I wrote this driver and use it daily on several machines for work, so
why not.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Acked-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

We don't really need to perform the ioremap "on demand" so it's simpler
just to do it from the probe function.  This also lets us eliminate the
UART_REG_SIZE constant and rely on the resource information passed in
from the DT or platform code.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This enables early console output if there is a chosen/stdout-path
property referencing a UART node with the "brcm,bcm6345-uart" compatible
string.  The bootloader sets up the pinmux and baud/parity/etc.
Tested on bcm3384 (MIPS, DT).
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This squashes a checkpatch warning on my new bcm3384 dts submission.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove the extra '<' character.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove incorrect "bcm963xx_uart" module name; add a list of known users;
tweak grammar/indentation/capitalization.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Acked-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The original non-DT bcm63xx clk code ignores the struct device argument
and looks up a global clock name.  DT platforms, by contrast, often just
use a phandle to reference a clock node with no "clock-output-names"
property.  Modify the UART driver to support both schemes.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This device was originally supported on bcm63xx only, but it shows up on
a wide variety of MIPS and ARM chipsets spanning multiple product lines.
Now that the driver has eliminated dependencies on bcm63xx-specific
header files, we can build it on any non-bcm63xx kernel.

Compile-tested on x86, both statically and as a module.  Tested for
functionality on bcm3384 (a new MIPS platform under active development).
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Acked-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

For the console poll usage, .poll_init() will perform deeper hardware
initialization to ensure the serial port is always active.
Signed-off-by: NDaniel Thompson <daniel.thompson@linaro.org>
Signed-off-by: NDirk Behme <dirk.behme@de.bosch.com>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: Huang Shijie <b32955@freescale.com>
Cc: linux-serial@vger.kernel.org
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

imx_put_poll_char() has been simplified to remove the code to disable
interrupts. The present code can corrupt register state when re-entered
from FIQ handler.

Switch to _relaxed() MMIO functions (which are safe for polled I/O and
needed to avoid taking spin locks).
Signed-off-by: NDaniel Thompson <daniel.thompson@linaro.org>
Signed-off-by: NDirk Behme <dirk.behme@de.bosch.com>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: Huang Shijie <b32955@freescale.com>
Cc: linux-serial@vger.kernel.org
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

We rely on probe order of this driver to determine the line number for
the uart port. This makes it impossible to know the line number
when these devices are populated via DT. Use the DT alias
mechanism to assign the line based on the aliases node.
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

We need to issue a reset if we ever change the value of the IPR
register on DM hardware. If we don't reset the hardware the RX
stale interrupt never triggers and the only way to trigger an RX
handling event is by filling up the fifo. This causes things like
getty to not work so well considering it might change the baud
rate a few times. Fix this by moving the reset on startup and any
reprogramming required after the reset to be after we change the
baud rate.
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

To properly support sysrq on uartDM hardware we need to properly
handle break characters. With the DM hardware the fifo can pack 4
characters at a time, where a break is indicated by an all zero
byte. Unfortunately, we can't differentiate between an all zero
byte for a break and an all zero byte of data, so try and do as
best we can. First unmask the RX break start interrupt and record
the interrupt when it arrives. Then while processing the fifo,
detect the break by searching for an all zero character as long
as we recently received an RX break start interrupt. This should
make sysrq work fairly well.

Cc: Frank Rowand <frank.rowand@sonymobile.com>
Cc: Daniel Thompson <daniel.thompson@linaro.org>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The handle_rx() path calls uart_handle_sysrq_char() with the port
lock held. This causes a spinlock recursion. Release and
reacquire the lock here to avoid this.

BUG: spinlock recursion on CPU#0, swapper/0
 lock: msm_uart_ports+0x1e0/0x2d0, .magic: dead4ead, .owner: swapper/0, .owner_cpu: 0
CPU: 0 PID: 0 Comm: swapper Not tainted 3.17.0-rc7-00012-gb38ee8265941 #69
[<c0013964>] (unwind_backtrace) from [<c0011f74>] (show_stack+0x10/0x14)
[<c0011f74>] (show_stack) from [<c004ed1c>] (do_raw_spin_lock+0x11c/0x13c)
[<c004ed1c>] (do_raw_spin_lock) from [<c02d44c0>] (msm_console_write+0x78/0x188)
[<c02d44c0>] (msm_console_write) from [<c0052880>] (call_console_drivers.constprop.22+0xb4/0x144)
[<c0052880>] (call_console_drivers.constprop.22) from [<c0053570>] (console_unlock+0x27c/0x4ac)
[<c0053570>] (console_unlock) from [<c0053bb4>] (vprintk_emit+0x1f4/0x5a8)
[<c0053bb4>] (vprintk_emit) from [<c04ad0ac>] (printk+0x30/0x40)
[<c04ad0ac>] (printk) from [<c02c2990>] (__handle_sysrq+0x58/0x1b8)
[<c02c2990>] (__handle_sysrq) from [<c02d41b0>] (msm_irq+0x694/0x6f8)
[<c02d41b0>] (msm_irq) from [<c0055740>] (handle_irq_event_percpu+0x58/0x270)
[<c0055740>] (handle_irq_event_percpu) from [<c0055994>] (handle_irq_event+0x3c/0x5c)
[<c0055994>] (handle_irq_event) from [<c0057e84>] (handle_level_irq+0x9c/0x138)
[<c0057e84>] (handle_level_irq) from [<c005509c>] (generic_handle_irq+0x24/0x38)
[<c005509c>] (generic_handle_irq) from [<c000f730>] (handle_IRQ+0x44/0xb0)
[<c000f730>] (handle_IRQ) from [<c0008518>] (msm_vic_handle_irq+0x44/0x64)
[<c0008518>] (msm_vic_handle_irq) from [<c04b5ac4>] (__irq_svc+0x44/0x7c)
Exception stack(0xc0719f68 to 0xc0719fb0)
9f60:                   00000001 00000001 00000000 c0722938 c0718000 c0769acc
9f80: 00000000 c0720098 c0769305 4117b362 c0769acc 00000000 01000000 c0719fb0
9fa0: c004cab0 c000f880 20000013 ffffffff
[<c04b5ac4>] (__irq_svc) from [<c000f880>] (arch_cpu_idle+0x20/0x30)
[<c000f880>] (arch_cpu_idle) from [<c004691c>] (cpu_startup_entry+0xf4/0x23c)
[<c004691c>] (cpu_startup_entry) from [<c06d8b70>] (start_kernel+0x32c/0x394)

Cc: Frank Rowand <frank.rowand@sonymobile.com>
Cc: Daniel Thompson <daniel.thompson@linaro.org>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Enables PPS support in mxs-auart serial driver to make PPS API working.
Signed-off-by: NJanusz Uzycki <j.uzycki@elproma.com.pl>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Handle CTS/DSR/RI/DCD GPIO interrupts in mxs-auart.
Signed-off-by: NJanusz Uzycki <j.uzycki@elproma.com.pl>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Dedicated CTS and RTS pins are unusable together with a lot of other
peripherals because they share the same line. Pinctrl is limited.

Moreover, the AUART controller doesn't handle DTR/DSR/DCD/RI signals,
so we have to control them via GPIO.

This patch permits to use GPIOs to control the CTS/RTS/DTR/DSR/DCD/RI
signals.
Signed-off-by: NJanusz Uzycki <j.uzycki@elproma.com.pl>
Reviewed-by: NRichard Genoud <richard.genoud@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Russell King:
The only thing which the .get_mctrl method is supposed to do is
to return the state of the /input/ lines, which are CTS, DCD, DSR, RI.
The output line state is stored in port->mctrl, and is added to
the returned value by serial_core when it's required.
RTS output state should not be returned from the .get_mctrl method.

This patch removes ctrl variable from mxs_auart_port
and removes useless reading back RTS line.

The ctrl variable in mxs_auart_port duplicated mctrl,
member of uart_port structure in serial_core.h.
The removed code from mxs_auart_set_mctrl() and mxs_auart_get_mctrl
duplicated uart_update_mctrl() and uart_tiocmget()
in serial_core.c.
Signed-off-by: NJanusz Uzycki <j.uzycki@elproma.com.pl>
Reviewed-by: NRichard Genoud <richard.genoud@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch swaps the use "tail" and "head" to fit the semantic of the linux
circular buffer documentation:
- head: the point at which the producer (the DMA controller) inserts items.
- tail: the point at which the consumer (the serial framework) finds the next
        item.

Besides the former code of the rx ring buffer didn't manage the case where
head < tail, which might lead to loss of data. To fix this bug the data are now
sent from the DMA buffer to the serial framework in two steps:
1 - First, we test if head < tail. If so, we copy the data from tail to the end
    of the DMA buffer then reset tail to zero.
2 - Finally, we copy data from tail to head then set tail to head.

In addition, since tty_insert_flip_string() may now be called twice,
atmel_flip_buffer_rx_dma() becomes less efficient than moving the calls
dma_sync_sg_for_cpu(), dma_sync_sg_for_device(), tty_insert_flip_string() and
tty_flip_buffer_push() directly into atmel_rx_from_dma().
Signed-off-by: NCyrille Pitchen <cyrille.pitchen@atmel.com>
Acked-by: NNicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Add the config symbol for Mediatek MT7620 SoC to the SERIAL_8250_RT288X
dependencies.
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The Ralink RT2880 SoC and its successors have an internal 8250 core. This core
needs the same quirks applied as the AMD AU1xxx uart. In addition to these
quirks, the ports memory region is only 0x100 unlike the AU1xxx which has a
size of 0x1000.
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Increase the maximum number of consoles possible to 10 since DRA7 now
has the maximum number of consoles possible. without doing this, for
example, enabling DRA7 UART10 results in internal data structures and
console cannot match up and we endup with a crash as follows:

[    1.903503] omap_uart 48424000.serial: [UART-1]: failure [serial_omap_probe]: -22
[    1.911437] omap_uart: probe of 48424000.serial failed with error -22
[    1.920379] Unable to handle kernel NULL pointer dereference at virtual address 00000004
[    1.928894] pgd = c0004000
[    1.931732] [00000004] *pgd=00000000
[    1.935485] Internal error: Oops: 5 [#1] SMP ARM
[    1.940338] Modules linked in:
[    1.943542] CPU: 1 PID: 12 Comm: kworker/1:0 Tainted: G        W      3.18.0-rc1-00001-g8821bc4-dirty #6
[    1.953521] task: ed8a2d00 ti: ed8a4000 task.ti: ed8a4000
[    1.959197] PC is at process_one_work+0x38/0x4a4
[    1.964050] LR is at 0x0
[    1.966705] pc : [<c00548e0>]    lr : [<00000000>]    psr: 40000093
[    1.966705] sp : ed8a5ea8  ip : ed8a5ec8  fp : eeb9abc0
[    1.978759] r10: ed8a4000  r9 : 00000008  r8 : ed842458
[    1.984252] r7 : 00000000  r6 : eeb9abc0  r5 : ed842440  r4 : edbf26a8
[    1.991119] r3 : 00000000  r2 : 00000000  r1 : 00000000  r0 : 00000000
[    1.997985] Flags: nZcv  IRQs off  FIQs on  Mode SVC_32  ISA ARM  Segment kernel
[    2.005767] Control: 10c5387d  Table: 8000406a  DAC: 00000015
[    2.011810] Process kworker/1:0 (pid: 12, stack limit = 0xed8a4248)
[    2.018371] Stack: (0xed8a5ea8 to 0xed8a6000)
[    2.022949] 5ea0:                   00000001 00000000 60000093 c008346c 00000001 ed8a5ec8
[    2.031555] 5ec0: c0054de4 eeb9abc0 ed8a4000 ed842458 00000008 ed8a4000 eeb9abc0 ed842440
[    2.040161] 5ee0: eeb9abc0 eeb9abf0 ed8a4000 ed842458 00000008 ed8a4000 eeb9abc0 c0054ec4
[    2.048767] 5f00: ed8a4000 eeb9ac4c a0000053 00000000 ed845bc0 ed842440 c0054d80 00000000
[    2.057342] 5f20: 00000000 00000000 00000000 c005999c 00000001 00000000 c05eaa64 ed842440
[    2.065948] 5f40: 00000000 00000000 dead4ead ffffffff ffffffff c097c680 00000000 00000000
[    2.074554] 5f60: c078acd4 ed8a5f64 ed8a5f64 00000000 00000000 dead4ead ffffffff ffffffff
[    2.083160] 5f80: c097c680 00000000 00000000 c078acd4 ed8a5f90 ed8a5f90 600000d3 ed845bc0
[    2.091766] 5fa0: c00598d4 00000000 00000000 c000e668 00000000 00000000 00000000 00000000
[    2.100341] 5fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[    2.108947] 5fe0: 00000000 00000000 00000000 00000000 00000013 00000000 90005148 11010482
[    2.117553] [<c00548e0>] (process_one_work) from [<c0054ec4>] (worker_thread+0x144/0x498)
[    2.126159] [<c0054ec4>] (worker_thread) from [<c005999c>] (kthread+0xc8/0xe4)
[    2.133758] [<c005999c>] (kthread) from [<c000e668>] (ret_from_fork+0x14/0x2c)
[    2.141357] Code: e1a04001 e1a05000 e893000f e31e0004 (e5978004)
[    2.147766] ---[ end trace 5798e2803311b69f ]---
<snip>

The final solution is to transition off to use 8250 driver and no
dependency on console structures and move away from omap-serial
driver, hence no major cleanups are done for this driver, so add a
simple check to prevent future crashes in a similar manner in case of
platform with additional ports.
Signed-off-by: NNishanth Menon <nm@ti.com>
Acked-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Commit 479e9b94 ("serial: Refactor uart_flush_buffer() from
uart_close") refactored the uart_flush_buffer() in uart_close() into
those drivers that define a flush_buffer() method. However, in the
amba-pl011 driver configured without CONFIG_DMA_ENGINE, flush_buffer()
is a NULL method, so the direct call fails to compile.

Check and call the flush_buffer() method through the ops table instead.
Reported-by: NFengguang Wu <fengguang.wu@intel.com>
Cc: Russell King <linux@arm.linux.org.uk>
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch fixes sparse warning:
"obsolete struct initializer, use C99 syntax"
Signed-off-by: NPeng Fan <van.freenix@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

devm_kzalloc() already warns if allocation fails; remove duplicate
message.
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

vt_get_kmsg_redirect() only has meaning to the console driver as
an alias for calling vt_kmsg_redirect(). Move the macro definition
to the only source file which uses it; remove from uapi/linux/vt.h
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The last vestige of ASYNC_SPLIT_TERMIOS was removed by commit
'cris: Remove obsolete ASYNC_SPLIT_TERMIOS behavior'. Mark the flag
as defunct in the uapi header.
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

ASYNC_SPLIT_TERMIOS behavior is a remnant of the long-dead /dev/cuaXX
callout device. Split termios handling was removed tree-wide in v2.5.71 by:

commit 99a21edebbfd8c29e39ee7fcc8a1ffa423657290
Author: Alexander Viro <viro@www.linux.org.uk>
Date:   Wed Jun 11 07:41:28 2003 -0700

    [PATCH] tty_driver refcounting

    killed the last remnants of callout stuff - we don't need to mess with
    storing termios privately anymore.

which pre-dated the re-introduction into the cris serial driver
in v2.6.7 by:

commit 311a5ffeda8ccb3f1f3840069f37234e043092d4
Author: Andrew Morton <akpm@osdl.org>
Date:   Mon May 31 18:52:29 2004 -0700

    [PATCH] CRIS architecture update

    From: "Mikael Starvik" <mikael.starvik@axis.com>

    - Lots of fixes from 2.4.

    - Updated for 2.6.6.

    - Added IDE driver

Cc: Mikael Starvik <starvik@axis.com>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

setup_serial_console() is obsolete and has been superseded by
early_serial_setup() which is called at the end of the function.

The IA64 arch does not call this function; only the m68k arch setup
calls this function.
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

serial_console_init() is not defined by the tty core; remove
declaration.

Note that the powerpc arch boot code contains a serial_console_init()
declaration in arch/powerpc/boot/ops.h which is restricted to
the powerpc arch boot.
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

pcxe_open() has no definition in mainline; remove declaration.
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The use of older function ptr calling style, (*fn)(), makes static
analysis more error-prone; replace with modern fn() style.
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

tty_hung_up_p() is equivalent to the open-coded test in tty_open().
Signed-off-by: NPeter Hurley <peter@hurleysoftware.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Precedence of & and >> is not the same and is not left to right.
shift has higher precedence and should be done after the mask.

Add parentheses around the masks.
Signed-off-by: NJoe Perches <joe@perches.com>
Acked-by: NDavid Sterba <dsterba@suse.cz>
Reviewed-by: NJiri Kosina <jkosina@suse.cz>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

There is no real value in displaying "Serial: IMX driver" in every boot.

The uart_register_driver() can fail and even so the "Serial: IMX driver" will
be displayed, which is not really helpful.

This is particularly annoying when booting multi_v7_defconfig kernel on a SoC
that is not a i.MX and even though this message gets displayed.
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

When kzalloc() fails the core MM will already complain about it, so there is
no need to have the error message locally.

Remove the unneeded error message.
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Instead of jumping to 'error_out1' label we can simplify the code and return the
error code directly.
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

On a imx system with ttymxc0, ttymxc1 and ttymxc4 registered we see the
following output from 'cat /proc/interrupts':

$ cat /proc/interrupts
           CPU0
...
 58:         39       GIC  58  2020000.serial
 67:        115       GIC  67  21f8000.i2c

The only uart irq that appears is ttymxc0, which is the console.

As ttymxc1 and ttymxc4 will only have their irq registered at imx_startup(),
they are not shown right after probe.

Transmitting to ttymxc1 and ttymxc4 will cause their irqs to be registered, but
the output shows:

$ echo "111111" > /dev/ttymxc1
$ echo "444444" > /dev/ttymxc4
$ cat /proc/interrupts
           CPU0
...
 58:        150       GIC  58  2020000.serial
 59:          1       GIC  59
 62:          1       GIC  62
 67:        115       GIC  67  21f8000.i2c

,which misses printing the associated device address.

In order to fix this, register all the irqs inside the probe function via
devm_request_irq(), which will correctly report the serial interrupts associated
with their correspondent serial device and also helps simplyfing the code by
avoiding the calls to free_irq().

$ echo "111111" > /dev/ttymxc1
$ echo "444444" > /dev/ttymxc4
$ cat /proc/interrupts

	   CPU0
....
 58:        202       GIC  58  2020000.serial
 59:          1       GIC  59  21e8000.serial
 62:          1       GIC  62  21f4000.serial
 67:        115       GIC  67  21f8000.i2c
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

For the case in which we just provide an address as an argument to the
earlycon console type like:

earlycon=msm_serial_dm,0xf991e000

We would report this as an IO port based mapping and not as MMIO.  Simple
fix to use the port->iotype to decide which message to print.
Signed-off-by: NKumar Gala <galak@codeaurora.org>
Acked-by: NMark RUtland <mark.rutland@arm.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>