The force_kill member of struct oom_control isn't needed if an order of -1
is used instead.  This is the same as order == -1 in struct
compact_control which requires full memory compaction.

This patch introduces no functional change.
Signed-off-by: NDavid Rientjes <rientjes@google.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
Cc: Michal Hocko <mhocko@suse.cz>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

There are essential elements to an oom context that are passed around to
multiple functions.

Organize these elements into a new struct, struct oom_control, that
specifies the context for an oom condition.

This patch introduces no functional change.
Signed-off-by: NDavid Rientjes <rientjes@google.com>
Acked-by: NMichal Hocko <mhocko@suse.cz>
Cc: Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

HVM_PARAM_CONSOLE_PFN is used to retrieved the console PFN for HVM
guest. It returns a PFN (aka GFN) and not a MFN.

Furthermore, use directly virt_to_gfn for both PV and HVM domain rather
than doing a special case for each of the them.
Signed-off-by: NJulien Grall <julien.grall@citrix.com>
Reviewed-by: NDavid Vrabel <david.vrabel@citrix.com>
Signed-off-by: NDavid Vrabel <david.vrabel@citrix.com>

Based on include/xen/mm.h [1], Linux is mistakenly using MFN when GFN
is meant, I suspect this is because the first support for Xen was for
PV. This resulted in some misimplementation of helpers on ARM and
confused developers about the expected behavior.

For instance, with pfn_to_mfn, we expect to get an MFN based on the name.
Although, if we look at the implementation on x86, it's returning a GFN.

For clarity and avoid new confusion, replace any reference to mfn with
gfn in any helpers used by PV drivers. The x86 code will still keep some
reference of pfn_to_mfn which may be used by all kind of guests
No changes as been made in the hypercall field, even
though they may be invalid, in order to keep the same as the defintion
in xen repo.

Note that page_to_mfn has been renamed to xen_page_to_gfn to avoid a
name to close to the KVM function gfn_to_page.

Take also the opportunity to simplify simple construction such
as pfn_to_mfn(page_to_pfn(page)) into xen_page_to_gfn. More complex clean up
will come in follow-up patches.

[1] http://xenbits.xen.org/gitweb/?p=xen.git;a=commitdiff;h=e758ed14f390342513405dd766e874934573e6cbSigned-off-by: NJulien Grall <julien.grall@citrix.com>
Reviewed-by: NStefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Acked-by: NWei Liu <wei.liu2@citrix.com>
Signed-off-by: NDavid Vrabel <david.vrabel@citrix.com>

This reverts commit 534e14e2 as with
this patch the serial console is broken on lots of platforms.
Reported-by: NMarc Zyngier <marc.zyngier@arm.com>
Cc: Jun Nie <jun.nie@linaro.org>
Acked-by: NWill Deacon <will.deacon@arm.com>
Tested-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This reverts commit 7b753f31 as with
this patch the serial console is broken on lots of platforms.
Reported-by: NMarc Zyngier <marc.zyngier@arm.com>
Cc: Jun Nie <jun.nie@linaro.org>
Acked-by: NWill Deacon <will.deacon@arm.com>
Tested-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This reverts commit 2c096a9e as with
this patch the serial console is broken on lots of platforms.
Reported-by: NMarc Zyngier <marc.zyngier@arm.com>
Cc: Jun Nie <jun.nie@linaro.org>
Acked-by: NWill Deacon <will.deacon@arm.com>
Tested-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This reverts commit 09dcc7df as with
this patch the serial console is broken on lots of platforms.
Reported-by: NMarc Zyngier <marc.zyngier@arm.com>
Cc: Jun Nie <jun.nie@linaro.org>
Acked-by: NWill Deacon <will.deacon@arm.com>
Tested-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This reverts commit 8cd90e50 as with
this patch the serial console is broken on lots of platforms.
Reported-by: NMarc Zyngier <marc.zyngier@arm.com>
Cc: Jun Nie <jun.nie@linaro.org>
Acked-by: NWill Deacon <will.deacon@arm.com>
Tested-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Weak header file declarations are error-prone because they make every
definition weak, and the linker chooses one based on link order (see
10629d71 ("PCI: Remove __weak annotation from pcibios_get_phb_of_node
decl")).

The most elegant solution is to have a weak default implementation and
allow a strong function to override it.  Then we don't have to test
whether a definition is present, and if there are ever multiple strong
definitions, we get a link error instead of calling a random definition.

Add a weak get_c0_fdc_int() definition with the default code and remove the
weak annotation from the declaration.
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>
Reviewed-by: NJames Hogan <james.hogan@imgtec.com>
Cc: Andrew Bresticker <abrestic@chromium.org>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/10687/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

This patch fixes several endianness issues detected when running the HVSI
driver in little endian mode.

These issues are raised in little endian mode because the data exchanged in
memory between the kernel and the hypervisor has to be in big endian
format. This exhibits as errors such as:

  irq: (null) didn't like hwirq-0x1000a00 to VIRQ16 mapping (rc=-22)
  hvsi_console_init: couldn't create irq mapping for 0x1000a00

The data structures already have endian annotations, and sparse is
generating numerous warnings based on those. This commit fixes all of
them.
Signed-off-by: NLaurent Dufour <ldufour@linux.vnet.ibm.com>
CC: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CC: Jiri Slaby <jslaby@suse.cz>
CC: linuxppc-dev@lists.ozlabs.org
CC: linux-kernel@vger.kernel.org
[mpe: Flesh out change log]
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

This change teaches the imx serial driver to save its
context and restore it across suspend and resume path.
To do so, it introduces serial_imx_restore_context()
and serial_imx_save_context() functions. They use
a shadow set of registers to save key registers
and restore them accordingly. These functions can
be reused on other situations, when the device
context is lost.

Cc: Fabio Estevam <festevam@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.com>
Cc: linux-serial@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: NEduardo Valentin <edubezval@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch sets RTSDEN bit when going into idle (Stop mode).
We add the RTSDEN for the case RTS is sent from
the remote connection. This way we allow the system
to wakeup when RTS is received.

Cc: Fabio Stevam <festevam@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.com>
Cc: linux-serial@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: NEduardo Valentin <edubezval@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This change is a code reorganization. Here we introduce
serial_imx_enable_wakeup() helper function to do
the job of configuring and preparing wakeup sources
on imx serial device. The idea is to allow other
parts of the code to call this function whenever
the device is known to go to idle.

Cc: Fabio Estevam <festevam@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.com>
Cc: linux-serial@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: NEduardo Valentin <edubezval@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The current code attempts to prepare clk_per and clk_ipg
before using the device. However, the result is an extra
prepare call on each clock. Here is the output of uart
clocks (only uart enabled and used as console):

$  grep uart /sys/kernel/debug/clk/clk_summary
 uart_serial           1            2    80000000          0 0
       uart           1            2    66000000          0 0

This patch balances the calls of prepares. The result is:

$  grep uart /sys/kernel/debug/clk/clk_summary
 uart_serial           1            1    80000000          0 0
       uart           1            1    66000000          0 0

Cc: Fabio Estevam <festevam@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.com>
Cc: linux-serial@vger.kernel.org
Cc: linux-pm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: NEduardo Valentin <edubezval@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

That bitfield is modified by read + or + write operation. If someone
sets any of the other two bits it might render the lock useless.

While at it, remove other bitfields as well to avoid more such
errors.

Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohn Ogness <john.ogness@linutronix.de>
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The 8250-omap driver requires the DMA-engine driver to support the pause
command in order to properly turn off programmed RX transfer before the
driver stars manually reading from the FIFO.
The lacking support of the requirement has been discovered recently. In
order to stay safe here we disable RX-DMA completly on probe.
The rx_dma_broken assignment on probe could be removed once we working
pause function in omap-dma.

Cc: <stable@vger.kernel.org>
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Since the serial core is informed that this device is doing
automatic flow control, it is not necessary to alter CTS and
DCTS of the MSR as the core no longer attempts stop the port
in uart_handle_cts_change().
Signed-off-by: NHuiquan Zhong <huiquan.zhong@intel.com>
Signed-off-by: NQipeng Zha <qipeng.zha@intel.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

port->lock get's initialized in uart_add_one_port(), no need to do it
in men_z135_probe().
Signed-off-by: NJohannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Fix panic caused by a race between men_z135_intr() and men_z135_set_termios().

men_z135_intr() and men_z135_set_termios() both hold the struct uart_port::lock
spinlock, but men_z135_intr() does a spin_lock_irqsave() and
men_z135_set_termios() does a normal spin_lock(), which can lead to a deadlock
when an interrupt is called while the lock is being helt by
men_z135_set_termios().

This was discovered using a insmod, hardware looppback send/receive, rmmod
stress test.
Signed-off-by: NJohannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: NPeter Hurley <peter@hurleysoftware.com>
Cc: Andreas Werner <andreas.werner@men.de>
Cc: stable@vger.kernel.org # v4.0+
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This way this device can be used with irtty-sir -
at least on Toshiba Satellite A20-S103 it is not configured by default
and needs PNP activation before it starts to respond on I/O ports.

This device has actually its own driver (ali-ircc),
but this driver seems to be non-functional for a very long time
(see http://permalink.gmane.org/gmane.linux.irda.general/484
http://permalink.gmane.org/gmane.network.protocols.obex.openobex.user/943
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=535070 ).
Signed-off-by: NMaciej Szmigiero <mail@maciej.szmigiero.name>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

SMSC IrCC SIR/FIR port should not be bound to by
(legacy) serial driver so its own driver (smsc-ircc2)
can bind to it.
Signed-off-by: NMaciej Szmigiero <mail@maciej.szmigiero.name>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Currently mxs-auart doesn't care correctly about the baud rate divisor.
According to reference manual the baud rate divisor must be between
0x000000EC and 0x003FFFC0. So calculate the possible baud rate range
and use it for uart_get_baud_rate().
Signed-off-by: NStefan Wahren <stefan.wahren@i2se.com>
Reviewed-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Whenever the UART device driver gets closed from userland, the driver
disables the UART unit and then stops the clocks to save power.

The bit which disabled the UART unit is described as:

 "UART Enable. If this bit is set to 1, the UART is enabled. Data
  transmission and reception occurs for the UART signals. When the
  UART is disabled in the middle of transmission or reception, it
  completes the current character before stopping."

The important part is the "it completes the current character". Whenever
a reception is ongoing when the UART gets disabled (including the clock
off) the statemachine freezes and "remembers" this state on the next
open() when re-enabling the unit's clock.

In this case we end up receiving an additional bogus character
immediately.

The solution in this change is to switch the AUART unit into its reset
state on close() and only release it from its reset state on the next
open().

Note: when the unit is also used as system console it is always 'in use',
so we cannot reset it on close().
Signed-off-by: NJuergen Borleis <jbe@pengutronix.de>
Reviewed-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This function clears the reset the AUART unit is in after system start
to make it work.
Signed-off-by: NJuergen Borleis <jbe@pengutronix.de>
Reviewed-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Preparation for uniform definition of ioremap, ioremap_wc, ioremap_wt,
and ioremap_cache, tree-wide.
Acked-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

Fix RS232/485 mode incorrect setting after S3/S4 resume for F81504/508/512

We had add RS232/485 RTS control with fecf27a3. But when it
resume from S3/S4, the mode register 0x40 + 0x08 * idx + 7 will
rewrite to 0x01 (RS232 mode).

This patch will modify 2 sections.

One is pci_fintek_init(), if it called when first init, it will
write mode register with 0x01. If it called from S3/S4 resume,
it's will get the relative port data and pass it to
pci_fintek_rs485_config() with NULL rs485 parameter.

The another modification is in pci_fintek_rs485_config(). It'll
re-apply old configuration when the parameter rs485 is NULL.
Signed-off-by: NPeter Hung <hpeter+linux_kernel@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

devtype structures are all declared as const.
Compiler does not complain because we cast
their pointers to save them in .driver_data.
Signed-off-by: NJakub Kicinski <kubakici@wp.pl>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

We currently register the uart driver during device probe
which makes it hard to support more than one chip.
Move the driver registration to module init/exit time and
preallocate space for up to 8 lines (4-8 chips).
Reported-by: NMichael Allwright <michael.allwright@upb.de>
Signed-off-by: NJakub Kicinski <kubakici@wp.pl>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

In preparation of supporting multiple devices we should save
the per-chip line number (0 or 1), because the uart_port line
will reflect system-wide uart number and will be offseted for
chips other than the first to register.
Signed-off-by: NJakub Kicinski <kubakici@wp.pl>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Support ZTE uart with some registers differing offset.
Probe as platform device for not AMBA IP ID is
available on ZTE uart.
Signed-off-by: NJun Nie <jun.nie@linaro.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Improve LCRH register access decision as ARM PL011 lcrh
register serve as both TX and RX, while other SOC may
implement TX and RX function with separated register.
Signed-off-by: NJun Nie <jun.nie@linaro.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Introduce register look up table as different SOC venders
may have different register offset for the some register.
Signed-off-by: NJun Nie <jun.nie@linaro.org>
Reviewed-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Introduce register accessor to ease loop up table access
in later patch.
Signed-off-by: NJun Nie <jun.nie@linaro.org>
Reviewed-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Rename regs with enumeration to generalize register names.
Signed-off-by: NJun Nie <jun.nie@linaro.org>
Reviewed-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NJun Nie <jun.nie@linaro.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Pericom PI7C9X795[1248] are Uno/Dual/Quad/Octal UART devices, this
patch enables them, also defines PCI_VENDOR_ID_PERICOM here.
Signed-off-by: NAdam Lee <adam.lee@canonical.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

So far DMA mode were activated when only number of bytes to send was
equal or greater than min_dma_size. Due to requirement that DMA transaction
buffer should be aligned to cache line size, the excessive bytes were
written to FIFO before starting DMA transaction. The problem occurred
when FIFO size were smaller than cache alignment, because writing all
excessive bytes to FIFO would fail. It happened in DMA mode when PIO
interrupts disabled, which caused driver hung.

The solution is to test if buffer is alligned to cache line size before
activating DMA mode, and if it's not, running PIO mode to align buffer
and then starting DMA transaction. In PIO mode, when interrupts are
enabled, lack of space in FIFO isn't the problem, so buffer aligning
will always finish with success.

Cc: <stable@vger.kernel.org> # v3.18+
Reported-by: NKrzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: NRobert Baldyga <r.baldyga@samsung.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Due to some of serial ports can have FIFO size smaller than cache line
size, and because of need to align DMA buffer address to cache line size,
it's necessary to calculate minimum number of bytes for which we want
to start DMA transaction to be at least cache line size. The simplest
way to meet this requirement is to get maximum of cache line size and
FIFO size.

Cc: <stable@vger.kernel.org> # v3.18+
Reported-by: NKrzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: NMarek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: NRobert Baldyga <r.baldyga@samsung.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

serial core does not use tty_port_open() or tty_port_close(); serial
core defines and extends it's own tty open() and close() methods
(uart_open() and uart_close(), respectively).

Remove the tty_port activate() and shutdown() initializations, and
the uart_port_activate() function, which is never called.

NB: uart_port_shutdown() is used by uart_close() => uart_shutdown()
call chain (but not via the tty_port methods).
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

uart_close() runs in non-atomic context only; use
spin_lock/unlock_irq instead of saving the interrupt state (which
== on).
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>