defined
but not used and when the OVERRIDE bit is cleared in tegra210_pll_init()
the code directly uses the bit number. Therefore, use the definition,
PLLU_BASE_OVERRIDE when clearing the OVERRIDE bit.
Signed-off-by: NJon Hunter <jonathanh@nvidia.com>
Acked-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

If the pll_u is not configured by the bootloader, then on kernel boot the
following warning is seen:

 clk_pll_wait_for_lock: Timed out waiting for pll pll_u_vco lock
 tegra_init_from_table: Failed to enable pll_u_out1
 ------------[ cut here ]------------
 WARNING: at drivers/clk/tegra/clk.c:269
 Modules linked in:

 CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.4.0-rc4-next-20151214+ #1
 Hardware name: NVIDIA Tegra210 P2371 reference board (E.1) (DT)
 task: ffffffc0bc0a0000 ti: ffffffc0bc0a8000 task.ti: ffffffc0bc0a8000
 PC is at tegra_init_from_table+0x140/0x164
 LR is at tegra_init_from_table+0x140/0x164
 pc : [<ffffffc0008fee78>] lr : [<ffffffc0008fee78>] pstate: 80000045
 sp : ffffffc0bc0abd50
 x29: ffffffc0bc0abd50 x28: ffffffc00090b8a8
 x27: ffffffc000a06000 x26: ffffffc0bc019780
 x25: ffffffc00086a708 x24: ffffffc00086a790
 x23: ffffffc0006d7188 x22: ffffffc0bc010000
 x21: 000000000000016e x20: ffffffc0bc00d100
 x19: ffffffc000944178 x18: 0000000000000007
 x17: 000000000000000e x16: 0000000000000001
 x15: 0000000000000007 x14: 000000000000000e
 x13: 0000000000000013 x12: 000000000000001a
 x11: 000000000000004d x10: 0000000000000750
 x9 : ffffffc0bc0a8000 x8 : ffffffc0bc0a07b0
 x7 : 0000000000000001 x6 : 0000000002d5f0f8
 x5 : 0000000000000000 x4 : 0000000000000000
 x3 : 0000000000000002 x2 : ffffffc000996724
 x1 : 0000000000000000 x0 : 0000000000000032

 ---[ end trace cbd20ae519e92ced ]---
 Call trace:
 [<ffffffc0008fee78>] tegra_init_from_table+0x140/0x164
 [<ffffffc000900ac8>] tegra210_clock_apply_init_table+0x20/0x28
 [<ffffffc0008fec40>] tegra_clocks_apply_init_table+0x18/0x24
 [<ffffffc00008291c>] do_one_initcall+0x90/0x194
 [<ffffffc0008cfab0>] kernel_init_freeable+0x148/0x1e8
 [<ffffffc000636bb0>] kernel_init+0x10/0xdc
 [<ffffffc000085cd0>] ret_from_fork+0x10/0x40
 clk_pll_wait_for_lock: Timed out waiting for pll pll_u_vco lock
 tegra_init_from_table: Failed to enable pll_u_out2
 ------------[ cut here ]------------

pll_u can be either controlled by software or hardware and this is
selected via the OVERRIDE bit in the pll_u base register. In the function
tegra210_pll_init(), the OVERRIDE bit for pll_u is cleared, which selects
hardware control of the pll. However, at the same time the pll_u clocks
are populated in the init_table for tegra210 and so software will try to
configure the pll_u if it is not already configured and hence, the above
warning is seen when the pll fails to lock. Remove the pll_u clocks from
the init_table so that software does not try to configure this pll on
boot.
Signed-off-by: NJon Hunter <jonathanh@nvidia.com>
Acked-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

The EMC clock sources for Tegra210 currently incorrectly include pll_c2
and pll_c3. However, both of these should have been pll_mb as shown in
the TRM. If Tegra210 happens to be configured such that the pll_mb is the
default clock for the EMC, as configured by the bootloader, then this will
cause a system hang on boot. This is because the kernel will disable the
pll_mb when disabling unused clock as it appears to be unused when it is
not.

Also add the additional pll_p clock source for the EMC.
Signed-off-by: NJon Hunter <jonathanh@nvidia.com>
Acked-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

The APB2APE clock for the audio subsystem is required for powering up the
audio power domain and accessing the various modules in this subsystem on
Tegra210 devices. Add this clock for Tegra210.
Signed-off-by: NJon Hunter <jonathanh@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

for_each_child_of_node() performs an of_node_get() on each iteration, so
before breaking out of the loop an of_node_put() is required.

Found using Coccinelle. The semantic patch used for this is as follows:

// <smpl>
@@
expression e;
local idexpression child;
@@

 for_each_child_of_node(root, child) {
   ... when != of_node_put(child)
       when != e = child
(
   return child;
|
+  of_node_put(child);
?  return ...;
)
   ...
 }
// </smpl>
Signed-off-by: NAmitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

The PLLE SS coefficients are different between Tegra210 and Tegra114.
Add SoC generation specific versions for Tegra114 and Tegra210 and use
them in their respective ->enable() callbacks.
Signed-off-by: NMark Kuo <mkuo@nvidia.com>
Signed-off-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

While enabling PLLE on both Tegra114 and Tegra210, we should be clearing
PLLE_MISC_VREG_BG_CTRL_MASK and PLLE_MISC_VREG_CTRL_MASK not setting
them. This patch fixes both places where we incorrectly set instead of
cleared those bits.
Signed-off-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

Software should not disable PLLE if PLLE is already put under hardware
control.
Signed-off-by: NMark Kuo <mkuo@nvidia.com>
Signed-off-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

The logic for calculating the input rate used when figuring out the
proper dynamic steps for pllx was incorrect. It is supposed to be
calculated using parent_rate / m but it was just using the parent rate
directly, therefore using the wrong step values.
Signed-off-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

Since the ->enable() callback is called with a spinlock held, we cannot
call potentially blocking functions such as clk_get_rate() or
clk_get_parent(), so use the unlocked versions instead.
Signed-off-by: NAndrew Bresticker <abrestic@chromium.org>
[rklein: Adapted from ChromeOS patch, removing pllu_enable cleanup as
it isn't present upstream]
Signed-off-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

When adding the nvenc clock, it was partially named msenc in the code.
Since the msenc clock isn't present in Tegra210 and has been replaced by
the nvenc clock, its misleading to see it present. Therefore, properly
rename it.
Signed-off-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

Some register for PLLM and PLLMB were named MISC0 but according to the
TRM, they have different names. Sync up the names to make it easier to
understand which register they are really referring to.
Signed-off-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

Requested by arm-soc maintainer Kevin Hilman in v9 review.
http://article.gmane.org/gmane.linux.ports.arm.kernel/456331Signed-off-by: NMarc Gonzalez <marc_gonzalez@sigmadesigns.com>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>

platform_device_register_simple() returns ERR_PTR on error.
Signed-off-by: NAxel Lin <axel.lin@ingics.com>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>

Silence the following sparse warning

drivers/clk/mvebu/dove-divider.c:252:14: warning: incorrect type in assignment (different address spaces)
drivers/clk/mvebu/dove-divider.c:252:14:    expected void *base
drivers/clk/mvebu/dove-divider.c:252:14:    got void [noderef] <asn:2>*
drivers/clk/mvebu/dove-divider.c:256:13: warning: incorrect type in argument 2 (different address spaces)
drivers/clk/mvebu/dove-divider.c:256:13:    expected void [noderef] <asn:2>*base
drivers/clk/mvebu/dove-divider.c:256:13:    got void *base
drivers/clk/mvebu/dove-divider.c:257:25: warning: incorrect type in argument 1 (different address spaces)
drivers/clk/mvebu/dove-divider.c:257:25:    expected void volatile [noderef] <asn:2>*iomem_cookie
drivers/clk/mvebu/dove-divider.c:257:25:    got void *base

Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>

Most PLL's don't actually have LOCK_ENABLE bits. However, most PLL's
also had that flag set, which meant that the clk code was trying to
enable locks, and inadvertantly flipping bits in other fields.

For PLLM, ensure the correct register is used for the misc_register.
PLL_MISC0 contains the EN_LCKDET bit which should be used for enabling
the lock, and PLLM_MISC1 shouldn't be used at all.

Lastly, remove some of the settings which would point to the EN_LCKDET
bits for some PLLs. There is no need to enable the locks, and that is
done as part of the set_defaults logic already.
Signed-off-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

VI-I2C has 16 bits available for its divider. Switch the divider width
to 16 instead of 8 so correct rates can be set.
Signed-off-by: NRhyland Klein <rklein@nvidia.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

Like the Yoga 900 models the Lenovo Yoga 700 does not have a
hw rfkill switch, and trying to read the hw rfkill switch through the
ideapad module causes it to always reported blocking breaking wifi.

This commit adds the Lenovo Yoga 700 to the no_hw_rfkill dmi list, fixing
the wifi breakage.

BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1295272
Tested-by: <dinyar.rabady+spam@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: NJosh Boyer <jwboyer@fedoraproject.org>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

This patch fixes compile time warnings when CONFIG_PM_SLEEP
is undefined. In this case sleep related counters are unused.
Signed-off-by: NSouvik Kumar Chakravarty <souvik.k.chakravarty@intel.com>
Signed-off-by: NDarren Hart <dvhart@linux.intel.com>

The extended flash address needs to be subtracted from bcm_tag flash
image offsets. Move this value to the bcm_tag header file.

Renamed define name to consistently use bcm963xx for flash layout
which should be considered a property of the board and not the SoC
(i.e. bcm63xx could theoretically be used on a board without CFE
or any flash).
Signed-off-by: NSimon Arlott <simon@fire.lp0.eu>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Brian Norris <computersforpeace@gmail.com>
Cc: Kevin Cernekee <cernekee@gmail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Cc: Jonas Gorski <jogo@openwrt.org>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Cc: MIPS Mailing List <linux-mips@linux-mips.org>
Cc: MTD Maling List <linux-mtd@lists.infradead.org>
Patchwork: https://patchwork.linux-mips.org/patch/11833/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Move Broadcom BCM963xx image tag data structure to include/linux/
so that drivers outside of mach-bcm63xx can use it.
Signed-off-by: NSimon Arlott <simon@fire.lp0.eu>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Brian Norris <computersforpeace@gmail.com>
Cc: Kevin Cernekee <cernekee@gmail.com>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Cc: Jonas Gorski <jogo@openwrt.org>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Cc: MIPS Mailing List <linux-mips@linux-mips.org>
Cc: MTD Maling List <linux-mtd@lists.infradead.org>
Patchwork: https://patchwork.linux-mips.org/patch/11832/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

This adds support for the interrupt controller present on PIC32 class
devices. It handles all internal and external interrupts. This controller
exists outside of the CPU core and is the arbitrator of all interrupts
(including interrupts from the CPU itself) before they are presented to
the CPU.

The following features are supported:
 - DT properties for EVIC and for devices/peripherals that use interrupt lines
 - Persistent and non-persistent interrupt handling
 - irqdomain and generic chip support
 - Configuration of external interrupt edge polarity
Signed-off-by: NCristian Birsan <cristian.birsan@microchip.com>
Signed-off-by: NJoshua Henderson <joshua.henderson@microchip.com>
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Cc: Jason Cooper <jason@lakedaemon.net>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/12092/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

There are many locations that do

  if (memory_was_allocated_by_vmalloc)
    vfree(ptr);
  else
    kfree(ptr);

but kvfree() can handle both kmalloc()ed memory and vmalloc()ed memory
using is_vmalloc_addr().  Unless callers have special reasons, we can
replace this branch with kvfree().  Please check and reply if you found
problems.
Signed-off-by: NTetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Acked-by: NMichal Hocko <mhocko@suse.com>
Acked-by: NJan Kara <jack@suse.com>
Acked-by: NRussell King <rmk+kernel@arm.linux.org.uk>
Reviewed-by: NAndreas Dilger <andreas.dilger@intel.com>
Acked-by: N"Rafael J. Wysocki" <rjw@rjwysocki.net>
Acked-by: NDavid Rientjes <rientjes@google.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Oleg Drokin <oleg.drokin@intel.com>
Cc: Boris Petkov <bp@suse.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

parallel to mutex_{lock,unlock,trylock,is_locked,lock_nested},
inode_foo(inode) being mutex_foo(&inode->i_mutex).

Please, use those for access to ->i_mutex; over the coming cycle
->i_mutex will become rwsem, with ->lookup() done with it held
only shared.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

gcc warns about the 'found' variable possibly being used uninitialized:

drivers/soc/qcom/spm.c: In function 'spm_dev_probe':
drivers/soc/qcom/spm.c:305:5: error: 'found' may be used uninitialized in this function [-Werror=maybe-uninitialized]

However, the code is correct because we know that there is
always at least one online CPU. This initializes the 'found'
variable to zero before the loop so the compiler knows
it does not have to warn about it.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

If the parameter given to the macro is replaced throughout the macro as
it is evaluated.  The intent is that the macro parameter should replace
the only the first parameter to container_of().  However, the way the
macro was written, it would also inadvertantly replace a structure field
name.  If a parameter of any other name is given to the macro, it will
fail to compile, if the structure does not contain a field of the same
name.  At worst, it will compile, and hide improper access of an
unintended field in the structure.

Change the macro parameter name, so it does not conflict with the
structure field name.
Signed-off-by: NAllen Hubbe <Allen.Hubbe@emc.com>
Acked-by: NDave Jiang <dave.jiang@intel.com>
Signed-off-by: NJon Mason <jdmason@kudzu.us>

This adds support for AMD's PCI-Express Non-Transparent Bridge
(NTB) device on the Zeppelin platform. The driver connnects to the
standard NTB sub-system interface, with modification to add hooks
for power management in a separate patch. The AMD NTB device has 3
memory windows, 16 doorbell, 16 scratch-pad registers, and supports
up to 16 PCIe lanes running a Gen3 speeds.
Signed-off-by: NXiangliang Yu <Xiangliang.Yu@amd.com>
Reviewed-by: NAllen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: NJon Mason <jdmason@kudzu.us>

The rbd_dev_destroy() function tests whether its argument is NULL
and then returns immediately. Thus the test around the call is not needed.

This issue was detected by using the Coccinelle software.
Signed-off-by: NMarkus Elfring <elfring@users.sourceforge.net>
Signed-off-by: NIlya Dryomov <idryomov@gmail.com>

The create_cq() can receive creation flags which were used
differently by two commits which added create_cq extended
command and cross-channel. The merged code caused to not
accept any flags at all.

This patch unifies the check into one function and one return
error code.

Fixes: 972ecb82 ("IB/mlx5: Add create_cq extended command")
Fixes: 051f2630 ("IB/mlx5: Add driver cross-channel support")
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

Added Raw Packet QP modify functionality which will enable user
space consumers to use it.

Since Raw Packet QP is built of SQ and RQ sub-objects, therefore
Raw Packet QP state changes are implemented by changing the state
of the sub-objects.
Signed-off-by: NMajd Dibbiny <majd@mellanox.com>
Reviewed-by: NMatan Barak <matanb@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

When modifying a QP, the desired operation was determined in
the mlx5_core using a transition table that takes the current
state, the final state, and returns the desired operation.

Since this logic will be used for Raw Packet QP, move the
operation table to the mlx5_ib.
Signed-off-by: NMajd Dibbiny <majd@mellanox.com>
Reviewed-by: NMatan Barak <matanb@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

When the user changes the Address Vector(AV) in the modify QP, he
provides an SL. This SL should be translated to Ethernet Priority
by taking the 3 LSB bits, and modify the QP's TIS according to this
Ethernet priority.
Signed-off-by: NMajd Dibbiny <majd@mellanox.com>
Reviewed-by: NMatan Barak <matanb@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

Since Raw Packet QP is composed of RQ and SQ, the IB QP's
state is derived from the sub-objects. Therefore we need
to query each one of the sub-objects, and decide on the
IB QP's state.
Signed-off-by: NMajd Dibbiny <majd@mellanox.com>
Reviewed-by: NMatan Barak <matanb@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>

This patch adds support for Raw Packet QP for the mlx5 device.

Raw Packet QP, unlike other QP types, has no matching mlx5_core_qp
object but rather it is built of RQ/SQ/TIR/TIS/TD mlx5_core object.

Since the SQ and RQ work-queue (WQ) buffers are not contiguous like
other QPs, we allocate separate buffers in the user-space and pass
the address of each one of them separately to the kernel.
Signed-off-by: NMajd Dibbiny <majd@mellanox.com>
Reviewed-by: NMatan Barak <matanb@mellanox.com>
Signed-off-by: NDoug Ledford <dledford@redhat.com>