Changing the deprecated custom_divisor field is an unprivileged
operation so after verifying that flag field does not contain any
privileged changes both updates can be carried out by any user.

Combine the two branches and drop the erroneous comment.

Note that private flags field is only used for ASYNC flags so there's no
need to try to retain any other bits when updating the flags.
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

The TIOCSSERIAL error handling is inconsistent at best, but drivers tend
to ignore requests to change parameters which cannot be changed rather
than return an error.

The FTDI driver ignores change requests for all immutable parameters but
baud_base so return success also in this case for consistency.
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

The TIOCGSERIAL ioctl can be used to set and retrieve the UART type for
legacy UARTs, but some USB serial drivers have been reporting back
random types in order to "make user-space happy".

Some applications have historically expected TIOCGSERIAL to be
implemented, but judging from the Debian sources, the port type not
being PORT_UNKNOWN is only used to check for the existence of legacy
serial ports (ttySn).

Drivers like ftdi_sio have been using PORT_UNKNOWN for twenty years (and
option for 10 years) without anyone complaining so let's stop reporting
back anything else.

In the unlikely event that this do cause problems, this should be fixed
tree-wide anyway (e.g. for all USB serial drivers and also CDC-ACM).
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The closing_wait parameter determines how long to wait for the transfer
buffers to drain during close and the default timeout of 30 seconds may
not be sufficient at low line speeds. In other cases, when for example
flow is stopped, the default timeout may instead be too long.

Add generic support for TIOCSSERIAL and TIOCGSERIAL with handling of the
three common parameters close_delay, closing_wait and line for the
benefit of all USB serial drivers while still allowing drivers to
implement further functionality through the existing callbacks.

This currently includes a few drivers that report their base baud clock
rate even if that is really only of interest when setting custom
divisors through the deprecated ASYNC_SPD_CUST interface; an interface
which only the FTDI driver actually implements.

Some drivers have also been reporting back a fake UART type, something
which should no longer be needed and will be dropped by a follow-on
patch.
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

Drivers should return -ENOTTY ("Inappropriate I/O control operation")
when an ioctl isn't supported, while -EINVAL is used for invalid
arguments.

Fix up the TIOCMGET, TIOCMSET and TIOCGICOUNT helpers which returned
-EINVAL when a USB serial driver did not implement the corresponding
methods.

Note that the TIOCMGET and TIOCMSET helpers predate git and do not get a
corresponding Fixes tag below.

Fixes: d281da7f ("tty: Make tiocgicount a handler")
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The port parameter is used to set the I/O port and does not make any
sense to use for USB serial devices.

The xmit_fifo_size parameter could be used to set the hardware transmit
fifo size of a legacy UART when it could not be detected, but the
interface is limited to eight bits and should be left unset when not
used.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds (not jiffies). The driver does not yet support
changing these, but let's report back the default values actually used
(0.5 and 30 seconds, respectively).
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The port parameter is used to set the I/O port and does not make any
sense to use for USB serial devices.

The baud_base parameter could be used to set the UART base clock when it
could not be detected but might as well be left unset when it is not
known.

Fix the usb_wwan TIOCGSERIAL implementation by dropping its custom
interpretation of the unused port and baud_base fields, which were set
to the port index and current line speed, respectively.

Fixes: 02303f73 ("usb-wwan: implement TIOCGSERIAL and TIOCSSERIAL to avoid blocking close(2)")
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

A non-privileged user has only ever been able to set the since long
deprecated ASYNC_SPD flags and trying to change any other *supported*
feature should result in -EPERM being returned. Setting the current
values for any supported features should return success.

Fix the usb_wwan implementation which instead indicated that the
TIOCSSERIAL ioctl was not even implemented when a non-privileged user
set the current values.

Fixes: 02303f73 ("usb-wwan: implement TIOCGSERIAL and TIOCSSERIAL to avoid blocking close(2)")
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

The port close_delay and closing_wait parameters set by TIOCSSERIAL are
specified in jiffies and not milliseconds.

Add the missing conversions so that the TIOCSSERIAL works as expected
also when HZ is not 1000.

Fixes: 02303f73 ("usb-wwan: implement TIOCGSERIAL and TIOCSSERIAL to avoid blocking close(2)")
Cc: stable@vger.kernel.org      # 2.6.38
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

Changing the port closing-wait parameter is a privileged operation so
make sure to return -EPERM if a regular user tries to change it.

Cc: stable@vger.kernel.org
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The port parameter is used to set the I/O port and does not make any
sense to use for USB serial devices.

The xmit_fifo_size parameter could be used to set the hardware transmit
fifo size of a legacy UART when it could not be detected, but the
interface is limited to eight bits and should be left unset when not
used.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds. The driver does not yet support changing
close_delay, but let's report back the default value actually used (0.5
seconds).
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The xmit_fifo_size parameter could be used to set the hardware transmit
fifo size of a legacy UART when it could not be detected, but the
interface is limited to eight bits and should be left unset when not
used.

Similarly, baud_base could be used to set the UART base clock when it
could not be detected but might as well be left unset when it is not
known.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds (not jiffies). The driver does not yet support
changing these, but let's report back the default values actually used
(0.5 and 30 seconds, respectively).

Fixes: 52af9545 ("USB: add USB serial ssu100 driver")
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The xmit_fifo_size parameter could be used to set the hardware transmit
fifo size of a legacy UART when it could not be detected, but the
interface is limited to eight bits and should be left unset when not
used.

Similarly, baud_base could be used to set the UART base clock when it
could not be detected but might as well be left unset when it is not
known.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds (not jiffies). The driver does not yet support
changing these, but let's report back the default values actually used
(0.5 and 30 seconds, respectively).

Fixes: f7a33e60 ("USB: serial: add quatech2 usb to serial driver")
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The port parameter is used to set the I/O port and does not make any
sense to use for USB serial devices.

The baud_base parameter could be used to set the UART base clock when it
could not be detected but might as well be left unset when it is not
known.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds. The driver does not yet support changing
these, but let's report back the default values actually used (0.5 and
30 seconds, respectively).
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The xmit_fifo_size parameter could be used to set the hardware transmit
fifo size of a legacy UART when it could not be detected, but the
interface is limited to eight bits and should be left unset when not
used.

Similarly, baud_base could be used to set the UART base clock when it
could not be detected but might as well be left unset when it is not
known.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds (not jiffies). The driver does not yet support
changing these, but let's report back the default values actually used
(0.5 and 30 seconds, respectively).

Fixes: faac64ad ("USB: serial: opticon: add serial line ioctls")
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The port parameter is used to set the I/O port and does not make any
sense to use for USB serial devices.

The xmit_fifo_size parameter could be used to set the hardware transmit
fifo size of a legacy UART when it could not be detected, but the
interface is limited to eight bits and should be left unset when not
used.

Similarly, baud_base could be used to set the UART base clock when it
could not be detected but might as well be left unset when it is not
known.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds (not jiffies). The driver does not yet support
changing these, but let's report back the default values actually used
(0.5 and 30 seconds, respectively).

Fixes: 3f542974 ("USB: Moschip 7840 USB-Serial Driver")
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The port parameter is used to set the I/O port and does not make any
sense to use for USB serial devices.

The xmit_fifo_size parameter could be used to set the hardware transmit
fifo size of a legacy UART when it could not be detected, but the
interface is limited to eight bits and should be left unset when not
used.

Similarly, baud_base could be used to set the UART base clock when it
could not be detected but might as well be left unset when it is not
known.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds (not jiffies). The driver does not yet support
changing these, but let's report back the default values actually used
(0.5 and 30 seconds, respectively).

Fixes: 0f64478c ("USB: add USB serial mos7720 driver")
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The port parameter is used to set the I/O port and does not make any
sense to use for USB serial devices.

The xmit_fifo_size parameter could be used to set the hardware transmit
fifo size of a legacy UART when it could not be detected, but the
interface is limited to eight bits and should be left unset when not
used.

Similarly, baud_base could be used to set the UART base clock when it
could not be detected but might as well be left unset when it is not
known.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds (not jiffies). The driver does not yet support
changing close_delay, but let's report back the default value actually
used (0.5 seconds).
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The port parameter is used to set the I/O port and does not make any
sense to use for USB serial devices.

The xmit_fifo_size parameter could be used to set the hardware transmit
fifo size of a legacy UART when it could not be detected, but the
interface is limited to eight bits and should be left unset when not
used.

Similarly, baud_base could be used to set the uart base clock when it
could not be detected, but might as well be left unset when it is not
known.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds (not jiffies). The driver does not yet support
changing these, but let's report back the default values actually used
(0.5 and 30 seconds, respectively).
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The FTDI driver is the only USB serial driver supporting the deprecated
ASYNC_SPD flags, which are reported back as they should by TIOCGSERIAL,
but the returned parameters did not include the line number.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds. The driver does not yet support changing
these, but let's report back the default values actually used (0.5 and
30 seconds, respectively).
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The port parameter is used to set the I/O port and does not make any
sense to use for USB serial devices.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds. The driver does not yet support changing
these, but let's report back the default values actually used (0.5 and
30 seconds, respectively).

Fixes: aac1fc38 ("USB: serial: add Fintek F81232 usb to serial driver")
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The port parameter is used to set the I/O port and does not make any
sense to use for USB serial devices.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds. The driver does not yet support changing
these, but let's report back the default values actually used (0.5 and
30 seconds, respectively).

Fixes: aac1fc38 ("USB: serial: add Fintek F81232 usb to serial driver")
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

TIOCSSERIAL is a horrid, underspecified, legacy interface which for most
serial devices is only useful for setting the close_delay and
closing_wait parameters.

The port parameter is used to set the I/O port and does not make any
sense to use for USB serial devices.

The baud_base parameter could be used to set the UART base clock when it
could not be detected but might as well be left unset when it is not
known.

The close_delay and closing_wait parameters returned by TIOCGSERIAL are
specified in centiseconds. The driver does not yet support changing
these, but let's report back the default values actually used (0.5 and
30 seconds, respectively).

Fixes: 2f430b4b ("USB: ark3116: Add TIOCGSERIAL and TIOCSSERIAL ioctl calls.")
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

Use the new multi-interface support in USB serial core to properly claim
also the control interface during probe. This prevents having another
driver claim the control interface and makes core allocate resources
also for the interrupt endpoint (currently unused).

Switch to probing only Communication Class interfaces and use the Union
functional descriptor to determine the corresponding data interface.
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

A single USB function can be implemented using a group of interfaces and
this is for example commonly used for Communication Class devices.

Add support for multi-interface functions to USB serial core and export
an interface that allows drivers to claim a second sibling interface.
The interface could easily be extended to allow claiming further
interfaces if ever needed.

When a driver claims a sibling interface in probe(), core allocates
resources for any bulk in, bulk out, interrupt in and interrupt out
endpoints found also on the sibling interface.

Disconnect is implemented so that unbinding either interface will
release the other interface while disconnect() is called precisely once.

Similarly, suspend() is called when the first sibling interface is
suspended and resume() is called when the last sibling interface is
resumed by USB core.
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

Refactor endpoint classification and replace the build-time
endpoint-array sanity checks with runtime checks in preparation for
handling endpoints from a sibling interface.
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

The suspending flag was added back in 2009 but no users ever followed.
Remove it.
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

The XR21V141X does not have a 5- or 6-bit mode, but the current
implementation failed to properly restore the old setting when CS5 or
CS6 was requested. Instead an invalid request would be sent to the
device.

Fixes: c2d405aa ("USB: serial: add MaxLinear/Exar USB to Serial driver")
Reviewed-by: NManivannan Sadhasivam <mani@kernel.org>
Cc: stable@vger.kernel.org	# 5.12
Signed-off-by: NJohan Hovold <johan@kernel.org>

The variable error is initialized to 0 and is set to 1 this
value is never read as it is on an immediate return path. The
only read of error is to check it is 0 and this check is always
true at that point of the code. The variable is redundant and
can be removed.

Addresses-Coverity: ("Unused value")
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NJohan Hovold <johan@kernel.org>

Use an alternate clock divider algorithm and bit ordering for the TA and
TB versions of the pl2303. It was discovered that these variants do not
produce the correct baud rates with the existing scheme.

see https://lore.kernel.org/r/3aee5708-7961-f464-8c5f-6685d96920d6@IEEE.orgSigned-off-by: NMichael G. Katzmann <michaelk@IEEE.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

Add names for the device types to be printed at probe when debugging is
enabled.

Note that the HXN type is referred to as G for now as that is the name
the vendor uses.
Signed-off-by: NJohan Hovold <johan@kernel.org>

Tighten the detection of the new HXN (G) type instead of assuming that
every device which doesn't support the old read request is an HXN.
Signed-off-by: NJohan Hovold <johan@kernel.org>

Rename the legacy type which is supposedly a PL2303H which came in two
variants (and which we handle the same way).
Signed-off-by: NJohan Hovold <johan@kernel.org>

Add support for detecting the HX, TA, TB and HXD device types and refuse
to bind to any unknown types.

Note that the HX type includes the XA variant, while the HXD type
includes the EA, RA and SA variants.
Signed-off-by: NJohan Hovold <johan@kernel.org>

Clean up the type detection somewhat in preparation for adding support
for more types.

Note this also fixes the type debug printk for the new HXN type.
Signed-off-by: NJohan Hovold <johan@kernel.org>

Forward declarations make the code larger and rewrites harder. Harder as
they are often omitted from global changes. Remove forward declarations
which are not really needed, i.e. the definition of the function is
before its first use.
Signed-off-by: NJiri Slaby <jslaby@suse.cz>
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
[ johan: update the prototype comments ]
Signed-off-by: NJohan Hovold <johan@kernel.org>

Forward declarations make the code larger, harder to follow and rewrite.
Harder as the declarations are often omitted from global changes. Remove
forward declarations which are not really needed, i.e. when the
definition of the function is before its first use.
Signed-off-by: NJiri Slaby <jslaby@suse.cz>
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NJohan Hovold <johan@kernel.org>

Doing a

	prctl(PR_SET_MM, PR_SET_MM_AUXV, addr, 1);

will copy 1 byte from userspace to (quite big) on-stack array
and then stash everything to mm->saved_auxv.
AT_NULL terminator will be inserted at the very end.

/proc/*/auxv handler will find that AT_NULL terminator
and copy original stack contents to userspace.

This devious scheme requires CAP_SYS_RESOURCE.
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Pull irq fixes from Thomas Gleixner:
 "A set of irqchip updates:

   - Make the GENERIC_IRQ_MULTI_HANDLER configuration correct

   - Add a missing DT compatible string for the Ingenic driver

   - Remove the pointless debugfs_file pointer from struct irqdomain"

* tag 'irq-urgent-2021-03-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/ingenic: Add support for the JZ4760
  dt-bindings/irq: Add compatible string for the JZ4760B
  irqchip: Do not blindly select CONFIG_GENERIC_IRQ_MULTI_HANDLER
  ARM: ep93xx: Select GENERIC_IRQ_MULTI_HANDLER directly
  irqdomain: Remove debugfs_file from struct irq_domain