sched/irq: Fix a infinite recursion problem that a recent change introduced into the i.MX6 SMP implementation.

sched/irq:  Fix an error in a assertion introduced in commit 37c9b3d5.  Noted by Masayuki Ishikawa.

sched/sched:  Extend the last global lock change to work with the lc823450-xgevk which does not support the atomic fetch add but does support disabling interprocessor interrupts.  Disabling interprocessor interrupts will also guarantee that the TCB addres calculation is atomic.

sched/sched:  Implements a global scheduler lock capability as part of SMP support.  This allows the scheduler to be locked with no knowledge or access to the TCB of the currently running task.  This is necessary because accessing the TCB of the currenlty running task is, itself, a non-atomic operation.  This global scheduler lock cpability was add just to support that atomic access to the TCB.

sched/sched:  Clean up some logic that I committed yesterday.  Add more comments and conditional logic to clarify the issues.

configs/stm32f429i-disco: Separate SPI4 from MTD init.  I was trying to attach a non-MTD peripheral to an STM32F429I Discovery Board's SPI4 port and was hitting compilation problems, since CONFIG_MTD and thus 'struct mtd_geometry_s' was not defined.  This patch separates SPI4 initialization logic from MTD initialization logic.

sched/sched:  SMP: Fix this_task() to be an atomic operation.  In the previous implementation, this_task() was defined in sched.h by using just a macro current_task(this_cpu()).  However, I found that this is not atomic and actually sometimes switching CPU happened in executing the macro when we tested audio steaming plus executing commands via telnet.  This change resolves this issue by implementing atomic this_task()in sched_thistask.c which is newly introduced.

spin_lock_irqsave() and spin_unlock_irqrestore() are only valid if the CPU supports global disabling of interrupts.

libc/stdio: in dtoa(), up_interrupt_context() is used in a debug assertion.  up_interrupt_context() is not available in the user-mode phase of the PROTECTED or KERNEL build configurations.  In those configurations, enabling libc floating point support and debug assertions will result in an undefined reference to up_interrupt_context().

    sched/:  Convert legitimate uses of task_create() to nxtask_create().  Review handling of returned values from all uses of kthread_create() (as well as nxtask_create()).
    graphics/:  Review return values for all calls to kthread_start() because it no longer returns an errno.
    drivers/:  threads started by drivers should be kernel threads, not user tasks.  Review return values for all calls to kthread_start() because it no longer returns an errno.
    configs/:  threads started by board bringup logic should be kernel threads, not user tasksi (part 2 of 2).
    sched/task:  Add nxtask_create().  Kthread_create() and nxtask_create() are internal OS functions and should not modify the errno variable.  configs/:  threads started by board bringup logic should be kernel threads, not user tasks.

sched/wdog:  wd_start() is an internal OS function and should not set the errno value.  Reviewed and updated every call to wd_start() to verify if return value is used and if so if the errno value is accessed.

SMP: Introduce a new global IRQ clearing logic and tasklist protection.

Approved-by: Gregory Nutt <gnutt@nuttx.org>

SAMDL: Added I2S register definitions

Approved-by: Gregory Nutt <gnutt@nuttx.org>

The previous implementation of clearing global IRQ in sched_addreadytorun()
and sched_removereadytorun() was done too early. As a result, nxsem_post()
would have a chance to enter the critical section even nxsem_wait() is
still not in blocked state. This patch moves clearing global IRQ controls
from sched_addreadytorun() and sched_removereadytorun() to sched_resumescheduler()
to ensure that nxsem_post() can enter the critical section correctly.

For this change, sched_resumescheduler.c is always necessary for SMP configuration.
In addition, by this change, task_exit() had to be modified so that it calls
sched_resumescheduler() because it calls sched_removescheduler() inside the
function, otherwise it will cause a deadlock.

However, I encountered another DEBUGASSERT() in sched_cpu_select() during
HTTP streaming aging test on lc823450-xgevk. Actually sched_cpu_select()
accesses the g_assignedtasks which might be changed by another CPU. Similarly,
other tasklists might be modified simultaneously if both CPUs are executing
scheduling logic. To avoid this, I introduced tasklist protetion APIs.

With these changes, SMP kernel stability has been much improved.

Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>

    binfmt/, configs/, grahics/, libc/, mm/, net/, sched/:  OS references to the errno variable should always use the set_errno(), get_errno() macros
    arch/arm/src/stm32 and stm32f7:  Architecture-specific code is not permitted to modify the errno variable.  drivers/ and libc/:  OS references to the errno variable should always use the set_errno(), get_errno() macros

binfmt/, drivers/, graphics/:  Fix several inappropriate accesses to get_errno() that were missed in previous changes (some going back to nuttx-.23).

sched/:  Fix several inappropriate accesses to get_errno() that were missed in previous changes (some going back to nuttx-.23).  Add new nxsched_setscheduler() and nxsched_getaffinity() which are equivalent to their counterparts without the nx on front.  These versions do not modify the errno value.  Changed all calls within the OS to use these newer versions of the functions.