ihidev(4): Fix locking and interrupt handler.

- Can't run iic_exec in softint because it does cv_wait, at least on
some i2c controllers -- defer to workqueue instead.

- Fix violations of locking rules:
. Do not take a lock at higher IPL than it is defined at!
. Do not sleep under a lock!
. Definitely do not sleep under a spin lock!
In this case, sc_intr_lock was defined at IPL_VM but used at IPL_TTY,
and i2c transactions -- possibly causing sleep for cv_wait -- were
issued under it.

But in this case, the interrupt handler needs only a single bit to
mark whether the work is pending, so just use atomic_swap for that.

- Use an adaptive lock (IPL_NONE) for i2c transactions.

- Detach children, and do so before freeing anything.

ihidev(4): Add missing includes and header guard.

Mark the sections that are conventionally separate files for hardware
interface (*reg.h) versus software state (*var.h).

branches: 1.4.10; 1.4.20;
Re-enable the intr / mask / softint / unmask dance now that the x86
interrupt issue is fixed. Verified working by ryoon@ (thanks!).

branches: 1.3.2;
Revert previous until issues can be sorted out.

The hid-over-i2c spec specifies that compliant devices use level-sensitive
interrupts. However, it's not safe to do i2c bus access in hard interrupt
context, and we must read the event data off the device in order to clear
the interrupt condition.

Address this by using acpi_intr_mask() to mask off the interrupt source
while a softint is pending to service the events, re-enabling it once
servicing is completed.

While here, re-factor the interrupt setup / tear-down code a bit to
eventually once day simplify supporting the FDT bindings for hid-over-i2c.

branches: 1.1.4;
Add drivers for HID over I2C devices, and a driver for I2C mices.
From OpenBSD.

Re-enable the intr / mask / softint / unmask dance now that the x86
interrupt issue is fixed. Verified working by ryoon@ (thanks!).

Revert previous until issues can be sorted out.

The hid-over-i2c spec specifies that compliant devices use level-sensitive
interrupts. However, it's not safe to do i2c bus access in hard interrupt
context, and we must read the event data off the device in order to clear
the interrupt condition.

Address this by using acpi_intr_mask() to mask off the interrupt source
while a softint is pending to service the events, re-enabling it once
servicing is completed.

While here, re-factor the interrupt setup / tear-down code a bit to
eventually once day simplify supporting the FDT bindings for hid-over-i2c.

Add drivers for HID over I2C devices, and a driver for I2C mices.
From OpenBSD.

Revert previous until issues can be sorted out.

The hid-over-i2c spec specifies that compliant devices use level-sensitive
interrupts. However, it's not safe to do i2c bus access in hard interrupt
context, and we must read the event data off the device in order to clear
the interrupt condition.

Address this by using acpi_intr_mask() to mask off the interrupt source
while a softint is pending to service the events, re-enabling it once
servicing is completed.

While here, re-factor the interrupt setup / tear-down code a bit to
eventually once day simplify supporting the FDT bindings for hid-over-i2c.

Add drivers for HID over I2C devices, and a driver for I2C mices.
From OpenBSD.

The hid-over-i2c spec specifies that compliant devices use level-sensitive
interrupts. However, it's not safe to do i2c bus access in hard interrupt
context, and we must read the event data off the device in order to clear
the interrupt condition.

Address this by using acpi_intr_mask() to mask off the interrupt source
while a softint is pending to service the events, re-enabling it once
servicing is completed.

While here, re-factor the interrupt setup / tear-down code a bit to
eventually once day simplify supporting the FDT bindings for hid-over-i2c.

Add drivers for HID over I2C devices, and a driver for I2C mices.
From OpenBSD.

Add drivers for HID over I2C devices, and a driver for I2C mices.
From OpenBSD.