This is only an API consolidation and should make things more readable
it replaces var * HZ / 1000 by msecs_to_jiffies(var).
Signed-off-by: NNicholas Mc Guire <hofrat@osadl.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

UBRC is a read-only register, so we should not store and restore it inside
imx_flush_buffer().
Reported-by: NFugang Duan <B38611@freescale.com>
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Acked-by: NFugang Duan <B38611@freescale.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

this patch fixes following sparse warnings:

bcm63xx_uart.c:857:43: warning: Using plain integer as NULL pointer
bcm63xx_uart.c:871:35: warning: Using plain integer as NULL pointer
Signed-off-by: NLad, Prabhakar <prabhakar.csengg@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Without this the module does not load automatically whenever suitable
platform device appears.
Reported-by: NJerome Blin <jerome.blin@intel.com>
Signed-off-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Building sprd_serial.o when CONFIG_PM_SLEEP is not defined triggers
these warnings:
    drivers/tty/serial/sprd_serial.c:755:12: warning: ‘sprd_suspend’ defined but not used [-Wunused-function]
     static int sprd_suspend(struct device *dev)
                ^
    drivers/tty/serial/sprd_serial.c:764:12: warning: ‘sprd_resume’ defined but not used [-Wunused-function]
     static int sprd_resume(struct device *dev)
                ^

Let's compile these functions only when CONFIG_PM_SLEEP is defined.
Signed-off-by: NPaul Bolle <pebolle@tiscali.nl>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Make of_device_id array const.
Signed-off-by: NSanjeev Sharma <Sanjeev_Sharma@mentor.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

No functional changes.
Signed-off-by: NMichal Simek <michal.simek@xilinx.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Convert the legacy system PM callbacks to new ones. Meanwhile, remove the
redudant calls to the PCI for changing a power state since it's done by bus
code.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

In the current kernel, the CONFIG_PPC_OF is always 'y' for the ppc
arch. So we don't need to check it with other ppc specific options.
Signed-off-by: NKevin Hao <haokexin@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Use helper functions to access current->state.
Direct assignments are prone to races and therefore buggy.

Thanks to Peter Zijlstra for the exact definition of the problem.
Suggested-By: NPeter Zijlstra <peterz@infradead.org>
Signed-off-by: NFabian Frederick <fabf@skynet.be>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

As more ARM platforms are moving into ARCH_MULTIPLATFORM, we can now
have integrator and versatile in the same kernel, and first one selects
this driver, causing a Kconfig warning:

warning: (ARCH_INTEGRATOR_AP) selects SERIAL_AMBA_PL010 which has unmet direct dependencies (TTY && HAS_IOMEM && ARM_AMBA && (BROKEN || !ARCH_VERSATILE))

It turns out that it has not been broken on versatile for a long time,
so we can remove the statement here.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Commit 5695be14 ("OOM, PM: OOM killed task shouldn't escape PM
suspend") has left a race window when OOM killer manages to
note_oom_kill after freeze_processes checks the counter.  The race
window is quite small and really unlikely and partial solution deemed
sufficient at the time of submission.

Tejun wasn't happy about this partial solution though and insisted on a
full solution.  That requires the full OOM and freezer's task freezing
exclusion, though.  This is done by this patch which introduces oom_sem
RW lock and turns oom_killer_disable() into a full OOM barrier.

oom_killer_disabled check is moved from the allocation path to the OOM
level and we take oom_sem for reading for both the check and the whole
OOM invocation.

oom_killer_disable() takes oom_sem for writing so it waits for all
currently running OOM killer invocations.  Then it disable all the further
OOMs by setting oom_killer_disabled and checks for any oom victims.
Victims are counted via mark_tsk_oom_victim resp.  unmark_oom_victim.  The
last victim wakes up all waiters enqueued by oom_killer_disable().
Therefore this function acts as the full OOM barrier.

The page fault path is covered now as well although it was assumed to be
safe before.  As per Tejun, "We used to have freezing points deep in file
system code which may be reacheable from page fault." so it would be
better and more robust to not rely on freezing points here.  Same applies
to the memcg OOM killer.

out_of_memory tells the caller whether the OOM was allowed to trigger and
the callers are supposed to handle the situation.  The page allocation
path simply fails the allocation same as before.  The page fault path will
retry the fault (more on that later) and Sysrq OOM trigger will simply
complain to the log.

Normally there wouldn't be any unfrozen user tasks after
try_to_freeze_tasks so the function will not block. But if there was an
OOM killer racing with try_to_freeze_tasks and the OOM victim didn't
finish yet then we have to wait for it. This should complete in a finite
time, though, because

	- the victim cannot loop in the page fault handler (it would die
	  on the way out from the exception)
	- it cannot loop in the page allocator because all the further
	  allocation would fail and __GFP_NOFAIL allocations are not
	  acceptable at this stage
	- it shouldn't be blocked on any locks held by frozen tasks
	  (try_to_freeze expects lockless context) and kernel threads and
	  work queues are not frozen yet
Signed-off-by: NMichal Hocko <mhocko@suse.cz>
Suggested-by: NTejun Heo <tj@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Cong Wang <xiyou.wangcong@gmail.com>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

While touching this area let's convert printk to pr_*.  This also makes
the printing of continuation lines done properly.
Signed-off-by: NMichal Hocko <mhocko@suse.cz>
Acked-by: NTejun Heo <tj@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Cong Wang <xiyou.wangcong@gmail.com>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

UARTs which do not trigger THRE interrupt if the fifo is already
empty when the interrupt is enabled need tx primed manually. These
UARTs are identified by the UART_BUG_TXEN flag to enable the
required workaround.

However, the current workaround is broken; if the fifo is already
empty but the shifter is still transmitting, then serial8250_tx_chars()
will not be called but no further THRE interrupt will occur, and
tx will stall. The appropriate check is for fifo empty (THRE), not
transmitter empty (TEMT).
Signed-off-by: NDick Hollenbeck <dick@softplc.com>
[pjh: rewrote commit log]
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This is the last missing piece to get a kernel booting to a prompt in qemu-cris.
Signed-off-by: NNiklas Cassel <nks@flawful.org>
Acked-by: NJesper Nilsson <jesper.nilsson@axis.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

sprop is unused in find_console_handle() since commit a752ee56
("tty: Update hypervisor tty drivers to use core stdout parsing code.")
Signed-off-by: NFabian Frederick <fabf@skynet.be>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The general agreed way to specify a fixed line number
for a serial console is to provide a "serial" alias
in the devicetree. Start parsing this property in
of_serial.
Signed-off-by: NLucas Stach <dev@lynxeye.de>
Acked-by: NOlof Johansson <olof@lixom.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

EarlyCon support depends on serial console infrastructure, so the code
implementing it should depend on CONFIG_SERIAL_SAMSUNG_CONSOLE.

This patch fixes the following build break:
CC [M]  drivers/tty/serial/samsung.o
drivers/tty/serial/samsung.c:2468:1: error: expected declaration specifiers or ‘...’ before string constant
drivers/tty/serial/samsung.c:2468:1: error: expected declaration specifiers or ‘...’ before ‘s3c2410_setup_earlycon’
drivers/tty/serial/samsung.c:2487:1: error: expected declaration specifiers or ‘...’ before string constant
drivers/tty/serial/samsung.c:2487:1: error: expected declaration specifiers or ‘...’ before ‘s3c2412_setup_earlycon’
drivers/tty/serial/samsung.c:2488:1: error: expected declaration specifiers or ‘...’ before string constant
drivers/tty/serial/samsung.c:2488:1: error: expected declaration specifiers or ‘...’ before ‘s3c2440_setup_earlycon’
drivers/tty/serial/samsung.c:2489:1: error: expected declaration specifiers or ‘...’ before string constant
drivers/tty/serial/samsung.c:2489:1: error: expected declaration specifiers or ‘...’ before ‘s3c6400_setup_earlycon’
drivers/tty/serial/samsung.c:2506:1: error: expected declaration specifiers or ‘...’ before string constant
drivers/tty/serial/samsung.c:2506:1: error: expected declaration specifiers or ‘...’ before ‘s5pv210_setup_earlycon’
drivers/tty/serial/samsung.c:2507:1: error: expected declaration specifiers or ‘...’ before string constant
drivers/tty/serial/samsung.c:2507:1: error: expected declaration specifiers or ‘...’ before ‘exynos4210_setup_earlycon’
drivers/tty/serial/samsung.c:2468:1: warning: ‘s3c2410_setup_earlycon’ defined but not used [-Wunused-function]
drivers/tty/serial/samsung.c:2487:1: warning: ‘s3c2412_setup_earlycon’ defined but not used [-Wunused-function]
drivers/tty/serial/samsung.c:2488:1: warning: ‘s3c2440_setup_earlycon’ defined but not used [-Wunused-function]
drivers/tty/serial/samsung.c:2489:1: warning: ‘s3c6400_setup_earlycon’ defined but not used [-Wunused-function]
drivers/tty/serial/samsung.c:2506:1: warning: ‘s5pv210_setup_earlycon’ defined but not used [-Wunused-function]
drivers/tty/serial/samsung.c:2507:1: warning: ‘exynos4210_setup_earlycon’ defined but not used [-Wunused-function]
make[3]: *** [drivers/tty/serial/samsung.o] Error 1
Reported-by: Nkbuild test robot <fengguang.wu@intel.com>
Signed-off-by: NMarek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/tty/serial/8250/8250_core.c:2503:6: sparse: symbol 'serial8250_set_divisor' was not declared. Should it be static?
Signed-off-by: NFengguang Wu <fengguang.wu@intel.com>
Cc: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Add a full sc9836-uart driver for SC9836 SoC which is based on the
spreadtrum sharkl64 platform.
This driver also support earlycon.
Originally-by: NLanqing Liu <lanqing.liu@spreadtrum.com>
Signed-off-by: NOrson Zhai <orson.zhai@spreadtrum.com>
Signed-off-by: NChunyan Zhang <chunyan.zhang@spreadtrum.com>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Reviewed-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Function s3c24xx_serial_start_tx_pio() enables interrupts if needed,
so we don't have to (or even we shouldn't) enable them before.
Signed-off-by: NRobert Baldyga <r.baldyga@samsung.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

tty_set_termios() is an internal helper intended for file scope use.

UART drivers which are capable of driving the RTS pin must
properly handle the tiocmset() method, regardless of termios settings.
A failure to do so is a UART driver bug and should be fixed there.
Do not use this interface to workaround UART driver bugs.

Cc: Johan Hedberg <johan.hedberg@gmail.com>
Cc: <linux-bluetooth@vger.kernel.org>
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Acked-by: NMarcel Holtmann <marcel@holtmann.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The OMAP UART ignores MCR[1] (ie., RTS) when in autoRTS mode. This
makes it impossible for either the serial core or userspace to
manually flow control the sender.

Disable autoRTS mode when RTS is lowered and restore the previous
mode when RTS is raised.
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Commit 88838d3112702 ("serial: omap_8250: Fix RTS handling") fixed
RTS pin control when in autoRTS mode.

New support added in "serial: core: Rework hw-assisted flow control support"
enables a much simpler approach; rather than masking out autoRTS
whenever writing the EFR register, use the UPSTAT_* mode to determine if
autoRTS should be enabled when raising RTS (in omap8250_set_mctrl()).
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

hw-assisted flow control support was added to the serial core
in v3.8 with commits,
dba05832 ("SERIAL: core: add hardware assisted h/w flow control support")
2cbacafd ("SERIAL: core: add hardware assisted s/w flow control support")
9aba8d5b ("SERIAL: core: add throttle/unthrottle callbacks for hardware
                assisted flow control")
Since then, additional requirements for serial core support have arisen.
Specifically,
1. Separate tx and rx flow control settings for UARTs which only support
   tx flow control (ie., autoCTS).
2. Disable sw-assisted CTS flow control in autoCTS mode
3. Support for RTS flow control by serial core and userspace in autoRTS mode

Distinguish mode from capability; introduce UPSTAT_AUTORTS, UPSTAT_AUTOCTS
and UPSTAT_AUTOXOFF which, when set by the uart driver, enable serial core
support for hw-assisted rx, hw-assisted tx and hw-assisted in-band/IXOFF
rx flow control, respectively. [Note: hw-assisted in-band/IXON tx flow
control does not require serial core support/intervention and can be
enabled by the uart driver when required.]

These modes must be set/reset in the driver's set_termios() method, based
on termios settings, and thus can be safely queried in any context in which
one of the port lock, port mutex or termios rwsem are held. Set these modes
in the 2 in-tree drivers, omap-serial and 8250_omap, which currently
use UPF_HARD_FLOW/UPF_SOFT_FLOW support.

Retain UPF_HARD_FLOW and UPF_SOFT_FLOW as capabilities; re-define
UPF_HARD_FLOW as both UPF_AUTO_RTS and UPF_AUTO_CTS to allow for distinct
and separate rx and tx flow control capabilities.

Disable sw-assisted CTS flow control when UPSTAT_AUTOCTS is enabled.
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The PXA variant of the 8250 UART adds a UART enable bit which must not
be cleared. Make the earlycon support maintain this bit if it is set.
This implies some initialization of the UART, but we cannot
unconditionally set the bit as some other variants require this bit to
be clear for other functions.
Signed-off-by: NRob Herring <robh@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: linux-serial@vger.kernel.org
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Add mrvl,pxa-uart and mrvl,mmp-uart compatible strings for the of_serial
driver. These are 8250 variants which have a port type of PORT_XSCALE.

There is also the serial driver pxa.c with these compatible strings
already. However, it can be replaced with the common 8250 driver. It has
some issues like it cannot coexist with the 8250 driver due to tty name
collision. That also means adding these compatible strings here should
not case a problem.
Signed-off-by: NRob Herring <robh@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: linux-serial@vger.kernel.org
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Add support for alias parsing from the DT. This allows for consistent
tty numbering.
Signed-off-by: NRob Herring <robh@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: linux-serial@vger.kernel.org
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The port shadow registers, ->fcr and ->mcr, must be protected from
concurrent updates. Relocate the shadow register updates in
serial8250_do_set_termios() to the port lock critical section.
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

When using no_console_suspend, the serial console may be powered off
anyway during system sleep. Upon resume, the port may be in its default
power-on state, but is expected to continue console i/o before the device
has received its pm callback. The resultant garbage i/o can cause all
kinds of havoc on the remote end.

Use the scratch register as a canary to discover if the console
has been powered-off. Write a non-zero value to the scratch register
at port suspend and reprogram the port before any console i/o if the
scratch register != canary before port resume.

This workaround is disabled for omap_8250 (which uses different divisor
programming).

Credit to Doug Anderson <dianders@chromium.org> for the idea of using
the scratch register canary to discover port power-down.

Cc: Doug Anderson <dianders@chromium.org>
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Exar XR17V35X PCIe uarts support a 4-bit fractional divisor register.
Refactor the divisor calculation from the divisor programming.

Allow a fractional result from serial8250_get_divisor() and pass this
result to serial8250_dl_write().

Simplify the calculation for quot and quot_frac. This was verified
to be identical to the results of the original calculation with a test
jig.

NB: The results were also compared with the divisor value chart
on pg 33 of the Exar XR17V352 datasheet, rev 1.0.3, here:
http://www.exar.com/common/content/document.ashx?id=1585
which differs from the calculated values by 1 in the fractional result.
This is because the calculated values are still rounded in the
fractional result, whereas the table values are truncated. Note
that the data error rate % values in the datasheet are for
rounded fractional results, as the truncated fractional results
have more error.

Cc: Joe Schultz <jschultz@xes-inc.com>
Cc: Aaron Sierra <asierra@xes-inc.com>
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Refactor divisor register programming into a new function,
serial8250_set_divisor; this allows serial console to reinitialize
early after resume from suspend.
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Refactor the computation of the LCR register value from termios c_cflag
into a new local function, serial8250_compute_lcr().
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The UART_BUG_QUOT workaround adjusts the divisor computed from the
baud rate by serial8250_get_divisor(). Move the workaround into
serial8250_get_divisor(), so that divisor-from-baud computation
is encapsulated.
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

16z135 IP Core has changed so the driver needs to be updated to respect
these changes. The following changes have been made:

* Don't invert the 16z135 modem status register when reading.
* Add module parameter to configure the (baud rate dependent) RX timeout.
  Character timeout in seconds = (timeout_reg * baud_reg * 4)/freq_reg.
* Enable the handling of UART core's automatic flow control feature.
  When AFE is active disable generation of modem status IRQs.
* Rework the handling of IRQs to be conform with newer FPGA versions and
  take precautions not to miss an interrupt because of the destructive read
  of the IIR register.
* Correct men_z135_handle_modem_status(), MSR is stat_reg[15:8] not
  stat_reg[7:0]
* Correct calling of uart_handle_{dcd,cts}_change()
* Reset CLOCAL when CRTSCTS is set
Signed-off-by: NJohannes Thumshirn <johannes.thumshirn@men.de>
Reviewed-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

When the uart port being suspended is a console and consoles are
not suspending (kernel command line contains no_console_suspend),
then no action is performed for that port, and the function can
return early.

If the function has not returned early, then one of the conditions
is not true, so the expression
   (console_suspend_enabled || !uart_console(uport))
must be true and can be eliminated.

Similarly, the expression
   (console_suspend_enabled && uart_console(uport))
simplifies to just uart_console(uport).
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

BRKINT and ISIG requires input and output flush when a signal char
is received. However, the order of operations is significant since
parallel i/o may be ongoing.

Merge the signal handling for BRKINT with ISIG handling.

Process the signal first. This ensures any ongoing i/o is aborted;
without this, a waiting writer may continue writing after the flush
occurs and after the signal char has been echoed.

Write lock the termios_rwsem, which excludes parallel writers from
pushing new i/o until after the output buffers are flushed; claiming
the write lock is necessary anyway to exclude parallel readers while
the read buffer is flushed.

Subclass the termios_rwsem for ptys since the slave pty performing
the flush may appear to reorder the termios_rwsem->tty buffer lock
lock order; adding annotation clarifies that
  slave tty_buffer lock-> slave termios_rwsem -> master tty_buffer lock
is a valid lock order.

Flush the echo buffer. In this context, the echo buffer is 'output'.
Otherwise, the output will appear discontinuous because the output buffer
was cleared which contains older output than the echo buffer.

Open-code the read buffer flush since the input worker does not need
kicking (this is the input worker).
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The pty driver does not clear its write buffer when commanded.
This is to avoid an apparent deadlock between parallel flushes from
both pty ends; specifically when handling either BRK or INTR input.
However, parallel flushes from this source is not possible since
the pty master can never be set to BRKINT or ISIG. Parallel flushes
from other sources are possible but these do not threaten deadlocks.

Annotate the tty buffer mutex for lockdep to represent the nested
tty_buffer locking which occurs when the pty slave is processing input
(its buffer mutex held) and receives INTR or BRK and acquires the
linked tty buffer mutex via tty_buffer_flush().
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Besides nested legacy_mutex locking which is required on pty pair
teardown, other nested pty operations require lock subclassing.

Move lock subclass definition to tty interface header, include/linux/tty.h,
and document its use.
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

In canon mode, the read buffer head will advance over the buffer tail
if the input > 4095 bytes without receiving a line termination char.

Discard additional input until a line termination is received.
Before evaluating for overflow, the 'room' value is normalized for
I_PARMRK and 1 byte is reserved for line termination (even in !icanon
mode, in case the mode is switched). The following table shows the
transform:

 actual buffer |  'room' value before overflow calc
  space avail  |    !I_PARMRK    |    I_PARMRK
 --------------------------------------------------
      0        |       -1        |       -1
      1        |        0        |        0
      2        |        1        |        0
      3        |        2        |        0
      4+       |        3        |        1

When !icanon or when icanon and the read buffer contains newlines,
normalized 'room' values of -1 and 0 are clamped to 0, and
'overflow' is 0, so read_head is not adjusted and the input i/o loop
exits (setting no_room if called from flush_to_ldisc()). No input
is discarded since the reader does have input available to read
which ensures forward progress.

When icanon and the read buffer does not contain newlines and the
normalized 'room' value is 0, then overflow and room are reset to 1,
so that the i/o loop will process the next input char normally
(except for parity errors which are ignored). Thus, erasures, signalling
chars, 7-bit mode, etc. will continue to be handled properly.

If the input char processed was not a line termination char, then
the canon_head index will not have advanced, so the normalized 'room'
value will now be -1 and 'overflow' will be set, which indicates the
read_head can safely be reset, effectively erasing the last char
processed.

If the input char processed was a line termination, then the
canon_head index will have advanced, so 'overflow' is cleared to 0,
the read_head is not reset, and 'room' is cleared to 0, which exits
the i/o loop (because the reader now have input available to read
which ensures forward progress).

Note that it is possible for a line termination to be received, and
for the reader to copy the line to the user buffer before the
input i/o loop is ready to process the next input char. This is
why the i/o loop recomputes the room/overflow state with every
input char while handling overflow.

Finally, if the input data was processed without receiving
a line termination (so that overflow is still set), the pty
driver must receive a write wakeup. A pty writer may be waiting
to write more data in n_tty_write() but without unthrottling
here that wakeup will not arrive, and forward progress will halt.
(Normally, the pty writer is woken when the reader reads data out
of the buffer and more space become available).
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>