waitpid():  Corrects two problems when CONFIG_SCHED_HAVE_PARENT+CONFIG_SCHED_CHILD_STATUS are enabled:  (1) Was erroring out if the waited for task had already exited, and (2) was not freeing resources when a wait was completed.

Master

* stm32f33xxx_hrtim.h: add some comments

* stm32_hrtim: add burst mode configuration, rename some definitions

* smps.h: add private data to the smps_s structure

* stm32_hrtim: cosmetics

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

Kinetis: First cut implementation of the alarm read function.  Pretty simple because the Kinetis RTC is just a 1Hz counter.

Correct naming of fields in struct alm_rdalarm_s.  Should not be the same as the corresponding fields of struct alm_setalarm_s.  The whole purpose of that naming convention is to keep the field names unique.

Add hooks for Boris Astardzhiev's RTC change for STM32L4 to Kinetis. Lower level logic not yet implemented.

Extend the RTC framework with an alarm read ioctl (RTC_RD_ALARM). Through it consumer could get configuration settings about previously scheduled hardware alarms (active status, hours, minutes, seconds).

Master

* stm32_dac.c: fix compilation when DMA disabled for channel

* smps.h: update some comments

* smps.c: more sanity checks

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

networking:  IGMP: Remove special support for interrupt level processing (there is none) and fix some timer cancellation logic.  In many files, correct comments.  There is no interrupt level processing in the networking layer.

stm32f7: do not enable read-modify-write on DTCM.  "AN 4667 - STM32F7 Series system architecture and performance" recommends to disable read-modify-write on DTCM: "If the DTCM-RAM is used as data location and the variables used are byte or/and halfword types, since there is no ECC management in this RAM on the STM32F7 Series, it is recommended to disable the read-modify-write of the DTCM-RAM in the DTCM interface (inthe DTCMCR register) to increase the performance."

Some STM32L4 DAC changes

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

net_timedwait() and net_lockedwait() no longer return errors via errno but returned negated errno values like other internal OS functions.

Signed-off-by: Juha Niskanen <juha.niskanen@haltian.com>

User should not be bothered by details like how many IP blocks there are. As no
current STM32L4 has second DAC block (channel 3), remove support for such
hypothetical hardware. DMA channels corrected.

Change-Id: I2cba7e55803871f1ff945538113f12cf5088f68d
Signed-off-by: Juha Niskanen <juha.niskanen@haltian.com>

Change-Id: Ibc6dc90b39b784b5534b8908eaf615bf1ddcb7ed
Signed-off-by: Juha Niskanen <juha.niskanen@haltian.com>

Actually write something to the DAC DMA buffer.

Change-Id: I1b2516ac26fb17f5242611b56be8926c5f40c2c7
Signed-off-by: Juha Niskanen <juha.niskanen@haltian.com>

This reverts commit ad2ef95d.

stm32 FLASH allow non blocking operation on constrained devices

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

   On a very memory constrained device with a single task. The
   sem_wait and sem_post operations can be disabled, to save space.
   The default is blocking enabled.

I'm worried about the stm32_tim_getcounter funtion. It returns always 32 bits. But major stm32 timers have 16 bits counters. I think, it's not a good idea to return the memory behind the TIMx_CNT register.  This changes adds the register size checking.